0: rm: cannot remove '/dev/shm/hsakmt_shared_mem': Operation not permitted
 0: rm: cannot remove '/dev/shm/sem.hsakmt_semaphore': Operation not permitted
58: rm: cannot remove '/dev/shm/hsakmt_shared_mem': Operation not permitted
58: rm: cannot remove '/dev/shm/sem.hsakmt_semaphore': Operation not permitted
53: rm: cannot remove '/dev/shm/hsakmt_shared_mem': Operation not permitted
53: rm: cannot remove '/dev/shm/sem.hsakmt_semaphore': Operation not permitted
52: rm: cannot remove '/dev/shm/hsakmt_shared_mem': Operation not permitted
52: rm: cannot remove '/dev/shm/sem.hsakmt_semaphore': Operation not permitted
48: rm: cannot remove '/dev/shm/hsakmt_shared_mem': Operation not permitted
48: rm: cannot remove '/dev/shm/sem.hsakmt_semaphore': Operation not permitted
56: rm: cannot remove '/dev/shm/hsakmt_shared_mem': Operation not permitted
56: rm: cannot remove '/dev/shm/sem.hsakmt_semaphore': Operation not permitted
 2: rm: cannot remove '/dev/shm/hsakmt_shared_mem': Operation not permitted
 2: rm: cannot remove '/dev/shm/sem.hsakmt_semaphore': Operation not permitted
54: rm: cannot remove '/dev/shm/hsakmt_shared_mem': Operation not permitted
54: rm: cannot remove '/dev/shm/sem.hsakmt_semaphore': Operation not permitted
62: rm: cannot remove '/dev/shm/hsakmt_shared_mem': Operation not permitted
62: rm: cannot remove '/dev/shm/sem.hsakmt_semaphore': Operation not permitted
60: rm: cannot remove '/dev/shm/hsakmt_shared_mem': Operation not permitted
60: rm: cannot remove '/dev/shm/sem.hsakmt_semaphore': Operation not permitted
24: rm: cannot remove '/dev/shm/hsakmt_shared_mem': Operation not permitted
24: rm: cannot remove '/dev/shm/sem.hsakmt_semaphore': Operation not permitted
 1: rm: cannot remove '/dev/shm/hsakmt_shared_mem': Operation not permitted
 1: rm: cannot remove '/dev/shm/sem.hsakmt_semaphore': Operation not permitted
57: rm: cannot remove '/dev/shm/hsakmt_shared_mem': Operation not permitted
57: rm: cannot remove '/dev/shm/sem.hsakmt_semaphore': Operation not permitted
51: rm: cannot remove '/dev/shm/hsakmt_shared_mem': Operation not permitted
51: rm: cannot remove '/dev/shm/sem.hsakmt_semaphore': Operation not permitted
55: rm: cannot remove '/dev/shm/hsakmt_shared_mem': Operation not permitted
55: rm: cannot remove '/dev/shm/sem.hsakmt_semaphore': Operation not permitted
61: rm: cannot remove '/dev/shm/hsakmt_shared_mem': Operation not permitted
61: rm: cannot remove '/dev/shm/sem.hsakmt_semaphore': Operation not permitted
25: rm: cannot remove '/dev/shm/hsakmt_shared_mem': Operation not permitted
25: rm: cannot remove '/dev/shm/sem.hsakmt_semaphore': Operation not permitted
26: rm: cannot remove '/dev/shm/hsakmt_shared_mem': Operation not permitted
26: rm: cannot remove '/dev/shm/sem.hsakmt_semaphore': Operation not permitted
10: rm: cannot remove '/dev/shm/hsakmt_shared_mem': Operation not permitted
10: rm: cannot remove '/dev/shm/sem.hsakmt_semaphore': Operation not permitted
46: rm: cannot remove '/dev/shm/hsakmt_shared_mem': Operation not permitted
46: rm: cannot remove '/dev/shm/sem.hsakmt_semaphore': Operation not permitted
 6: rm: cannot remove '/dev/shm/hsakmt_shared_mem': Operation not permitted
 6: rm: cannot remove '/dev/shm/sem.hsakmt_semaphore': Operation not permitted
20: rm: cannot remove '/dev/shm/hsakmt_shared_mem': Operation not permitted
20: rm: cannot remove '/dev/shm/sem.hsakmt_semaphore': Operation not permitted
40: rm: cannot remove '/dev/shm/hsakmt_shared_mem': Operation not permitted
40: rm: cannot remove '/dev/shm/sem.hsakmt_semaphore': Operation not permitted
38: rm: cannot remove '/dev/shm/hsakmt_shared_mem': Operation not permitted
38: rm: cannot remove '/dev/shm/rocm_smi_card0': Operation not permitted
38: rm: cannot remove '/dev/shm/rocm_smi_card1': Operation not permitted
38: rm: cannot remove '/dev/shm/rocm_smi_card2': Operation not permitted
38: rm: cannot remove '/dev/shm/rocm_smi_card3': Operation not permitted
63: rm: cannot remove '/dev/shm/hsakmt_shared_mem': Operation not permitted
63: rm: cannot remove '/dev/shm/sem.hsakmt_semaphore': Operation not permitted
59: rm: cannot remove '/dev/shm/hsakmt_shared_mem': Operation not permitted
59: rm: cannot remove '/dev/shm/sem.hsakmt_semaphore': Operation not permitted
21: rm: cannot remove '/dev/shm/hsakmt_shared_mem': Operation not permitted
21: rm: cannot remove '/dev/shm/sem.hsakmt_semaphore': Operation not permitted
 7: rm: cannot remove '/dev/shm/hsakmt_shared_mem': Operation not permitted
 7: rm: cannot remove '/dev/shm/sem.hsakmt_semaphore': Operation not permitted
 9: rm: cannot remove '/dev/shm/hsakmt_shared_mem': Operation not permitted
 9: rm: cannot remove '/dev/shm/sem.hsakmt_semaphore': Operation not permitted
39: rm: cannot remove '/dev/shm/hsakmt_shared_mem': Operation not permitted
39: rm: cannot remove '/dev/shm/rocm_smi_card0': Operation not permitted
39: rm: cannot remove '/dev/shm/rocm_smi_card1': Operation not permitted
39: rm: cannot remove '/dev/shm/rocm_smi_card2': Operation not permitted
19: rm: cannot remove '/dev/shm/hsakmt_shared_mem': Operation not permitted
19: rm: cannot remove '/dev/shm/sem.hsakmt_semaphore': Operation not permitted
23: rm: cannot remove '/dev/shm/hsakmt_shared_mem': Operation not permitted
23: rm: cannot remove '/dev/shm/sem.hsakmt_semaphore': Operation not permitted
45: rm: cannot remove '/dev/shm/hsakmt_shared_mem': Operation not permitted
45: rm: cannot remove '/dev/shm/sem.hsakmt_semaphore': Operation not permitted
 4: rm: cannot remove '/dev/shm/hsakmt_shared_mem': Operation not permitted
 4: rm: cannot remove '/dev/shm/sem.hsakmt_semaphore': Operation not permitted
44: rm: cannot remove '/dev/shm/hsakmt_shared_mem': Operation not permitted
44: rm: cannot remove '/dev/shm/sem.hsakmt_semaphore': Operation not permitted
42: rm: cannot remove '/dev/shm/hsakmt_shared_mem': Operation not permitted
42: rm: cannot remove '/dev/shm/sem.hsakmt_semaphore': Operation not permitted
18: rm: cannot remove '/dev/shm/hsakmt_shared_mem': Operation not permitted
18: rm: cannot remove '/dev/shm/sem.hsakmt_semaphore': Operation not permitted
22: rm: cannot remove '/dev/shm/hsakmt_shared_mem': Operation not permitted
22: rm: cannot remove '/dev/shm/sem.hsakmt_semaphore': Operation not permitted
30: rm: cannot remove '/dev/shm/hsakmt_shared_mem': Operation not permitted
30: rm: cannot remove '/dev/shm/sem.hsakmt_semaphore': Operation not permitted
28: rm: cannot remove '/dev/shm/hsakmt_shared_mem': Operation not permitted
28: rm: cannot remove '/dev/shm/sem.hsakmt_semaphore': Operation not permitted
32: rm: cannot remove '/dev/shm/hsakmt_shared_mem': Operation not permitted
32: rm: cannot remove '/dev/shm/sem.hsakmt_semaphore': Operation not permitted
16: rm: cannot remove '/dev/shm/hsakmt_shared_mem': Operation not permitted
16: rm: cannot remove '/dev/shm/sem.hsakmt_semaphore': Operation not permitted
14: rm: cannot remove '/dev/shm/hsakmt_shared_mem': Operation not permitted
14: rm: cannot remove '/dev/shm/sem.hsakmt_semaphore': Operation not permitted
38: rm: cannot remove '/dev/shm/rocm_smi_card4': Operation not permitted
38: rm: cannot remove '/dev/shm/rocm_smi_card5': Operation not permitted
38: rm: cannot remove '/dev/shm/rocm_smi_card6': Operation not permitted
38: rm: cannot remove '/dev/shm/rocm_smi_card7': Operation not permitted
38: rm: cannot remove '/dev/shm/sem.hsakmt_semaphore': Operation not permitted
39: rm: cannot remove '/dev/shm/rocm_smi_card3': Operation not permitted
39: rm: cannot remove '/dev/shm/rocm_smi_card4': Operation not permitted
39: rm: cannot remove '/dev/shm/rocm_smi_card5': Operation not permitted
39: rm: cannot remove '/dev/shm/rocm_smi_card6': Operation not permitted
 5: rm: cannot remove '/dev/shm/hsakmt_shared_mem': Operation not permitted
 5: rm: cannot remove '/dev/shm/sem.hsakmt_semaphore': Operation not permitted
11: rm: cannot remove '/dev/shm/hsakmt_shared_mem': Operation not permitted
11: rm: cannot remove '/dev/shm/sem.hsakmt_semaphore': Operation not permitted
47: rm: cannot remove '/dev/shm/hsakmt_shared_mem': Operation not permitted
47: rm: cannot remove '/dev/shm/sem.hsakmt_semaphore': Operation not permitted
43: rm: cannot remove '/dev/shm/hsakmt_shared_mem': Operation not permitted
43: rm: cannot remove '/dev/shm/sem.hsakmt_semaphore': Operation not permitted
41: rm: cannot remove '/dev/shm/hsakmt_shared_mem': Operation not permitted
41: rm: cannot remove '/dev/shm/sem.hsakmt_semaphore': Operation not permitted
17: rm: cannot remove '/dev/shm/hsakmt_shared_mem': Operation not permitted
17: rm: cannot remove '/dev/shm/sem.hsakmt_semaphore': Operation not permitted
 3: rm: cannot remove '/dev/shm/hsakmt_shared_mem': Operation not permitted
 3: rm: cannot remove '/dev/shm/sem.hsakmt_semaphore': Operation not permitted
29: rm: cannot remove '/dev/shm/hsakmt_shared_mem': Operation not permitted
29: rm: cannot remove '/dev/shm/rocm_smi_card0': Operation not permitted
31: rm: cannot remove '/dev/shm/hsakmt_shared_mem': Operation not permitted
31: rm: cannot remove '/dev/shm/sem.hsakmt_semaphore': Operation not permitted
39: rm: cannot remove '/dev/shm/rocm_smi_card7': Operation not permitted
39: rm: cannot remove '/dev/shm/sem.hsakmt_semaphore': Operation not permitted
34: rm: cannot remove '/dev/shm/hsakmt_shared_mem': Operation not permitted
34: rm: cannot remove '/dev/shm/sem.hsakmt_semaphore': Operation not permitted
36: rm: cannot remove '/dev/shm/hsakmt_shared_mem': Operation not permitted
36: rm: cannot remove '/dev/shm/sem.hsakmt_semaphore': Operation not permitted
29: rm: cannot remove '/dev/shm/rocm_smi_card1': Operation not permitted
29: rm: cannot remove '/dev/shm/rocm_smi_card2': Operation not permitted
49: rm: cannot remove '/dev/shm/hsakmt_shared_mem': Operation not permitted
49: rm: cannot remove '/dev/shm/sem.hsakmt_semaphore': Operation not permitted
37: rm: cannot remove '/dev/shm/hsakmt_shared_mem': Operation not permitted
37: rm: cannot remove '/dev/shm/sem.hsakmt_semaphore': Operation not permitted
27: rm: cannot remove '/dev/shm/hsakmt_shared_mem': Operation not permitted
27: rm: cannot remove '/dev/shm/rocm_smi_card0': Operation not permitted
27: rm: cannot remove '/dev/shm/rocm_smi_card1': Operation not permitted
27: rm: cannot remove '/dev/shm/rocm_smi_card2': Operation not permitted
29: rm: cannot remove '/dev/shm/rocm_smi_card3': Operation not permitted
27: rm: cannot remove '/dev/shm/rocm_smi_card3': Operation not permitted
27: rm: cannot remove '/dev/shm/rocm_smi_card4': Operation not permitted
27: rm: cannot remove '/dev/shm/rocm_smi_card5': Operation not permitted
27: rm: cannot remove '/dev/shm/rocm_smi_card6': Operation not permitted
29: rm: cannot remove '/dev/shm/rocm_smi_card4': Operation not permitted
27: rm: cannot remove '/dev/shm/rocm_smi_card7': Operation not permitted
27: rm: cannot remove '/dev/shm/sem.hsakmt_semaphore': Operation not permitted
29: rm: cannot remove '/dev/shm/rocm_smi_card5': Operation not permitted
29: rm: cannot remove '/dev/shm/rocm_smi_card6': Operation not permitted
29: rm: cannot remove '/dev/shm/rocm_smi_card7': Operation not permitted
29: rm: cannot remove '/dev/shm/sem.hsakmt_semaphore': Operation not permitted
15: rm: cannot remove '/dev/shm/hsakmt_shared_mem': Operation not permitted
15: rm: cannot remove '/dev/shm/sem.hsakmt_semaphore': Operation not permitted
12: rm: cannot remove '/dev/shm/hsakmt_shared_mem': Operation not permitted
12: rm: cannot remove '/dev/shm/sem.hsakmt_semaphore': Operation not permitted
50: rm: cannot remove '/dev/shm/hsakmt_shared_mem': Operation not permitted
50: rm: cannot remove '/dev/shm/sem.hsakmt_semaphore': Operation not permitted
35: rm: cannot remove '/dev/shm/hsakmt_shared_mem': Operation not permitted
35: rm: cannot remove '/dev/shm/sem.hsakmt_semaphore': Operation not permitted
13: rm: cannot remove '/dev/shm/hsakmt_shared_mem': Operation not permitted
13: rm: cannot remove '/dev/shm/sem.hsakmt_semaphore': Operation not permitted
33: rm: cannot remove '/dev/shm/hsakmt_shared_mem': Operation not permitted
33: rm: cannot remove '/dev/shm/sem.hsakmt_semaphore': Operation not permitted
 8: rm: cannot remove '/dev/shm/hsakmt_shared_mem': Operation not permitted
 8: rm: cannot remove '/dev/shm/sem.hsakmt_semaphore': Operation not permitted
 0: Successfully preprocessed all matching files.
 0: Detected CUDA files, patching ldflags
 0: Emitting ninja build file /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/Megatron-DeepSpeed/megatron/fused_kernels/build/build.ninja...
 0: Building extension module scaled_upper_triang_masked_softmax_cuda...
 0: Allowing ninja to set a default number of workers... (overridable by setting the environment variable MAX_JOBS=N)
 0: Loading extension module scaled_upper_triang_masked_softmax_cuda...
 0: Successfully preprocessed all matching files.
 0: Detected CUDA files, patching ldflags
 0: Emitting ninja build file /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/Megatron-DeepSpeed/megatron/fused_kernels/build/build.ninja...
 0: Building extension module scaled_masked_softmax_cuda...
 0: Allowing ninja to set a default number of workers... (overridable by setting the environment variable MAX_JOBS=N)
 0: Loading extension module scaled_masked_softmax_cuda...
 0: Successfully preprocessed all matching files.
 0: Detected CUDA files, patching ldflags
 0: Emitting ninja build file /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/Megatron-DeepSpeed/megatron/fused_kernels/build/build.ninja...
 0: Building extension module fused_mix_prec_layer_norm_cuda...
 0: Allowing ninja to set a default number of workers... (overridable by setting the environment variable MAX_JOBS=N)
 0: Loading extension module fused_mix_prec_layer_norm_cuda...
 0: Successfully preprocessed all matching files.
 0: Successfully preprocessed all matching files.
 0: Successfully preprocessed all matching files.
 0: Successfully preprocessed all matching files.
 0: Successfully preprocessed all matching files.
28: Successfully preprocessed all matching files.
28: Successfully preprocessed all matching files.
17: Successfully preprocessed all matching files.
14: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
14:   warnings.warn(
14: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
14:   warnings.warn(
14: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
14:   warnings.warn(
14: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
14:   warnings.warn(
14: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
14:   warnings.warn(
14: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
14:   warnings.warn(
40: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
40:   warnings.warn(
40: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
40:   warnings.warn(
22: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
22:   warnings.warn(
40: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
40:   warnings.warn(
22: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
22:   warnings.warn(
22: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
22:   warnings.warn(
22: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
22:   warnings.warn(
22: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
22:   warnings.warn(
15: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
15:   warnings.warn(
40: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
40:   warnings.warn(
40: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
40:   warnings.warn(
44: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
44:   warnings.warn(
44: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
44:   warnings.warn(
 6: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 6:   warnings.warn(
44: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
44:   warnings.warn(
44: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
44:   warnings.warn(
44: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
44:   warnings.warn(
22: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
22:   warnings.warn(
15: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
15:   warnings.warn(
15: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
15:   warnings.warn(
15: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
15:   warnings.warn(
55: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
55:   warnings.warn(
55: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
55:   warnings.warn(
55: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
55:   warnings.warn(
15: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
15:   warnings.warn(
62: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
62:   warnings.warn(
62: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
62:   warnings.warn(
40: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
40:   warnings.warn(
55: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
55:   warnings.warn(
55: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
55:   warnings.warn(
15: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
15:   warnings.warn(
62: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
62:   warnings.warn(
 6: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 6:   warnings.warn(
 6: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 6:   warnings.warn(
 6: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 6:   warnings.warn(
44: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
44:   warnings.warn(
 6: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 6:   warnings.warn(
12: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
12:   warnings.warn(
12: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
12:   warnings.warn(
12: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
12:   warnings.warn(
 4: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 4:   warnings.warn(
 6: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 6:   warnings.warn(
55: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
55:   warnings.warn(
62: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
62:   warnings.warn(
62: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
62:   warnings.warn(
12: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
12:   warnings.warn(
12: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
12:   warnings.warn(
12: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
12:   warnings.warn(
 4: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 4:   warnings.warn(
 4: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 4:   warnings.warn(
62: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
62:   warnings.warn(
 4: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 4:   warnings.warn(
31: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
31:   warnings.warn(
31: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
31:   warnings.warn(
31: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
31:   warnings.warn(
 4: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 4:   warnings.warn(
28: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
28:   warnings.warn(
28: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
28:   warnings.warn(
28: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
28:   warnings.warn(
31: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
31:   warnings.warn(
31: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
31:   warnings.warn(
 4: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 4:   warnings.warn(
28: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
28:   warnings.warn(
34: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
34:   warnings.warn(
34: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
34:   warnings.warn(
28: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
28:   warnings.warn(
34: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
34:   warnings.warn(
34: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
34:   warnings.warn(
34: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
34:   warnings.warn(
31: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
31:   warnings.warn(
51: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
51:   warnings.warn(
51: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
51:   warnings.warn(
51: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
51:   warnings.warn(
21: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
21:   warnings.warn(
21: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
21:   warnings.warn(
21: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
21:   warnings.warn(
28: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
28:   warnings.warn(
34: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
34:   warnings.warn(
21: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
21:   warnings.warn(
21: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
21:   warnings.warn(
48: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
48:   warnings.warn(
21: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
21:   warnings.warn(
48: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
48:   warnings.warn(
48: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
48:   warnings.warn(
51: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
51:   warnings.warn(
51: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
51:   warnings.warn(
50: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
50:   warnings.warn(
50: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
50:   warnings.warn(
53: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
53:   warnings.warn(
53: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
53:   warnings.warn(
51: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
51:   warnings.warn(
53: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
53:   warnings.warn(
53: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
53:   warnings.warn(
54: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
54:   warnings.warn(
41: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
41:   warnings.warn(
41: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
41:   warnings.warn(
53: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
53:   warnings.warn(
48: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
48:   warnings.warn(
48: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
48:   warnings.warn(
38: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
38:   warnings.warn(
38: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
38:   warnings.warn(
41: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
41:   warnings.warn(
38: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
38:   warnings.warn(
38: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
38:   warnings.warn(
38: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
38:   warnings.warn(
50: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
50:   warnings.warn(
41: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
41:   warnings.warn(
41: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
41:   warnings.warn(
49: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
49:   warnings.warn(
49: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
49:   warnings.warn(
50: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
50:   warnings.warn(
36: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
36:   warnings.warn(
50: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
50:   warnings.warn(
49: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
49:   warnings.warn(
49: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
49:   warnings.warn(
41: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
41:   warnings.warn(
 3: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 3:   warnings.warn(
 3: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 3:   warnings.warn(
 3: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 3:   warnings.warn(
49: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
49:   warnings.warn(
53: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
53:   warnings.warn(
50: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
50:   warnings.warn(
38: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
38:   warnings.warn(
54: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
54:   warnings.warn(
54: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
54:   warnings.warn(
54: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
54:   warnings.warn(
36: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
36:   warnings.warn(
49: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
49:   warnings.warn(
54: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
54:   warnings.warn(
36: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
36:   warnings.warn(
36: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
36:   warnings.warn(
30: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
30:   warnings.warn(
36: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
36:   warnings.warn(
 3: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 3:   warnings.warn(
 3: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 3:   warnings.warn(
30: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
30:   warnings.warn(
30: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
30:   warnings.warn(
13: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
13:   warnings.warn(
30: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
30:   warnings.warn(
30: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
30:   warnings.warn(
36: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
36:   warnings.warn(
13: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
13:   warnings.warn(
13: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
13:   warnings.warn(
13: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
13:   warnings.warn(
54: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
54:   warnings.warn(
13: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
13:   warnings.warn(
 3: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 3:   warnings.warn(
58: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
58:   warnings.warn(
18: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
18:   warnings.warn(
18: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
18:   warnings.warn(
18: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
18:   warnings.warn(
58: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
58:   warnings.warn(
47: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
47:   warnings.warn(
47: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
47:   warnings.warn(
47: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
47:   warnings.warn(
47: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
47:   warnings.warn(
13: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
13:   warnings.warn(
18: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
18:   warnings.warn(
59: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
59:   warnings.warn(
59: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
59:   warnings.warn(
59: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
59:   warnings.warn(
 8: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 8:   warnings.warn(
47: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
47:   warnings.warn(
45: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
45:   warnings.warn(
18: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
18:   warnings.warn(
58: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
58:   warnings.warn(
58: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
58:   warnings.warn(
58: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
58:   warnings.warn(
 8: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 8:   warnings.warn(
45: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
45:   warnings.warn(
45: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
45:   warnings.warn(
18: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
18:   warnings.warn(
 8: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 8:   warnings.warn(
 8: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 8:   warnings.warn(
47: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
47:   warnings.warn(
59: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
59:   warnings.warn(
59: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
59:   warnings.warn(
25: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
25:   warnings.warn(
45: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
45:   warnings.warn(
45: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
45:   warnings.warn(
58: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
58:   warnings.warn(
45: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
45:   warnings.warn(
 8: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 8:   warnings.warn(
26: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
26:   warnings.warn(
32: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
32:   warnings.warn(
43: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
43:   warnings.warn(
43: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
43:   warnings.warn(
59: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
59:   warnings.warn(
25: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
25:   warnings.warn(
16: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
16:   warnings.warn(
16: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
16:   warnings.warn(
25: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
25:   warnings.warn(
16: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
16:   warnings.warn(
25: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
25:   warnings.warn(
16: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
16:   warnings.warn(
 8: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 8:   warnings.warn(
25: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
25:   warnings.warn(
25: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
25:   warnings.warn(
16: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
16:   warnings.warn(
35: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
35:   warnings.warn(
35: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
35:   warnings.warn(
33: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
33:   warnings.warn(
33: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
33:   warnings.warn(
32: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
32:   warnings.warn(
32: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
32:   warnings.warn(
32: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
32:   warnings.warn(
43: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
43:   warnings.warn(
43: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
43:   warnings.warn(
43: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
43:   warnings.warn(
33: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
33:   warnings.warn(
33: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
33:   warnings.warn(
32: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
32:   warnings.warn(
16: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
16:   warnings.warn(
43: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
43:   warnings.warn(
33: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
33:   warnings.warn(
26: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
26:   warnings.warn(
26: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
26:   warnings.warn(
26: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
26:   warnings.warn(
 1: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 1:   warnings.warn(
32: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
32:   warnings.warn(
 1: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 1:   warnings.warn(
 1: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 1:   warnings.warn(
35: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
35:   warnings.warn(
35: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
35:   warnings.warn(
35: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
35:   warnings.warn(
56: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
56:   warnings.warn(
56: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
56:   warnings.warn(
56: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
56:   warnings.warn(
26: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
26:   warnings.warn(
56: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
56:   warnings.warn(
26: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
26:   warnings.warn(
56: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
56:   warnings.warn(
 1: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 1:   warnings.warn(
 1: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 1:   warnings.warn(
35: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
35:   warnings.warn(
33: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
33:   warnings.warn(
 9: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 9:   warnings.warn(
56: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
56:   warnings.warn(
46: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
46:   warnings.warn(
46: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
46:   warnings.warn(
46: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
46:   warnings.warn(
46: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
46:   warnings.warn(
46: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
46:   warnings.warn(
 9: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 9:   warnings.warn(
 9: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 9:   warnings.warn(
 9: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 9:   warnings.warn(
 9: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 9:   warnings.warn(
60: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
60:   warnings.warn(
60: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
60:   warnings.warn(
60: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
60:   warnings.warn(
 5: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 5:   warnings.warn(
29: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
29:   warnings.warn(
29: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
29:   warnings.warn(
60: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
60:   warnings.warn(
60: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
60:   warnings.warn(
20: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
20:   warnings.warn(
20: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
20:   warnings.warn(
57: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
57:   warnings.warn(
 5: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 5:   warnings.warn(
 5: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 5:   warnings.warn(
60: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
60:   warnings.warn(
20: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
20:   warnings.warn(
20: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
20:   warnings.warn(
 5: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 5:   warnings.warn(
46: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
46:   warnings.warn(
20: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
20:   warnings.warn(
 9: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 9:   warnings.warn(
 5: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 5:   warnings.warn(
37: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
37:   warnings.warn(
37: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
37:   warnings.warn(
37: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
37:   warnings.warn(
10: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
10:   warnings.warn(
10: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
10:   warnings.warn(
57: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
57:   warnings.warn(
37: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
37:   warnings.warn(
37: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
37:   warnings.warn(
57: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
57:   warnings.warn(
29: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
29:   warnings.warn(
29: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
29:   warnings.warn(
10: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
10:   warnings.warn(
20: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
20:   warnings.warn(
29: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
29:   warnings.warn(
10: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
10:   warnings.warn(
10: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
10:   warnings.warn(
57: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
57:   warnings.warn(
57: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
57:   warnings.warn(
29: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
29:   warnings.warn(
 5: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 5:   warnings.warn(
24: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
24:   warnings.warn(
24: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
24:   warnings.warn(
24: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
24:   warnings.warn(
57: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
57:   warnings.warn(
39: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
39:   warnings.warn(
10: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
10:   warnings.warn(
37: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
37:   warnings.warn(
 0: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 0:   warnings.warn(
 0: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 0:   warnings.warn(
27: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
27:   warnings.warn(
27: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
27:   warnings.warn(
 0: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 0:   warnings.warn(
 0: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 0:   warnings.warn(
24: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
24:   warnings.warn(
39: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
39:   warnings.warn(
11: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
11:   warnings.warn(
 0: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 0:   warnings.warn(
24: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
24:   warnings.warn(
39: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
39:   warnings.warn(
11: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
11:   warnings.warn(
19: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
19:   warnings.warn(
11: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
11:   warnings.warn(
39: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
39:   warnings.warn(
39: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
39:   warnings.warn(
19: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
19:   warnings.warn(
19: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
19:   warnings.warn(
19: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
19:   warnings.warn(
24: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
24:   warnings.warn(
39: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
39:   warnings.warn(
11: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
11:   warnings.warn(
11: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
11:   warnings.warn(
19: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
19:   warnings.warn(
27: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
27:   warnings.warn(
27: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
27:   warnings.warn(
27: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
27:   warnings.warn(
52: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
52:   warnings.warn(
11: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
11:   warnings.warn(
42: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
42:   warnings.warn(
19: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
19:   warnings.warn(
 0: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 0:   warnings.warn(
 2: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 2:   warnings.warn(
27: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
27:   warnings.warn(
63: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
63:   warnings.warn(
 7: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 7:   warnings.warn(
 7: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 7:   warnings.warn(
 7: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 7:   warnings.warn(
 7: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 7:   warnings.warn(
42: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
42:   warnings.warn(
42: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
42:   warnings.warn(
 7: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 7:   warnings.warn(
42: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
42:   warnings.warn(
42: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
42:   warnings.warn(
42: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
42:   warnings.warn(
52: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
52:   warnings.warn(
52: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
52:   warnings.warn(
52: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
52:   warnings.warn(
52: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
52:   warnings.warn(
 2: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 2:   warnings.warn(
 2: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 2:   warnings.warn(
 2: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 2:   warnings.warn(
 2: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 2:   warnings.warn(
63: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
63:   warnings.warn(
 2: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 2:   warnings.warn(
 7: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 7:   warnings.warn(
63: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
63:   warnings.warn(
52: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
52:   warnings.warn(
63: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
63:   warnings.warn(
63: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
63:   warnings.warn(
23: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
23:   warnings.warn(
23: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
23:   warnings.warn(
23: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
23:   warnings.warn(
17: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
17:   warnings.warn(
17: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
17:   warnings.warn(
23: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
23:   warnings.warn(
23: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
23:   warnings.warn(
63: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
63:   warnings.warn(
23: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
23:   warnings.warn(
17: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
17:   warnings.warn(
17: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
17:   warnings.warn(
17: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
17:   warnings.warn(
61: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
61:   warnings.warn(
61: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
61:   warnings.warn(
17: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
17:   warnings.warn(
61: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
61:   warnings.warn(
61: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
61:   warnings.warn(
61: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
61:   warnings.warn(
61: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
61:   warnings.warn(
48: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
48:   warnings.warn(
30: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
30:   warnings.warn(
 1: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 1:   warnings.warn(
28: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
28:   warnings.warn(
60: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
60:   warnings.warn(
15: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
15:   warnings.warn(
 0: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 0:   warnings.warn(
32: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
32:   warnings.warn(
 1: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 1:   warnings.warn(
33: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
33:   warnings.warn(
47: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
47:   warnings.warn(
 0: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 0:   warnings.warn(
32: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
32:   warnings.warn(
42: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
42:   warnings.warn(
15: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
15:   warnings.warn(
41: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
41:   warnings.warn(
47: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
47:   warnings.warn(
20: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
20:   warnings.warn(
52: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
52:   warnings.warn(
10: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
10:   warnings.warn(
30: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
30:   warnings.warn(
62: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
62:   warnings.warn(
28: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
28:   warnings.warn(
60: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
60:   warnings.warn(
24: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
24:   warnings.warn(
56: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
56:   warnings.warn(
 1: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 1:   warnings.warn(
33: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
33:   warnings.warn(
17: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
17:   warnings.warn(
49: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
49:   warnings.warn(
 9: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 9:   warnings.warn(
 4: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 4:   warnings.warn(
36: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
36:   warnings.warn(
 4: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 4:   warnings.warn(
36: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
36:   warnings.warn(
31: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
31:   warnings.warn(
30: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
30:   warnings.warn(
63: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
63:   warnings.warn(
49: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
49:   warnings.warn(
17: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
17:   warnings.warn(
23: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
23:   warnings.warn(
62: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
62:   warnings.warn(
55: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
55:   warnings.warn(
23: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
23:   warnings.warn(
 7: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 7:   warnings.warn(
55: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
55:   warnings.warn(
11: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
11:   warnings.warn(
39: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
39:   warnings.warn(
 7: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 7:   warnings.warn(
43: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
43:   warnings.warn(
39: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
39:   warnings.warn(
20: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
20:   warnings.warn(
52: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
52:   warnings.warn(
59: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
59:   warnings.warn(
27: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
27:   warnings.warn(
25: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
25:   warnings.warn(
 2: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 2:   warnings.warn(
57: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
57:   warnings.warn(
25: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
25:   warnings.warn(
57: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
57:   warnings.warn(
34: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
34:   warnings.warn(
10: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
10:   warnings.warn(
 8: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 8:   warnings.warn(
42: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
42:   warnings.warn(
11: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
11:   warnings.warn(
40: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
40:   warnings.warn(
43: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
43:   warnings.warn(
19: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
19:   warnings.warn(
51: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
51:   warnings.warn(
19: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
19:   warnings.warn(
51: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
51:   warnings.warn(
18: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
18:   warnings.warn(
50: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
50:   warnings.warn(
59: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
59:   warnings.warn(
27: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
27:   warnings.warn(
24: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
24:   warnings.warn(
56: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
56:   warnings.warn(
22: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
22:   warnings.warn(
54: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
54:   warnings.warn(
 2: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 2:   warnings.warn(
34: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
34:   warnings.warn(
37: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
37:   warnings.warn(
22: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
22:   warnings.warn(
54: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
54:   warnings.warn(
16: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
16:   warnings.warn(
48: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
48:   warnings.warn(
 5: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 5:   warnings.warn(
31: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
31:   warnings.warn(
63: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
63:   warnings.warn(
 5: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 5:   warnings.warn(
37: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
37:   warnings.warn(
26: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
26:   warnings.warn(
14: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
14:   warnings.warn(
18: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
18:   warnings.warn(
50: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
50:   warnings.warn(
45: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
45:   warnings.warn(
 9: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 9:   warnings.warn(
41: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
41:   warnings.warn(
58: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
58:   warnings.warn(
46: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
46:   warnings.warn(
29: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
29:   warnings.warn(
29: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
29:   warnings.warn(
38: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
38:   warnings.warn(
61: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
61:   warnings.warn(
61: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
61:   warnings.warn(
21: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
21:   warnings.warn(
 6: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 6:   warnings.warn(
 6: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 6:   warnings.warn(
 3: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 3:   warnings.warn(
26: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
26:   warnings.warn(
58: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
58:   warnings.warn(
16: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
16:   warnings.warn(
35: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
35:   warnings.warn(
 3: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 3:   warnings.warn(
48: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
48:   warnings.warn(
21: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
21:   warnings.warn(
35: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
35:   warnings.warn(
53: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
53:   warnings.warn(
53: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
53:   warnings.warn(
38: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
38:   warnings.warn(
13: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
13:   warnings.warn(
14: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
14:   warnings.warn(
46: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
46:   warnings.warn(
13: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
13:   warnings.warn(
45: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
45:   warnings.warn(
12: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
12:   warnings.warn(
12: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
12:   warnings.warn(
44: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
44:   warnings.warn(
44: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
44:   warnings.warn(
 8: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
 8:   warnings.warn(
40: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/torch/distributed/distributed_c10d.py:429: UserWarning: torch.distributed.distributed_c10d._get_global_rank is deprecated please use torch.distributed.distributed_c10d.get_global_rank instead
40:   warnings.warn(
13: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 6: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 7: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
16: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
19: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
27: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
34: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
39: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
43: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
44: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
48: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
52: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
53: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 0: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
58: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
60: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
61: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
62: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
63: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
32: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
13: Emitting ninja build file /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu/utils/build.ninja...
13: Building extension module utils...
13: Allowing ninja to set a default number of workers... (overridable by setting the environment variable MAX_JOBS=N)
32: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
32: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
32: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
32: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
32: 
32: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
32: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
34: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
34: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
34: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
34: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
34: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
34: 
34: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
33: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
33: 
33: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
33: 
33: 
33: 
33: 
33: 
36: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
36: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
36: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
36: 
36: 
36: 
36: 
36: 
35: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
35: 
35: 
35: 
35: 
35: 
35: 
35: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
39: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
39: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
37: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
37: 
37: 
37: 
37: 
37: 
37: 
37: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
39: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
39: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
39: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
39: 
39: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
38: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
38: 
38: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
38: 
38: 
38: 
38: 
38: 
43: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
43: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
43: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
43: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
43: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
43: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
43: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
40: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
40: 
40: 
40: 
40: 
40: 
40: 
40: 
44: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
44: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
44: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
44: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
44: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
44: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
44: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
41: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
41: 
41: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
41: 
41: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
41: 
41: 
41: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
42: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
42: 
42: 
42: 
42: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
42: 
42: 
42: 
48: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
48: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
48: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
48: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
48: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
48: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
48: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
46: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
46: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
46: 
46: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
46: 
46: 
46: 
46: 
45: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
45: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
45: 
45: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
45: 
45: 
45: 
45: 
47: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
47: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
47: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
47: 
47: 
47: 
47: 
47: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
52: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
52: 
49: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
49: 
49: 
49: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
49: 
49: 
49: 
49: 
52: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
52: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
52: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
52: 
52: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
53: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
53: 
53: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
53: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
53: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
53: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
53: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
50: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
50: 
50: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
50: 
50: 
50: 
50: 
50: 
51: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
51: 
51: 
51: 
51: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
51: 
51: 
51: 
58: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
54: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
54: 
54: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
54: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
54: 
54: 
54: 
58: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
54: 
58: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
58: 
58: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
58: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
58: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
55: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
55: 
55: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
55: 
55: 
55: 
55: 
55: 
56: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
56: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
56: 
56: 
56: 
56: 
56: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
56: 
60: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
60: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
60: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
60: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
60: 
60: 
60: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
57: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
57: 
57: 
57: 
57: 
57: 
57: 
57: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
61: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
61: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
61: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
61: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
61: 
61: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
61: 
62: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
62: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
63: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
59: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
59: 
59: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
59: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
59: 
59: 
59: 
59: 
62: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
62: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
62: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
62: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
62: 
63: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
63: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
63: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
63: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
63: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
63: 
13: Loading extension module utils...
51: Loading extension module utils...
51: Loading extension module utils...
51: Loading extension module utils...
51: Loading extension module utils...
51: Loading extension module utils...
50: Loading extension module utils...
51: Loading extension module utils...
50: Loading extension module utils...
51: Loading extension module utils...
50: Loading extension module utils...
51: Loading extension module utils...
50: Loading extension module utils...
50: Loading extension module utils...
50: Loading extension module utils...
50: Loading extension module utils...
50: Loading extension module utils...
58: Loading extension module utils...
 0: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
58: Loading extension module utils...
 0: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 0: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 0: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
58: Loading extension module utils...
 0: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 0: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
58: Loading extension module utils...
 0: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
58: Loading extension module utils...
 2: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 2: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 2: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 2: 
 2: 
 2: 
 2: 
 2: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 1: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 1: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 1: 
 1: 
 1: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 1: 
 1: 
 1: 
 3: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 3: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 3: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 3: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 3: 
 3: 
 3: 
 3: 
 7: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 6: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 7: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 7: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 6: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 6: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 7: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 6: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 7: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 7: 
 6: 
 7: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 6: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 6: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 4: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 4: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 4: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 4: 
 4: 
 4: 
 4: 
 4: 
 5: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 5: 
 5: 
 5: 
 5: 
 5: 
 5: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 5: 
 8: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 8: 
 8: 
 8: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 8: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 8: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 8: 
 8: 
13: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
13: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
13: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
13: 
 9: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 9: 
 9: 
 9: 
 9: 
 9: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 9: 
 9: 
13: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
13: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
13: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
10: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
10: 
11: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
11: 
11: 
11: 
11: 
10: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
10: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
10: 
10: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
11: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
11: 
11: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
10: 
10: 
12: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
12: 
12: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
12: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
12: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
12: 
12: 
12: 
16: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
16: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
16: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
16: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
16: 
16: 
16: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
17: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
17: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
14: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
14: 
14: 
14: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
14: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
14: 
14: 
14: 
17: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
17: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
17: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
17: 
17: 
17: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
15: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
15: 
15: 
15: 
15: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
15: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
15: 
15: 
19: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
19: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
19: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
19: 
19: 
19: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
19: 
18: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
18: 
18: 
18: 
18: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
18: 
18: 
18: 
20: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
20: 
20: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
20: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
20: 
20: 
20: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
20: 
21: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
21: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
21: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
21: 
21: 
21: 
21: 
21: 
22: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
22: 
22: 
22: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
22: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
22: 
22: 
22: 
23: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
23: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
23: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
23: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
23: 
23: 
23: 
23: 
24: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
27: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
24: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
24: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
24: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
24: 
24: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
24: 
24: 
27: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
27: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
27: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
28: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
28: 
28: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
27: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
27: 
28: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
27: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
28: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
28: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
28: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
28: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
25: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
25: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
25: 
25: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
25: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
25: 
25: 
25: 
26: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
26: 
26: 
26: 
26: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
26: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
26: 
26: 
29: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
29: 
29: 
29: 
29: 
29: 
29: 
29: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
30: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
30: 
30: 
30: 
30: 
31: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
30: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
30: 
30: 
31: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
31: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
31: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
31: 
31: 
31: 
31: 
 0: Emitting ninja build file /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu/utils/build.ninja...
 0: Building extension module utils...
 0: Allowing ninja to set a default number of workers... (overridable by setting the environment variable MAX_JOBS=N)
 0: Loading extension module utils...
 4: Loading extension module utils...
 4: Loading extension module utils...
 5: Loading extension module utils...
 4: Loading extension module utils...
 5: Loading extension module utils...
 4: Loading extension module utils...
 5: Loading extension module utils...
 4: Loading extension module utils...
 5: Loading extension module utils...
 4: Loading extension module utils...
 5: Loading extension module utils...
 4: Loading extension module utils...
 5: Loading extension module utils...
 4: Loading extension module utils...
 5: Loading extension module utils...
 5: Loading extension module utils...
13: Loading extension module utils...
13: Loading extension module utils...
 8: Loading extension module utils...
13: Loading extension module utils...
13: Loading extension module utils...
 8: Loading extension module utils...
13: Loading extension module utils...
 8: Loading extension module utils...
 8: Loading extension module utils...
13: Loading extension module utils...
13: Loading extension module utils...
 8: Loading extension module utils...
 8: Loading extension module utils...
 8: Loading extension module utils...
 8: Loading extension module utils...
 9: Loading extension module utils...
 9: Loading extension module utils...
 9: Loading extension module utils...
 9: Loading extension module utils...
 9: Loading extension module utils...
 9: Loading extension module utils...
 9: Loading extension module utils...
 9: Loading extension module utils...
11: Loading extension module utils...
 6: Loading extension module utils...
 7: Loading extension module utils...
11: Loading extension module utils...
11: Loading extension module utils...
11: Loading extension module utils...
10: Loading extension module utils...
16: Loading extension module utils...
11: Loading extension module utils...
10: Loading extension module utils...
16: Loading extension module utils...
11: Loading extension module utils...
10: Loading extension module utils...
16: Loading extension module utils...
11: Loading extension module utils...
10: Loading extension module utils...
16: Loading extension module utils...
11: Loading extension module utils...
10: Loading extension module utils...
16: Loading extension module utils...
10: Loading extension module utils...
19: Loading extension module utils...
16: Loading extension module utils...
10: Loading extension module utils...
16: Loading extension module utils...
16: Loading extension module utils...
10: Loading extension module utils...
12: Loading extension module utils...
12: Loading extension module utils...
12: Loading extension module utils...
12: Loading extension module utils...
12: Loading extension module utils...
12: Loading extension module utils...
12: Loading extension module utils...
12: Loading extension module utils...
19: Loading extension module utils...
19: Loading extension module utils...
17: Loading extension module utils...
17: Loading extension module utils...
14: Loading extension module utils...
17: Loading extension module utils...
19: Loading extension module utils...
14: Loading extension module utils...
17: Loading extension module utils...
19: Loading extension module utils...
14: Loading extension module utils...
17: Loading extension module utils...
19: Loading extension module utils...
17: Loading extension module utils...
14: Loading extension module utils...
19: Loading extension module utils...
17: Loading extension module utils...
14: Loading extension module utils...
19: Loading extension module utils...
14: Loading extension module utils...
17: Loading extension module utils...
14: Loading extension module utils...
14: Loading extension module utils...
15: Loading extension module utils...
15: Loading extension module utils...
15: Loading extension module utils...
15: Loading extension module utils...
15: Loading extension module utils...
15: Loading extension module utils...
15: Loading extension module utils...
15: Loading extension module utils...
18: Loading extension module utils...
18: Loading extension module utils...
18: Loading extension module utils...
18: Loading extension module utils...
18: Loading extension module utils...
18: Loading extension module utils...
18: Loading extension module utils...
18: Loading extension module utils...
20: Loading extension module utils...
20: Loading extension module utils...
20: Loading extension module utils...
20: Loading extension module utils...
20: Loading extension module utils...
20: Loading extension module utils...
21: Loading extension module utils...
20: Loading extension module utils...
21: Loading extension module utils...
20: Loading extension module utils...
21: Loading extension module utils...
27: Loading extension module utils...
21: Loading extension module utils...
21: Loading extension module utils...
21: Loading extension module utils...
21: Loading extension module utils...
27: Loading extension module utils...
21: Loading extension module utils...
27: Loading extension module utils...
27: Loading extension module utils...
27: Loading extension module utils...
27: Loading extension module utils...
27: Loading extension module utils...
27: Loading extension module utils...
24: Loading extension module utils...
24: Loading extension module utils...
24: Loading extension module utils...
24: Loading extension module utils...
24: Loading extension module utils...
24: Loading extension module utils...
23: Loading extension module utils...
22: Loading extension module utils...
24: Loading extension module utils...
23: Loading extension module utils...
22: Loading extension module utils...
23: Loading extension module utils...
22: Loading extension module utils...
23: Loading extension module utils...
22: Loading extension module utils...
24: Loading extension module utils...
23: Loading extension module utils...
22: Loading extension module utils...
23: Loading extension module utils...
22: Loading extension module utils...
23: Loading extension module utils...
22: Loading extension module utils...
23: Loading extension module utils...
28: Loading extension module utils...
22: Loading extension module utils...
28: Loading extension module utils...
28: Loading extension module utils...
28: Loading extension module utils...
28: Loading extension module utils...
28: Loading extension module utils...
28: Loading extension module utils...
28: Loading extension module utils...
25: Loading extension module utils...
25: Loading extension module utils...
25: Loading extension module utils...
25: Loading extension module utils...
25: Loading extension module utils...
26: Loading extension module utils...
25: Loading extension module utils...
26: Loading extension module utils...
25: Loading extension module utils...
25: Loading extension module utils...
26: Loading extension module utils...
26: Loading extension module utils...
26: Loading extension module utils...
26: Loading extension module utils...
26: Loading extension module utils...
26: Loading extension module utils...
29: Loading extension module utils...
29: Loading extension module utils...
29: Loading extension module utils...
29: Loading extension module utils...
29: Loading extension module utils...
29: Loading extension module utils...
29: Loading extension module utils...
29: Loading extension module utils...
30: Loading extension module utils...
30: Loading extension module utils...
30: Loading extension module utils...
30: Loading extension module utils...
30: Loading extension module utils...
30: Loading extension module utils...
31: Loading extension module utils...
31: Loading extension module utils...
30: Loading extension module utils...
31: Loading extension module utils...
30: Loading extension module utils...
31: Loading extension module utils...
31: Loading extension module utils...
31: Loading extension module utils...
31: Loading extension module utils...
31: Loading extension module utils...
34: Loading extension module utils...
39: Loading extension module utils...
43: Loading extension module utils...
44: Loading extension module utils...
32: Loading extension module utils...
48: Loading extension module utils...
32: Loading extension module utils...
32: Loading extension module utils...
32: Loading extension module utils...
32: Loading extension module utils...
32: Loading extension module utils...
32: Loading extension module utils...
34: Loading extension module utils...
34: Loading extension module utils...
34: Loading extension module utils...
34: Loading extension module utils...
34: Loading extension module utils...
34: Loading extension module utils...
34: Loading extension module utils...
52: Loading extension module utils...
53: Loading extension module utils...
33: Loading extension module utils...
33: Loading extension module utils...
33: Loading extension module utils...
33: Loading extension module utils...
33: Loading extension module utils...
33: Loading extension module utils...
33: Loading extension module utils...
39: Loading extension module utils...
33: Loading extension module utils...
39: Loading extension module utils...
39: Loading extension module utils...
39: Loading extension module utils...
39: Loading extension module utils...
39: Loading extension module utils...
39: Loading extension module utils...
36: Loading extension module utils...
36: Loading extension module utils...
36: Loading extension module utils...
36: Loading extension module utils...
36: Loading extension module utils...
36: Loading extension module utils...
35: Loading extension module utils...
36: Loading extension module utils...
36: Loading extension module utils...
35: Loading extension module utils...
35: Loading extension module utils...
 0: Loading extension module utils...
35: Loading extension module utils...
35: Loading extension module utils...
35: Loading extension module utils...
35: Loading extension module utils...
35: Loading extension module utils...
37: Loading extension module utils...
43: Loading extension module utils...
37: Loading extension module utils...
43: Loading extension module utils...
58: Loading extension module utils...
37: Loading extension module utils...
43: Loading extension module utils...
37: Loading extension module utils...
43: Loading extension module utils...
37: Loading extension module utils...
43: Loading extension module utils...
37: Loading extension module utils...
43: Loading extension module utils...
37: Loading extension module utils...
43: Loading extension module utils...
37: Loading extension module utils...
44: Loading extension module utils...
38: Loading extension module utils...
38: Loading extension module utils...
38: Loading extension module utils...
60: Loading extension module utils...
44: Loading extension module utils...
44: Loading extension module utils...
38: Loading extension module utils...
44: Loading extension module utils...
44: Loading extension module utils...
44: Loading extension module utils...
38: Loading extension module utils...
44: Loading extension module utils...
38: Loading extension module utils...
38: Loading extension module utils...
38: Loading extension module utils...
61: Loading extension module utils...
41: Loading extension module utils...
41: Loading extension module utils...
41: Loading extension module utils...
40: Loading extension module utils...
41: Loading extension module utils...
40: Loading extension module utils...
40: Loading extension module utils...
63: Loading extension module utils...
62: Loading extension module utils...
41: Loading extension module utils...
40: Loading extension module utils...
40: Loading extension module utils...
41: Loading extension module utils...
41: Loading extension module utils...
40: Loading extension module utils...
41: Loading extension module utils...
40: Loading extension module utils...
40: Loading extension module utils...
42: Loading extension module utils...
48: Loading extension module utils...
42: Loading extension module utils...
48: Loading extension module utils...
42: Loading extension module utils...
42: Loading extension module utils...
48: Loading extension module utils...
48: Loading extension module utils...
42: Loading extension module utils...
48: Loading extension module utils...
42: Loading extension module utils...
48: Loading extension module utils...
42: Loading extension module utils...
48: Loading extension module utils...
42: Loading extension module utils...
46: Loading extension module utils...
45: Loading extension module utils...
46: Loading extension module utils...
45: Loading extension module utils...
46: Loading extension module utils...
45: Loading extension module utils...
46: Loading extension module utils...
45: Loading extension module utils...
46: Loading extension module utils...
45: Loading extension module utils...
46: Loading extension module utils...
46: Loading extension module utils...
45: Loading extension module utils...
32: Loading extension module utils...
46: Loading extension module utils...
45: Loading extension module utils...
45: Loading extension module utils...
47: Loading extension module utils...
52: Loading extension module utils...
47: Loading extension module utils...
52: Loading extension module utils...
52: Loading extension module utils...
47: Loading extension module utils...
52: Loading extension module utils...
47: Loading extension module utils...
52: Loading extension module utils...
52: Loading extension module utils...
47: Loading extension module utils...
52: Loading extension module utils...
47: Loading extension module utils...
47: Loading extension module utils...
53: Loading extension module utils...
47: Loading extension module utils...
53: Loading extension module utils...
53: Loading extension module utils...
53: Loading extension module utils...
53: Loading extension module utils...
53: Loading extension module utils...
53: Loading extension module utils...
49: Loading extension module utils...
49: Loading extension module utils...
49: Loading extension module utils...
49: Loading extension module utils...
49: Loading extension module utils...
49: Loading extension module utils...
49: Loading extension module utils...
49: Loading extension module utils...
58: Loading extension module utils...
 0: Loading extension module utils...
58: Loading extension module utils...
 0: Loading extension module utils...
 0: Loading extension module utils...
 0: Loading extension module utils...
 0: Loading extension module utils...
 0: Loading extension module utils...
60: Loading extension module utils...
60: Loading extension module utils...
54: Loading extension module utils...
54: Loading extension module utils...
55: Loading extension module utils...
55: Loading extension module utils...
60: Loading extension module utils...
54: Loading extension module utils...
60: Loading extension module utils...
54: Loading extension module utils...
60: Loading extension module utils...
55: Loading extension module utils...
54: Loading extension module utils...
60: Loading extension module utils...
55: Loading extension module utils...
60: Loading extension module utils...
54: Loading extension module utils...
55: Loading extension module utils...
54: Loading extension module utils...
54: Loading extension module utils...
55: Loading extension module utils...
61: Loading extension module utils...
55: Loading extension module utils...
61: Loading extension module utils...
55: Loading extension module utils...
61: Loading extension module utils...
56: Loading extension module utils...
61: Loading extension module utils...
61: Loading extension module utils...
56: Loading extension module utils...
61: Loading extension module utils...
61: Loading extension module utils...
56: Loading extension module utils...
62: Loading extension module utils...
56: Loading extension module utils...
62: Loading extension module utils...
56: Loading extension module utils...
62: Loading extension module utils...
56: Loading extension module utils...
56: Loading extension module utils...
56: Loading extension module utils...
63: Loading extension module utils...
62: Loading extension module utils...
62: Loading extension module utils...
57: Loading extension module utils...
63: Loading extension module utils...
62: Loading extension module utils...
62: Loading extension module utils...
63: Loading extension module utils...
57: Loading extension module utils...
63: Loading extension module utils...
57: Loading extension module utils...
63: Loading extension module utils...
57: Loading extension module utils...
63: Loading extension module utils...
57: Loading extension module utils...
63: Loading extension module utils...
57: Loading extension module utils...
57: Loading extension module utils...
57: Loading extension module utils...
59: Loading extension module utils...
59: Loading extension module utils...
59: Loading extension module utils...
59: Loading extension module utils...
59: Loading extension module utils...
59: Loading extension module utils...
59: Loading extension module utils...
59: Loading extension module utils...
 7: Loading extension module utils...
 6: Loading extension module utils...
 7: Loading extension module utils...
 2: Loading extension module utils...
 7: Loading extension module utils...
 1: Loading extension module utils...
 2: Loading extension module utils...
 6: Loading extension module utils...
 7: Loading extension module utils...
 1: Loading extension module utils...
 2: Loading extension module utils...
 6: Loading extension module utils...
 7: Loading extension module utils...
 2: Loading extension module utils...
 6: Loading extension module utils...
 7: Loading extension module utils...
 2: Loading extension module utils...
 3: Loading extension module utils...
 1: Loading extension module utils...
 7: Loading extension module utils...
 2: Loading extension module utils...
 1: Loading extension module utils...
 3: Loading extension module utils...
 6: Loading extension module utils...
 2: Loading extension module utils...
 1: Loading extension module utils...
 6: Loading extension module utils...
 2: Loading extension module utils...
 6: Loading extension module utils...
 1: Loading extension module utils...
 3: Loading extension module utils...
 1: Loading extension module utils...
 3: Loading extension module utils...
 3: Loading extension module utils...
 1: Loading extension module utils...
 3: Loading extension module utils...
 3: Loading extension module utils...
 3: Loading extension module utils...
41: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
41: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
41: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
41: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
41: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
41: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
41: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
41: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
41: No modifications detected for re-loaded extension module utils, skipping build step...
41: No modifications detected for re-loaded extension module utils, skipping build step...Loading extension module utils...
41: 
41: Loading extension module utils...
41: No modifications detected for re-loaded extension module utils, skipping build step...
41: Loading extension module utils...
41: No modifications detected for re-loaded extension module utils, skipping build step...
41: Loading extension module utils...
41: No modifications detected for re-loaded extension module utils, skipping build step...
41: Loading extension module utils...
41: No modifications detected for re-loaded extension module utils, skipping build step...No modifications detected for re-loaded extension module utils, skipping build step...
41: Loading extension module utils...
41: 
41: Loading extension module utils...
41: No modifications detected for re-loaded extension module utils, skipping build step...
41: Loading extension module utils...
34: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 8: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 8: 
31: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
31: 
31: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
34: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
34: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
29: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
29: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 8: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 8: 
 8: 
 8: 
 8: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
31: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
13: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 9: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
34: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
34: 
34: 
34: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 8: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
29: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
29: 
31: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
31: 
31: 
33: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
33: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
34: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
29: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
29: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
31: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
33: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
13: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 9: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
34: No modifications detected for re-loaded extension module utils, skipping build step...
34: Loading extension module utils...
33: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
13: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 9: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
29: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
33: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
13: 
 9: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
29: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
33: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 9: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 9: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
33: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
13: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
13: 
33: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
13: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 9: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 9: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
13: No modifications detected for re-loaded extension module utils, skipping build step...
13: Loading extension module utils...
13: No modifications detected for re-loaded extension module utils, skipping build step...No modifications detected for re-loaded extension module utils, skipping build step...
13: 
 8: No modifications detected for re-loaded extension module utils, skipping build step...
 8: Loading extension module utils...
13: Loading extension module utils...Loading extension module utils...
13: 
13: No modifications detected for re-loaded extension module utils, skipping build step...
13: Loading extension module utils...
13: No modifications detected for re-loaded extension module utils, skipping build step...
13: Loading extension module utils...
31: No modifications detected for re-loaded extension module utils, skipping build step...No modifications detected for re-loaded extension module utils, skipping build step...
31: 
31: Loading extension module utils...Loading extension module utils...
31: 
33: No modifications detected for re-loaded extension module utils, skipping build step...
33: Loading extension module utils...
 8: No modifications detected for re-loaded extension module utils, skipping build step...
 8: Loading extension module utils...
13: No modifications detected for re-loaded extension module utils, skipping build step...
13: Loading extension module utils...
29: No modifications detected for re-loaded extension module utils, skipping build step...
29: Loading extension module utils...
31: No modifications detected for re-loaded extension module utils, skipping build step...
31: Loading extension module utils...
 8: No modifications detected for re-loaded extension module utils, skipping build step...
 8: Loading extension module utils...
13: No modifications detected for re-loaded extension module utils, skipping build step...
13: Loading extension module utils...
13: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 8: No modifications detected for re-loaded extension module utils, skipping build step...
 8: Loading extension module utils...
31: No modifications detected for re-loaded extension module utils, skipping build step...No modifications detected for re-loaded extension module utils, skipping build step...
31: 
31: Loading extension module utils...Loading extension module utils...No modifications detected for re-loaded extension module utils, skipping build step...
31: 
31: 
13: No modifications detected for re-loaded extension module utils, skipping build step...
 8: No modifications detected for re-loaded extension module utils, skipping build step...
 8: Loading extension module utils...
29: No modifications detected for re-loaded extension module utils, skipping build step...
29: Loading extension module utils...
13: Loading extension module utils...
 8: No modifications detected for re-loaded extension module utils, skipping build step...
 8: Loading extension module utils...No modifications detected for re-loaded extension module utils, skipping build step...
31: Loading extension module utils...
31: No modifications detected for re-loaded extension module utils, skipping build step...
31: Loading extension module utils...
 9: No modifications detected for re-loaded extension module utils, skipping build step...
 9: Loading extension module utils...
 8: 
 8: Loading extension module utils...
31: No modifications detected for re-loaded extension module utils, skipping build step...
31: Loading extension module utils...
 9: No modifications detected for re-loaded extension module utils, skipping build step...
 9: Loading extension module utils...
29: No modifications detected for re-loaded extension module utils, skipping build step...
29: Loading extension module utils...
29: No modifications detected for re-loaded extension module utils, skipping build step...
29: Loading extension module utils...
 8: No modifications detected for re-loaded extension module utils, skipping build step...
 8: Loading extension module utils...
29: No modifications detected for re-loaded extension module utils, skipping build step...
29: Loading extension module utils...
33: No modifications detected for re-loaded extension module utils, skipping build step...
33: Loading extension module utils...
 9: No modifications detected for re-loaded extension module utils, skipping build step...
 9: Loading extension module utils...
29: No modifications detected for re-loaded extension module utils, skipping build step...
29: Loading extension module utils...
 9: No modifications detected for re-loaded extension module utils, skipping build step...
29: No modifications detected for re-loaded extension module utils, skipping build step...
29: Loading extension module utils...
 9: Loading extension module utils...
33: No modifications detected for re-loaded extension module utils, skipping build step...
33: No modifications detected for re-loaded extension module utils, skipping build step...Loading extension module utils...
33: 
 9: No modifications detected for re-loaded extension module utils, skipping build step...
 9: Loading extension module utils...
33: Loading extension module utils...
 9: No modifications detected for re-loaded extension module utils, skipping build step...
 9: Loading extension module utils...
33: No modifications detected for re-loaded extension module utils, skipping build step...
33: Loading extension module utils...
 9: No modifications detected for re-loaded extension module utils, skipping build step...
 9: Loading extension module utils...
33: No modifications detected for re-loaded extension module utils, skipping build step...
33: Loading extension module utils...
29: No modifications detected for re-loaded extension module utils, skipping build step...
29: Loading extension module utils...
33: No modifications detected for re-loaded extension module utils, skipping build step...
33: Loading extension module utils...
33: No modifications detected for re-loaded extension module utils, skipping build step...
33: Loading extension module utils...
 9: No modifications detected for re-loaded extension module utils, skipping build step...
 9: Loading extension module utils...
43: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
43: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
43: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
43: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
43: 
43: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
43: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
43: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
43: No modifications detected for re-loaded extension module utils, skipping build step...
43: No modifications detected for re-loaded extension module utils, skipping build step...Loading extension module utils...
43: 
43: Loading extension module utils...
43: No modifications detected for re-loaded extension module utils, skipping build step...
43: Loading extension module utils...
43: No modifications detected for re-loaded extension module utils, skipping build step...No modifications detected for re-loaded extension module utils, skipping build step...No modifications detected for re-loaded extension module utils, skipping build step...
43: 
43: 
43: Loading extension module utils...Loading extension module utils...Loading extension module utils...
43: 
43: 
43: No modifications detected for re-loaded extension module utils, skipping build step...
43: Loading extension module utils...No modifications detected for re-loaded extension module utils, skipping build step...
43: 
43: Loading extension module utils...
50: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
50: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
50: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
50: 
50: 
50: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
50: 
50: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
50: No modifications detected for re-loaded extension module utils, skipping build step...
50: Loading extension module utils...
50: No modifications detected for re-loaded extension module utils, skipping build step...
50: Loading extension module utils...
50: No modifications detected for re-loaded extension module utils, skipping build step...
50: No modifications detected for re-loaded extension module utils, skipping build step...Loading extension module utils...
50: 
50: Loading extension module utils...
50: No modifications detected for re-loaded extension module utils, skipping build step...
50: Loading extension module utils...No modifications detected for re-loaded extension module utils, skipping build step...
50: 
50: Loading extension module utils...
50: No modifications detected for re-loaded extension module utils, skipping build step...
50: Loading extension module utils...
50: No modifications detected for re-loaded extension module utils, skipping build step...
50: Loading extension module utils...
26: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
26: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
32: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
26: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
26: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
26: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
32: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
32: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
26: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
26: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
26: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
32: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
32: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
32: 
32: 
32: 
26: No modifications detected for re-loaded extension module utils, skipping build step...
26: Loading extension module utils...
32: No modifications detected for re-loaded extension module utils, skipping build step...
32: Loading extension module utils...
58: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
58: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
32: No modifications detected for re-loaded extension module utils, skipping build step...
32: Loading extension module utils...
32: No modifications detected for re-loaded extension module utils, skipping build step...
58: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
58: 
32: Loading extension module utils...
58: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
58: 
32: No modifications detected for re-loaded extension module utils, skipping build step...No modifications detected for re-loaded extension module utils, skipping build step...No modifications detected for re-loaded extension module utils, skipping build step...
32: 
32: 
32: No modifications detected for re-loaded extension module utils, skipping build step...Loading extension module utils...
32: 
32: Loading extension module utils...Loading extension module utils...
32: Loading extension module utils...
32: 
58: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
58: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
58: No modifications detected for re-loaded extension module utils, skipping build step...
58: Loading extension module utils...
32: No modifications detected for re-loaded extension module utils, skipping build step...
32: Loading extension module utils...
26: No modifications detected for re-loaded extension module utils, skipping build step...
26: Loading extension module utils...
58: No modifications detected for re-loaded extension module utils, skipping build step...
58: Loading extension module utils...
58: No modifications detected for re-loaded extension module utils, skipping build step...
58: Loading extension module utils...
58: No modifications detected for re-loaded extension module utils, skipping build step...
58: Loading extension module utils...
26: No modifications detected for re-loaded extension module utils, skipping build step...
26: Loading extension module utils...
26: No modifications detected for re-loaded extension module utils, skipping build step...
58: No modifications detected for re-loaded extension module utils, skipping build step...No modifications detected for re-loaded extension module utils, skipping build step...
58: 
58: Loading extension module utils...
58: Loading extension module utils...No modifications detected for re-loaded extension module utils, skipping build step...
58: 
26: Loading extension module utils...
26: No modifications detected for re-loaded extension module utils, skipping build step...
26: Loading extension module utils...
58: Loading extension module utils...
58: No modifications detected for re-loaded extension module utils, skipping build step...
58: Loading extension module utils...
26: No modifications detected for re-loaded extension module utils, skipping build step...
26: No modifications detected for re-loaded extension module utils, skipping build step...Loading extension module utils...
26: 
26: Loading extension module utils...
26: No modifications detected for re-loaded extension module utils, skipping build step...
26: Loading extension module utils...
27: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
27: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
27: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
27: 
27: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
27: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
27: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
27: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
27: No modifications detected for re-loaded extension module utils, skipping build step...
27: Loading extension module utils...
27: No modifications detected for re-loaded extension module utils, skipping build step...
27: Loading extension module utils...
27: No modifications detected for re-loaded extension module utils, skipping build step...
27: Loading extension module utils...
27: No modifications detected for re-loaded extension module utils, skipping build step...
27: Loading extension module utils...
27: No modifications detected for re-loaded extension module utils, skipping build step...
27: Loading extension module utils...
27: No modifications detected for re-loaded extension module utils, skipping build step...
27: Loading extension module utils...
60: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
60: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
60: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
60: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
27: No modifications detected for re-loaded extension module utils, skipping build step...
27: Loading extension module utils...
27: No modifications detected for re-loaded extension module utils, skipping build step...
27: Loading extension module utils...
60: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
60: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
60: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
60: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
60: No modifications detected for re-loaded extension module utils, skipping build step...
60: Loading extension module utils...
60: No modifications detected for re-loaded extension module utils, skipping build step...
60: Loading extension module utils...
60: No modifications detected for re-loaded extension module utils, skipping build step...No modifications detected for re-loaded extension module utils, skipping build step...
60: 
60: Loading extension module utils...Loading extension module utils...
60: 
60: No modifications detected for re-loaded extension module utils, skipping build step...
60: Loading extension module utils...
53: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
60: No modifications detected for re-loaded extension module utils, skipping build step...
60: No modifications detected for re-loaded extension module utils, skipping build step...Loading extension module utils...
60: 
60: Loading extension module utils...
53: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
53: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
60: No modifications detected for re-loaded extension module utils, skipping build step...
60: Loading extension module utils...
53: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
53: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
53: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
53: 
53: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
53: No modifications detected for re-loaded extension module utils, skipping build step...
53: Loading extension module utils...
53: No modifications detected for re-loaded extension module utils, skipping build step...
53: No modifications detected for re-loaded extension module utils, skipping build step...No modifications detected for re-loaded extension module utils, skipping build step...Loading extension module utils...
53: 
53: 
53: Loading extension module utils...Loading extension module utils...
53: 
53: No modifications detected for re-loaded extension module utils, skipping build step...
53: Loading extension module utils...
53: No modifications detected for re-loaded extension module utils, skipping build step...
53: Loading extension module utils...
53: No modifications detected for re-loaded extension module utils, skipping build step...
53: Loading extension module utils...
53: No modifications detected for re-loaded extension module utils, skipping build step...
53: Loading extension module utils...
 7: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 7: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 7: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 7: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 7: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 7: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 7: 
 7: 
 7: No modifications detected for re-loaded extension module utils, skipping build step...
 7: Loading extension module utils...
 7: No modifications detected for re-loaded extension module utils, skipping build step...No modifications detected for re-loaded extension module utils, skipping build step...No modifications detected for re-loaded extension module utils, skipping build step...
 7: 
 7: 
 7: Loading extension module utils...Loading extension module utils...
 7: Loading extension module utils...
 7: 
 7: No modifications detected for re-loaded extension module utils, skipping build step...
 7: Loading extension module utils...
 7: No modifications detected for re-loaded extension module utils, skipping build step...No modifications detected for re-loaded extension module utils, skipping build step...
 7: No modifications detected for re-loaded extension module utils, skipping build step...
 7: 
 7: Loading extension module utils...Loading extension module utils...
 7: 
 7: Loading extension module utils...
34: No modifications detected for re-loaded extension module utils, skipping build step...
34: Loading extension module utils...
34: No modifications detected for re-loaded extension module utils, skipping build step...
34: Loading extension module utils...
34: No modifications detected for re-loaded extension module utils, skipping build step...
34: No modifications detected for re-loaded extension module utils, skipping build step...Loading extension module utils...
34: 
34: Loading extension module utils...
34: No modifications detected for re-loaded extension module utils, skipping build step...
34: Loading extension module utils...
34: No modifications detected for re-loaded extension module utils, skipping build step...
34: Loading extension module utils...
34: No modifications detected for re-loaded extension module utils, skipping build step...
34: Loading extension module utils...
11: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
11: 
11: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
11: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
38: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
11: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
11: 
38: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
11: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
11: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
38: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
19: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
19: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
38: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
19: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
38: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
38: 
38: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
38: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
19: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
19: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
19: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
19: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
19: 
19: No modifications detected for re-loaded extension module utils, skipping build step...No modifications detected for re-loaded extension module utils, skipping build step...
19: 
19: Loading extension module utils...
19: Loading extension module utils...
19: No modifications detected for re-loaded extension module utils, skipping build step...
19: Loading extension module utils...
19: No modifications detected for re-loaded extension module utils, skipping build step...
19: Loading extension module utils...
11: No modifications detected for re-loaded extension module utils, skipping build step...
11: Loading extension module utils...
19: No modifications detected for re-loaded extension module utils, skipping build step...No modifications detected for re-loaded extension module utils, skipping build step...
19: 
19: Loading extension module utils...No modifications detected for re-loaded extension module utils, skipping build step...Loading extension module utils...
19: 
19: 
19: No modifications detected for re-loaded extension module utils, skipping build step...Loading extension module utils...
19: 
19: Loading extension module utils...
11: No modifications detected for re-loaded extension module utils, skipping build step...
11: Loading extension module utils...
39: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
39: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
38: No modifications detected for re-loaded extension module utils, skipping build step...
38: Loading extension module utils...No modifications detected for re-loaded extension module utils, skipping build step...
38: 
38: Loading extension module utils...
11: No modifications detected for re-loaded extension module utils, skipping build step...
11: Loading extension module utils...
39: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
39: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
39: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
11: No modifications detected for re-loaded extension module utils, skipping build step...
11: Loading extension module utils...
39: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
39: 
39: 
11: No modifications detected for re-loaded extension module utils, skipping build step...
38: No modifications detected for re-loaded extension module utils, skipping build step...No modifications detected for re-loaded extension module utils, skipping build step...
38: 
38: Loading extension module utils...Loading extension module utils...No modifications detected for re-loaded extension module utils, skipping build step...
38: 
38: 
11: Loading extension module utils...
11: No modifications detected for re-loaded extension module utils, skipping build step...
11: Loading extension module utils...
38: Loading extension module utils...
11: No modifications detected for re-loaded extension module utils, skipping build step...
11: Loading extension module utils...
38: No modifications detected for re-loaded extension module utils, skipping build step...
38: Loading extension module utils...
38: No modifications detected for re-loaded extension module utils, skipping build step...
38: Loading extension module utils...
39: No modifications detected for re-loaded extension module utils, skipping build step...
39: Loading extension module utils...
38: No modifications detected for re-loaded extension module utils, skipping build step...
38: Loading extension module utils...
39: No modifications detected for re-loaded extension module utils, skipping build step...
39: Loading extension module utils...
11: No modifications detected for re-loaded extension module utils, skipping build step...
11: Loading extension module utils...
39: No modifications detected for re-loaded extension module utils, skipping build step...
39: Loading extension module utils...No modifications detected for re-loaded extension module utils, skipping build step...
39: 
39: No modifications detected for re-loaded extension module utils, skipping build step...Loading extension module utils...
39: 
39: Loading extension module utils...
39: No modifications detected for re-loaded extension module utils, skipping build step...
39: Loading extension module utils...
39: No modifications detected for re-loaded extension module utils, skipping build step...No modifications detected for re-loaded extension module utils, skipping build step...
39: 
39: Loading extension module utils...Loading extension module utils...
39: 
 0: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 0: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 0: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 0: 
 0: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 0: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 0: No modifications detected for re-loaded extension module utils, skipping build step...
 0: Loading extension module utils...
 0: No modifications detected for re-loaded extension module utils, skipping build step...
 0: Loading extension module utils...
 0: No modifications detected for re-loaded extension module utils, skipping build step...No modifications detected for re-loaded extension module utils, skipping build step...No modifications detected for re-loaded extension module utils, skipping build step...
 0: 
 0: 
 0: Loading extension module utils...
 0: Loading extension module utils...Loading extension module utils...
 0: 
 0: No modifications detected for re-loaded extension module utils, skipping build step...
54: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 0: Loading extension module utils...
54: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
54: 
 0: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
54: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
54: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
54: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
54: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 0: No modifications detected for re-loaded extension module utils, skipping build step...
54: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 0: Loading extension module utils...
54: No modifications detected for re-loaded extension module utils, skipping build step...
54: Loading extension module utils...
54: No modifications detected for re-loaded extension module utils, skipping build step...
54: Loading extension module utils...
54: No modifications detected for re-loaded extension module utils, skipping build step...
54: Loading extension module utils...
54: No modifications detected for re-loaded extension module utils, skipping build step...
54: Loading extension module utils...
54: No modifications detected for re-loaded extension module utils, skipping build step...
54: Loading extension module utils...
54: No modifications detected for re-loaded extension module utils, skipping build step...
54: Loading extension module utils...
54: No modifications detected for re-loaded extension module utils, skipping build step...
54: Loading extension module utils...
54: No modifications detected for re-loaded extension module utils, skipping build step...
63: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
63: 
63: 
54: Loading extension module utils...
36: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
36: 
63: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
63: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
63: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
63: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
36: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
36: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
36: 
36: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
36: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
36: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
47: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
47: 
47: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
47: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
63: No modifications detected for re-loaded extension module utils, skipping build step...No modifications detected for re-loaded extension module utils, skipping build step...
63: 
63: Loading extension module utils...Loading extension module utils...
63: 
47: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
47: 
47: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
47: 
63: No modifications detected for re-loaded extension module utils, skipping build step...No modifications detected for re-loaded extension module utils, skipping build step...
63: 
63: Loading extension module utils...Loading extension module utils...
63: 
63: No modifications detected for re-loaded extension module utils, skipping build step...
63: Loading extension module utils...
63: No modifications detected for re-loaded extension module utils, skipping build step...
63: Loading extension module utils...
63: No modifications detected for re-loaded extension module utils, skipping build step...
63: Loading extension module utils...
63: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
17: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
63: No modifications detected for re-loaded extension module utils, skipping build step...
63: Loading extension module utils...
17: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
17: 
17: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
17: 
17: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
17: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
17: 
47: No modifications detected for re-loaded extension module utils, skipping build step...
47: Loading extension module utils...
36: No modifications detected for re-loaded extension module utils, skipping build step...
36: Loading extension module utils...
45: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
45: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
36: No modifications detected for re-loaded extension module utils, skipping build step...
36: Loading extension module utils...
45: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
45: 
47: No modifications detected for re-loaded extension module utils, skipping build step...
47: Loading extension module utils...
17: No modifications detected for re-loaded extension module utils, skipping build step...
17: Loading extension module utils...
45: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
45: 
45: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
36: No modifications detected for re-loaded extension module utils, skipping build step...No modifications detected for re-loaded extension module utils, skipping build step...
36: 
47: No modifications detected for re-loaded extension module utils, skipping build step...
47: Loading extension module utils...
45: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
36: No modifications detected for re-loaded extension module utils, skipping build step...Loading extension module utils...Loading extension module utils...
36: 
36: 
36: Loading extension module utils...
47: No modifications detected for re-loaded extension module utils, skipping build step...
47: Loading extension module utils...
47: No modifications detected for re-loaded extension module utils, skipping build step...
36: No modifications detected for re-loaded extension module utils, skipping build step...
36: Loading extension module utils...No modifications detected for re-loaded extension module utils, skipping build step...
47: Loading extension module utils...
47: No modifications detected for re-loaded extension module utils, skipping build step...
47: Loading extension module utils...
36: 
36: Loading extension module utils...
47: No modifications detected for re-loaded extension module utils, skipping build step...
47: Loading extension module utils...
47: No modifications detected for re-loaded extension module utils, skipping build step...
47: Loading extension module utils...
36: No modifications detected for re-loaded extension module utils, skipping build step...
36: Loading extension module utils...
17: No modifications detected for re-loaded extension module utils, skipping build step...
17: Loading extension module utils...
48: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
17: No modifications detected for re-loaded extension module utils, skipping build step...No modifications detected for re-loaded extension module utils, skipping build step...
17: 
17: Loading extension module utils...Loading extension module utils...
17: 
48: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
48: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
17: No modifications detected for re-loaded extension module utils, skipping build step...
17: No modifications detected for re-loaded extension module utils, skipping build step...Loading extension module utils...
17: 
17: Loading extension module utils...
48: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
48: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
48: 
45: No modifications detected for re-loaded extension module utils, skipping build step...
45: Loading extension module utils...
17: No modifications detected for re-loaded extension module utils, skipping build step...
17: Loading extension module utils...
48: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
48: 
17: No modifications detected for re-loaded extension module utils, skipping build step...
17: Loading extension module utils...
48: No modifications detected for re-loaded extension module utils, skipping build step...
48: Loading extension module utils...
45: No modifications detected for re-loaded extension module utils, skipping build step...
45: Loading extension module utils...
48: No modifications detected for re-loaded extension module utils, skipping build step...
48: Loading extension module utils...
45: No modifications detected for re-loaded extension module utils, skipping build step...
45: Loading extension module utils...
48: No modifications detected for re-loaded extension module utils, skipping build step...No modifications detected for re-loaded extension module utils, skipping build step...
48: 
48: Loading extension module utils...Loading extension module utils...
48: 
45: No modifications detected for re-loaded extension module utils, skipping build step...
45: Loading extension module utils...
48: No modifications detected for re-loaded extension module utils, skipping build step...No modifications detected for re-loaded extension module utils, skipping build step...No modifications detected for re-loaded extension module utils, skipping build step...
48: 
48: 
45: No modifications detected for re-loaded extension module utils, skipping build step...
45: Loading extension module utils...
48: Loading extension module utils...Loading extension module utils...Loading extension module utils...
48: No modifications detected for re-loaded extension module utils, skipping build step...
48: 
48: 
48: Loading extension module utils...
45: No modifications detected for re-loaded extension module utils, skipping build step...
45: Loading extension module utils...
45: No modifications detected for re-loaded extension module utils, skipping build step...
45: Loading extension module utils...No modifications detected for re-loaded extension module utils, skipping build step...
45: 
45: Loading extension module utils...
57: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
57: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
57: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
57: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
57: 
57: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
57: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
57: 
57: No modifications detected for re-loaded extension module utils, skipping build step...
57: Loading extension module utils...
57: No modifications detected for re-loaded extension module utils, skipping build step...
57: Loading extension module utils...
57: No modifications detected for re-loaded extension module utils, skipping build step...
57: Loading extension module utils...
57: No modifications detected for re-loaded extension module utils, skipping build step...
57: Loading extension module utils...No modifications detected for re-loaded extension module utils, skipping build step...
57: 
57: Loading extension module utils...
57: No modifications detected for re-loaded extension module utils, skipping build step...
57: Loading extension module utils...
57: No modifications detected for re-loaded extension module utils, skipping build step...
57: Loading extension module utils...
57: No modifications detected for re-loaded extension module utils, skipping build step...
57: Loading extension module utils...
49: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
49: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
49: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
49: 
49: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
49: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
49: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
49: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
49: No modifications detected for re-loaded extension module utils, skipping build step...
49: Loading extension module utils...
49: No modifications detected for re-loaded extension module utils, skipping build step...
49: Loading extension module utils...
30: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
30: 
30: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
30: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
30: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
30: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
30: 
30: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
49: No modifications detected for re-loaded extension module utils, skipping build step...
49: Loading extension module utils...
49: No modifications detected for re-loaded extension module utils, skipping build step...
49: Loading extension module utils...
49: No modifications detected for re-loaded extension module utils, skipping build step...
49: Loading extension module utils...
49: No modifications detected for re-loaded extension module utils, skipping build step...
49: Loading extension module utils...
49: No modifications detected for re-loaded extension module utils, skipping build step...
49: Loading extension module utils...
49: No modifications detected for re-loaded extension module utils, skipping build step...
49: Loading extension module utils...
30: No modifications detected for re-loaded extension module utils, skipping build step...
30: Loading extension module utils...
30: No modifications detected for re-loaded extension module utils, skipping build step...
30: Loading extension module utils...
30: No modifications detected for re-loaded extension module utils, skipping build step...
30: Loading extension module utils...
30: No modifications detected for re-loaded extension module utils, skipping build step...
30: Loading extension module utils...
30: No modifications detected for re-loaded extension module utils, skipping build step...
 2: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
30: Loading extension module utils...
30: No modifications detected for re-loaded extension module utils, skipping build step...
30: Loading extension module utils...
30: No modifications detected for re-loaded extension module utils, skipping build step...
30: Loading extension module utils...
 2: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 2: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 2: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
30: No modifications detected for re-loaded extension module utils, skipping build step...
30: Loading extension module utils...
 2: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 2: 
 2: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 2: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 2: No modifications detected for re-loaded extension module utils, skipping build step...
 2: Loading extension module utils...
 3: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 2: No modifications detected for re-loaded extension module utils, skipping build step...
 2: Loading extension module utils...
 2: No modifications detected for re-loaded extension module utils, skipping build step...
 2: Loading extension module utils...
 2: No modifications detected for re-loaded extension module utils, skipping build step...
 2: Loading extension module utils...
 3: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 3: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 3: 
 3: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 2: No modifications detected for re-loaded extension module utils, skipping build step...No modifications detected for re-loaded extension module utils, skipping build step...
 3: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 3: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 2: 
 2: Loading extension module utils...Loading extension module utils...
 3: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 2: 
 2: No modifications detected for re-loaded extension module utils, skipping build step...
 2: Loading extension module utils...
 2: No modifications detected for re-loaded extension module utils, skipping build step...
 2: Loading extension module utils...
55: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
55: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 1: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 1: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
55: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 1: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 1: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 1: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 3: No modifications detected for re-loaded extension module utils, skipping build step...
 3: Loading extension module utils...
 3: No modifications detected for re-loaded extension module utils, skipping build step...
 1: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
55: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
55: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 3: Loading extension module utils...
 1: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
55: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 1: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
55: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
55: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 3: No modifications detected for re-loaded extension module utils, skipping build step...
 3: Loading extension module utils...
 3: No modifications detected for re-loaded extension module utils, skipping build step...
 3: Loading extension module utils...
 3: No modifications detected for re-loaded extension module utils, skipping build step...
 3: Loading extension module utils...
 3: No modifications detected for re-loaded extension module utils, skipping build step...
 3: Loading extension module utils...
 3: No modifications detected for re-loaded extension module utils, skipping build step...
 3: Loading extension module utils...
 3: No modifications detected for re-loaded extension module utils, skipping build step...
 3: Loading extension module utils...
55: No modifications detected for re-loaded extension module utils, skipping build step...No modifications detected for re-loaded extension module utils, skipping build step...
55: 
55: Loading extension module utils...Loading extension module utils...
55: 
55: No modifications detected for re-loaded extension module utils, skipping build step...
55: Loading extension module utils...
 1: No modifications detected for re-loaded extension module utils, skipping build step...
 1: Loading extension module utils...
 1: No modifications detected for re-loaded extension module utils, skipping build step...
 1: Loading extension module utils...
55: No modifications detected for re-loaded extension module utils, skipping build step...
55: Loading extension module utils...
 1: No modifications detected for re-loaded extension module utils, skipping build step...
 1: Loading extension module utils...
55: No modifications detected for re-loaded extension module utils, skipping build step...
55: Loading extension module utils...
 1: No modifications detected for re-loaded extension module utils, skipping build step...No modifications detected for re-loaded extension module utils, skipping build step...
 1: 
 1: Loading extension module utils...Loading extension module utils...
 1: 
55: No modifications detected for re-loaded extension module utils, skipping build step...No modifications detected for re-loaded extension module utils, skipping build step...
55: 
 1: No modifications detected for re-loaded extension module utils, skipping build step...
55: Loading extension module utils...Loading extension module utils...
55: 
 1: Loading extension module utils...
 1: No modifications detected for re-loaded extension module utils, skipping build step...
 1: Loading extension module utils...
55: No modifications detected for re-loaded extension module utils, skipping build step...
55: Loading extension module utils...
 1: No modifications detected for re-loaded extension module utils, skipping build step...
 1: Loading extension module utils...
59: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
59: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
59: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
59: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
59: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
59: 
59: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
59: 
59: No modifications detected for re-loaded extension module utils, skipping build step...
59: Loading extension module utils...
59: No modifications detected for re-loaded extension module utils, skipping build step...
59: Loading extension module utils...
59: No modifications detected for re-loaded extension module utils, skipping build step...
59: Loading extension module utils...
59: No modifications detected for re-loaded extension module utils, skipping build step...
59: Loading extension module utils...
59: No modifications detected for re-loaded extension module utils, skipping build step...
59: Loading extension module utils...No modifications detected for re-loaded extension module utils, skipping build step...
59: 
59: Loading extension module utils...
59: No modifications detected for re-loaded extension module utils, skipping build step...
59: Loading extension module utils...
59: No modifications detected for re-loaded extension module utils, skipping build step...
59: Loading extension module utils...
35: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
35: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
35: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
35: 
35: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
35: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
35: 
35: 
35: No modifications detected for re-loaded extension module utils, skipping build step...
35: Loading extension module utils...
35: No modifications detected for re-loaded extension module utils, skipping build step...
35: Loading extension module utils...
35: No modifications detected for re-loaded extension module utils, skipping build step...
35: Loading extension module utils...
35: No modifications detected for re-loaded extension module utils, skipping build step...
35: Loading extension module utils...
35: No modifications detected for re-loaded extension module utils, skipping build step...
35: Loading extension module utils...
35: No modifications detected for re-loaded extension module utils, skipping build step...
35: Loading extension module utils...
35: No modifications detected for re-loaded extension module utils, skipping build step...
35: Loading extension module utils...
35: No modifications detected for re-loaded extension module utils, skipping build step...
35: Loading extension module utils...
56: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
56: 
56: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
56: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
56: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
56: 
56: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
56: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
56: No modifications detected for re-loaded extension module utils, skipping build step...
56: Loading extension module utils...
56: No modifications detected for re-loaded extension module utils, skipping build step...
56: Loading extension module utils...No modifications detected for re-loaded extension module utils, skipping build step...
56: 
56: Loading extension module utils...
56: No modifications detected for re-loaded extension module utils, skipping build step...
56: Loading extension module utils...No modifications detected for re-loaded extension module utils, skipping build step...
56: 
56: Loading extension module utils...
56: No modifications detected for re-loaded extension module utils, skipping build step...
56: Loading extension module utils...
56: No modifications detected for re-loaded extension module utils, skipping build step...
56: Loading extension module utils...
56: No modifications detected for re-loaded extension module utils, skipping build step...
56: Loading extension module utils...
51: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
51: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
51: 
51: 
51: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
51: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
51: 
51: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
51: No modifications detected for re-loaded extension module utils, skipping build step...
51: Loading extension module utils...
51: No modifications detected for re-loaded extension module utils, skipping build step...No modifications detected for re-loaded extension module utils, skipping build step...
51: 
51: Loading extension module utils...Loading extension module utils...
51: 
51: No modifications detected for re-loaded extension module utils, skipping build step...
51: No modifications detected for re-loaded extension module utils, skipping build step...Loading extension module utils...
51: 
51: Loading extension module utils...
51: No modifications detected for re-loaded extension module utils, skipping build step...
51: No modifications detected for re-loaded extension module utils, skipping build step...Loading extension module utils...
51: 
51: Loading extension module utils...
51: No modifications detected for re-loaded extension module utils, skipping build step...
51: Loading extension module utils...
37: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
37: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
37: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
37: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
37: 
37: 
37: 
37: 
37: No modifications detected for re-loaded extension module utils, skipping build step...
37: Loading extension module utils...
37: No modifications detected for re-loaded extension module utils, skipping build step...
37: No modifications detected for re-loaded extension module utils, skipping build step...Loading extension module utils...
37: 
37: Loading extension module utils...
37: No modifications detected for re-loaded extension module utils, skipping build step...
37: No modifications detected for re-loaded extension module utils, skipping build step...Loading extension module utils...
37: 
37: Loading extension module utils...
37: No modifications detected for re-loaded extension module utils, skipping build step...
37: Loading extension module utils...No modifications detected for re-loaded extension module utils, skipping build step...
37: 
37: Loading extension module utils...
37: No modifications detected for re-loaded extension module utils, skipping build step...
37: Loading extension module utils...
 6: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 6: 
 6: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 6: 
 6: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 6: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 6: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 6: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 6: No modifications detected for re-loaded extension module utils, skipping build step...
 6: Loading extension module utils...
 6: No modifications detected for re-loaded extension module utils, skipping build step...
 6: Loading extension module utils...
 6: No modifications detected for re-loaded extension module utils, skipping build step...
 6: Loading extension module utils...
 6: No modifications detected for re-loaded extension module utils, skipping build step...
 6: Loading extension module utils...
 6: No modifications detected for re-loaded extension module utils, skipping build step...
 6: Loading extension module utils...
 6: No modifications detected for re-loaded extension module utils, skipping build step...
 6: Loading extension module utils...
 6: No modifications detected for re-loaded extension module utils, skipping build step...
 6: No modifications detected for re-loaded extension module utils, skipping build step...Loading extension module utils...
 6: 
 6: Loading extension module utils...
14: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
14: 
14: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
14: 
14: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
14: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
14: 
14: 
14: No modifications detected for re-loaded extension module utils, skipping build step...
14: Loading extension module utils...
14: No modifications detected for re-loaded extension module utils, skipping build step...
14: Loading extension module utils...No modifications detected for re-loaded extension module utils, skipping build step...
14: 
14: Loading extension module utils...
14: No modifications detected for re-loaded extension module utils, skipping build step...
14: Loading extension module utils...
14: No modifications detected for re-loaded extension module utils, skipping build step...
14: Loading extension module utils...
14: No modifications detected for re-loaded extension module utils, skipping build step...
14: No modifications detected for re-loaded extension module utils, skipping build step...Loading extension module utils...
14: 
14: Loading extension module utils...
14: No modifications detected for re-loaded extension module utils, skipping build step...
14: Loading extension module utils...
 4: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 4: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 4: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 4: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 4: 
 4: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 4: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 4: 
 4: No modifications detected for re-loaded extension module utils, skipping build step...
 4: Loading extension module utils...
 4: No modifications detected for re-loaded extension module utils, skipping build step...
 4: Loading extension module utils...
 4: No modifications detected for re-loaded extension module utils, skipping build step...No modifications detected for re-loaded extension module utils, skipping build step...
 4: 
 4: Loading extension module utils...Loading extension module utils...
 4: 
 4: No modifications detected for re-loaded extension module utils, skipping build step...No modifications detected for re-loaded extension module utils, skipping build step...
 4: No modifications detected for re-loaded extension module utils, skipping build step...
 4: Loading extension module utils...
 4: 
 4: Loading extension module utils...Loading extension module utils...
 4: 
 4: No modifications detected for re-loaded extension module utils, skipping build step...
 4: Loading extension module utils...
61: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
61: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
61: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
61: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
61: 
61: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
61: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
61: 
61: No modifications detected for re-loaded extension module utils, skipping build step...
61: Loading extension module utils...
61: No modifications detected for re-loaded extension module utils, skipping build step...
61: Loading extension module utils...No modifications detected for re-loaded extension module utils, skipping build step...
61: 
61: Loading extension module utils...
61: No modifications detected for re-loaded extension module utils, skipping build step...No modifications detected for re-loaded extension module utils, skipping build step...
61: 
61: Loading extension module utils...Loading extension module utils...
61: 
61: No modifications detected for re-loaded extension module utils, skipping build step...
61: Loading extension module utils...
61: No modifications detected for re-loaded extension module utils, skipping build step...
61: Loading extension module utils...
61: No modifications detected for re-loaded extension module utils, skipping build step...
61: Loading extension module utils...
 5: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 5: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 5: 
 5: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 5: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 5: 
 5: 
 5: 
 5: No modifications detected for re-loaded extension module utils, skipping build step...
 5: Loading extension module utils...
 5: No modifications detected for re-loaded extension module utils, skipping build step...
 5: Loading extension module utils...
 5: No modifications detected for re-loaded extension module utils, skipping build step...
 5: Loading extension module utils...
 5: No modifications detected for re-loaded extension module utils, skipping build step...
 5: Loading extension module utils...
 5: No modifications detected for re-loaded extension module utils, skipping build step...
 5: Loading extension module utils...No modifications detected for re-loaded extension module utils, skipping build step...
 5: 
 5: Loading extension module utils...
 5: No modifications detected for re-loaded extension module utils, skipping build step...
 5: Loading extension module utils...
 5: No modifications detected for re-loaded extension module utils, skipping build step...
 5: Loading extension module utils...
15: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
15: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
15: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
15: 
15: 
15: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
15: 
15: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
15: No modifications detected for re-loaded extension module utils, skipping build step...
15: Loading extension module utils...
15: No modifications detected for re-loaded extension module utils, skipping build step...
15: Loading extension module utils...
15: No modifications detected for re-loaded extension module utils, skipping build step...
15: Loading extension module utils...
15: No modifications detected for re-loaded extension module utils, skipping build step...
15: Loading extension module utils...
15: No modifications detected for re-loaded extension module utils, skipping build step...
15: Loading extension module utils...
15: No modifications detected for re-loaded extension module utils, skipping build step...
15: Loading extension module utils...
15: No modifications detected for re-loaded extension module utils, skipping build step...
15: Loading extension module utils...
15: No modifications detected for re-loaded extension module utils, skipping build step...
15: Loading extension module utils...
25: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
25: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
25: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
25: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
25: 
25: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
25: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
25: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
25: No modifications detected for re-loaded extension module utils, skipping build step...
25: Loading extension module utils...
25: No modifications detected for re-loaded extension module utils, skipping build step...
25: Loading extension module utils...
25: No modifications detected for re-loaded extension module utils, skipping build step...
25: Loading extension module utils...
25: No modifications detected for re-loaded extension module utils, skipping build step...
25: Loading extension module utils...
25: No modifications detected for re-loaded extension module utils, skipping build step...
25: Loading extension module utils...
25: No modifications detected for re-loaded extension module utils, skipping build step...
25: Loading extension module utils...No modifications detected for re-loaded extension module utils, skipping build step...
25: 
25: Loading extension module utils...
25: No modifications detected for re-loaded extension module utils, skipping build step...
25: Loading extension module utils...
42: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
42: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
42: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
42: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
42: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
42: 
42: 
42: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
42: No modifications detected for re-loaded extension module utils, skipping build step...
42: Loading extension module utils...
42: No modifications detected for re-loaded extension module utils, skipping build step...
42: Loading extension module utils...
42: No modifications detected for re-loaded extension module utils, skipping build step...
42: Loading extension module utils...
42: No modifications detected for re-loaded extension module utils, skipping build step...
42: No modifications detected for re-loaded extension module utils, skipping build step...Loading extension module utils...
42: 
42: Loading extension module utils...
42: No modifications detected for re-loaded extension module utils, skipping build step...
42: No modifications detected for re-loaded extension module utils, skipping build step...Loading extension module utils...
42: 
42: Loading extension module utils...
42: No modifications detected for re-loaded extension module utils, skipping build step...
42: Loading extension module utils...
20: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
20: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
20: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
20: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
20: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
20: 
20: 
20: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
20: No modifications detected for re-loaded extension module utils, skipping build step...
20: Loading extension module utils...
20: No modifications detected for re-loaded extension module utils, skipping build step...
20: Loading extension module utils...
20: No modifications detected for re-loaded extension module utils, skipping build step...
20: Loading extension module utils...
20: No modifications detected for re-loaded extension module utils, skipping build step...
20: Loading extension module utils...
20: No modifications detected for re-loaded extension module utils, skipping build step...
20: Loading extension module utils...
20: No modifications detected for re-loaded extension module utils, skipping build step...
20: Loading extension module utils...
20: No modifications detected for re-loaded extension module utils, skipping build step...
20: Loading extension module utils...
20: No modifications detected for re-loaded extension module utils, skipping build step...
20: Loading extension module utils...
23: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
23: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
23: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
23: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
23: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
23: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
23: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
23: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
23: No modifications detected for re-loaded extension module utils, skipping build step...
23: Loading extension module utils...
23: No modifications detected for re-loaded extension module utils, skipping build step...
23: Loading extension module utils...
23: No modifications detected for re-loaded extension module utils, skipping build step...
23: Loading extension module utils...
23: No modifications detected for re-loaded extension module utils, skipping build step...
23: Loading extension module utils...
23: No modifications detected for re-loaded extension module utils, skipping build step...
23: Loading extension module utils...
23: No modifications detected for re-loaded extension module utils, skipping build step...
23: Loading extension module utils...No modifications detected for re-loaded extension module utils, skipping build step...
23: 
23: Loading extension module utils...
23: No modifications detected for re-loaded extension module utils, skipping build step...
23: Loading extension module utils...
12: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
12: 
12: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
12: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
12: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
12: 
12: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
12: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
12: No modifications detected for re-loaded extension module utils, skipping build step...
12: Loading extension module utils...
12: No modifications detected for re-loaded extension module utils, skipping build step...
12: Loading extension module utils...
12: No modifications detected for re-loaded extension module utils, skipping build step...
12: Loading extension module utils...
12: No modifications detected for re-loaded extension module utils, skipping build step...
12: Loading extension module utils...
12: No modifications detected for re-loaded extension module utils, skipping build step...
12: Loading extension module utils...
12: No modifications detected for re-loaded extension module utils, skipping build step...
12: Loading extension module utils...No modifications detected for re-loaded extension module utils, skipping build step...
12: 
12: Loading extension module utils...
12: No modifications detected for re-loaded extension module utils, skipping build step...
12: Loading extension module utils...
24: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
24: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
24: 
24: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
24: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
24: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
24: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
24: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
24: No modifications detected for re-loaded extension module utils, skipping build step...
24: Loading extension module utils...
24: No modifications detected for re-loaded extension module utils, skipping build step...
24: Loading extension module utils...
24: No modifications detected for re-loaded extension module utils, skipping build step...
24: Loading extension module utils...
24: No modifications detected for re-loaded extension module utils, skipping build step...
24: Loading extension module utils...
24: No modifications detected for re-loaded extension module utils, skipping build step...
24: Loading extension module utils...
24: No modifications detected for re-loaded extension module utils, skipping build step...
24: Loading extension module utils...
24: No modifications detected for re-loaded extension module utils, skipping build step...
24: Loading extension module utils...
24: No modifications detected for re-loaded extension module utils, skipping build step...
24: Loading extension module utils...
22: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
22: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
22: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
22: 
22: 
22: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
22: 
22: 
22: No modifications detected for re-loaded extension module utils, skipping build step...
22: Loading extension module utils...
22: No modifications detected for re-loaded extension module utils, skipping build step...
22: Loading extension module utils...
22: No modifications detected for re-loaded extension module utils, skipping build step...
22: Loading extension module utils...
22: No modifications detected for re-loaded extension module utils, skipping build step...
22: Loading extension module utils...
22: No modifications detected for re-loaded extension module utils, skipping build step...
22: Loading extension module utils...
22: No modifications detected for re-loaded extension module utils, skipping build step...
22: Loading extension module utils...No modifications detected for re-loaded extension module utils, skipping build step...
22: 
22: Loading extension module utils...
22: No modifications detected for re-loaded extension module utils, skipping build step...
22: Loading extension module utils...
52: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
52: 
52: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
52: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
52: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
52: 
52: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
52: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
52: No modifications detected for re-loaded extension module utils, skipping build step...
52: Loading extension module utils...
52: No modifications detected for re-loaded extension module utils, skipping build step...
52: Loading extension module utils...
52: No modifications detected for re-loaded extension module utils, skipping build step...
52: No modifications detected for re-loaded extension module utils, skipping build step...No modifications detected for re-loaded extension module utils, skipping build step...Loading extension module utils...
52: 
52: 
52: Loading extension module utils...Loading extension module utils...
52: 
52: No modifications detected for re-loaded extension module utils, skipping build step...
52: Loading extension module utils...
52: No modifications detected for re-loaded extension module utils, skipping build step...No modifications detected for re-loaded extension module utils, skipping build step...
52: 
52: Loading extension module utils...
52: Loading extension module utils...
21: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
21: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
21: 
21: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
21: 
21: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
21: 
21: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
21: No modifications detected for re-loaded extension module utils, skipping build step...
21: Loading extension module utils...
21: No modifications detected for re-loaded extension module utils, skipping build step...No modifications detected for re-loaded extension module utils, skipping build step...
21: 
21: Loading extension module utils...No modifications detected for re-loaded extension module utils, skipping build step...Loading extension module utils...
21: 
21: 
21: No modifications detected for re-loaded extension module utils, skipping build step...Loading extension module utils...
21: 
21: Loading extension module utils...
21: No modifications detected for re-loaded extension module utils, skipping build step...
21: Loading extension module utils...
21: No modifications detected for re-loaded extension module utils, skipping build step...
21: Loading extension module utils...
21: No modifications detected for re-loaded extension module utils, skipping build step...
21: Loading extension module utils...
44: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
44: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
44: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
44: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
44: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
44: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
44: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
44: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
44: No modifications detected for re-loaded extension module utils, skipping build step...
44: Loading extension module utils...
44: No modifications detected for re-loaded extension module utils, skipping build step...
44: Loading extension module utils...
44: No modifications detected for re-loaded extension module utils, skipping build step...No modifications detected for re-loaded extension module utils, skipping build step...
44: 
44: No modifications detected for re-loaded extension module utils, skipping build step...Loading extension module utils...Loading extension module utils...
44: 
44: 
44: No modifications detected for re-loaded extension module utils, skipping build step...Loading extension module utils...
44: 
44: Loading extension module utils...
44: No modifications detected for re-loaded extension module utils, skipping build step...
44: Loading extension module utils...
44: No modifications detected for re-loaded extension module utils, skipping build step...
44: Loading extension module utils...
28: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
28: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
28: 
28: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
28: 
28: 
28: 
28: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
28: No modifications detected for re-loaded extension module utils, skipping build step...
28: Loading extension module utils...
28: No modifications detected for re-loaded extension module utils, skipping build step...
28: Loading extension module utils...
28: No modifications detected for re-loaded extension module utils, skipping build step...No modifications detected for re-loaded extension module utils, skipping build step...
28: 
28: Loading extension module utils...Loading extension module utils...
28: 
28: No modifications detected for re-loaded extension module utils, skipping build step...
28: Loading extension module utils...
28: No modifications detected for re-loaded extension module utils, skipping build step...
28: No modifications detected for re-loaded extension module utils, skipping build step...Loading extension module utils...
28: 
28: Loading extension module utils...
28: No modifications detected for re-loaded extension module utils, skipping build step...
28: Loading extension module utils...
40: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
40: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
40: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
40: 
40: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
40: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
40: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
40: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
18: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
18: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
16: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
16: 
18: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
18: 
16: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
16: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
18: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
16: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
16: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
16: 
16: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
18: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
18: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
18: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
10: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
16: No modifications detected for re-loaded extension module utils, skipping build step...
16: Loading extension module utils...
10: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
10: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
16: No modifications detected for re-loaded extension module utils, skipping build step...
16: Loading extension module utils...
16: No modifications detected for re-loaded extension module utils, skipping build step...
10: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
16: Loading extension module utils...
10: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
10: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
10: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
16: No modifications detected for re-loaded extension module utils, skipping build step...No modifications detected for re-loaded extension module utils, skipping build step...
16: 
16: Loading extension module utils...Loading extension module utils...
16: 
16: No modifications detected for re-loaded extension module utils, skipping build step...No modifications detected for re-loaded extension module utils, skipping build step...
16: 
16: Loading extension module utils...
16: Loading extension module utils...
10: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
16: No modifications detected for re-loaded extension module utils, skipping build step...
16: Loading extension module utils...
62: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
62: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
18: No modifications detected for re-loaded extension module utils, skipping build step...
18: Loading extension module utils...
62: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
62: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
46: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
46: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
62: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
62: 
46: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
62: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
62: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
40: No modifications detected for re-loaded extension module utils, skipping build step...
40: No modifications detected for re-loaded extension module utils, skipping build step...Loading extension module utils...
40: 
40: Loading extension module utils...
18: No modifications detected for re-loaded extension module utils, skipping build step...
18: Loading extension module utils...
46: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
46: 
46: 
40: No modifications detected for re-loaded extension module utils, skipping build step...
40: Loading extension module utils...
40: No modifications detected for re-loaded extension module utils, skipping build step...
10: No modifications detected for re-loaded extension module utils, skipping build step...
10: Loading extension module utils...
40: Loading extension module utils...No modifications detected for re-loaded extension module utils, skipping build step...
40: 
40: Loading extension module utils...
40: No modifications detected for re-loaded extension module utils, skipping build step...
62: No modifications detected for re-loaded extension module utils, skipping build step...
62: Loading extension module utils...
46: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
46: 
40: Loading extension module utils...
40: No modifications detected for re-loaded extension module utils, skipping build step...
40: Loading extension module utils...
62: No modifications detected for re-loaded extension module utils, skipping build step...No modifications detected for re-loaded extension module utils, skipping build step...
62: 
62: Loading extension module utils...Loading extension module utils...
62: 
10: No modifications detected for re-loaded extension module utils, skipping build step...No modifications detected for re-loaded extension module utils, skipping build step...
10: 
10: Loading extension module utils...Loading extension module utils...
10: 
40: No modifications detected for re-loaded extension module utils, skipping build step...
40: Loading extension module utils...
62: No modifications detected for re-loaded extension module utils, skipping build step...No modifications detected for re-loaded extension module utils, skipping build step...
62: 
62: Loading extension module utils...
62: Loading extension module utils...
18: No modifications detected for re-loaded extension module utils, skipping build step...
18: Loading extension module utils...No modifications detected for re-loaded extension module utils, skipping build step...
18: 
62: No modifications detected for re-loaded extension module utils, skipping build step...No modifications detected for re-loaded extension module utils, skipping build step...
62: 
62: Loading extension module utils...Loading extension module utils...No modifications detected for re-loaded extension module utils, skipping build step...
62: 
62: 
62: Loading extension module utils...
18: Loading extension module utils...
10: No modifications detected for re-loaded extension module utils, skipping build step...
10: Loading extension module utils...
18: No modifications detected for re-loaded extension module utils, skipping build step...
18: Loading extension module utils...
18: No modifications detected for re-loaded extension module utils, skipping build step...
18: Loading extension module utils...No modifications detected for re-loaded extension module utils, skipping build step...
18: 
10: No modifications detected for re-loaded extension module utils, skipping build step...
10: No modifications detected for re-loaded extension module utils, skipping build step...Loading extension module utils...
10: 
10: Loading extension module utils...
18: Loading extension module utils...
10: No modifications detected for re-loaded extension module utils, skipping build step...
10: Loading extension module utils...
18: No modifications detected for re-loaded extension module utils, skipping build step...
18: Loading extension module utils...
10: No modifications detected for re-loaded extension module utils, skipping build step...
10: Loading extension module utils...
46: No modifications detected for re-loaded extension module utils, skipping build step...
46: Loading extension module utils...
46: No modifications detected for re-loaded extension module utils, skipping build step...
46: Loading extension module utils...
46: No modifications detected for re-loaded extension module utils, skipping build step...
46: No modifications detected for re-loaded extension module utils, skipping build step...Loading extension module utils...
46: 
46: Loading extension module utils...
46: No modifications detected for re-loaded extension module utils, skipping build step...
46: No modifications detected for re-loaded extension module utils, skipping build step...Loading extension module utils...
46: 
46: Loading extension module utils...
46: No modifications detected for re-loaded extension module utils, skipping build step...
46: Loading extension module utils...
46: No modifications detected for re-loaded extension module utils, skipping build step...
46: Loading extension module utils...
 0: Using /pfs/lustrep4/users/muennighoff/.cache/torch_extensions/py39_cpu as PyTorch extensions root...
 0: No modifications detected for re-loaded extension module utils, skipping build step...
 0: Loading extension module utils...
 0: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/Megatron-DeepSpeed/megatron/utils.py:349: UserWarning: Parameter count with the embeddings will be inaccurate with PP > 1, as the first and last stage hold several copies of the embeddings
 0:   warnings.warn("Parameter count with the embeddings will be inaccurate with PP > 1, as the first and last stage hold several copies of the embeddings")
38: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
38:   if inputs.grad is not None:
38: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
38:   if inputs.grad is not None:
38: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
38:   if inputs.grad is not None:
38: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
38:   if inputs.grad is not None:
38: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
38:   if inputs.grad is not None:
42: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
42:   if inputs.grad is not None:
42: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
42:   if inputs.grad is not None:
42: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
42:   if inputs.grad is not None:
42: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
42:   if inputs.grad is not None:
42: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
42:   if inputs.grad is not None:
42: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
42:   if inputs.grad is not None:
38: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
38:   if inputs.grad is not None:
43: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
43:   if inputs.grad is not None:
43: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
43:   if inputs.grad is not None:
43: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
43:   if inputs.grad is not None:
43: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
43:   if inputs.grad is not None:
43: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
43:   if inputs.grad is not None:
43: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
43:   if inputs.grad is not None:
53: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
53:   if inputs.grad is not None:
53: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
53:   if inputs.grad is not None:
53: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
53:   if inputs.grad is not None:
53: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
53:   if inputs.grad is not None:
53: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
53:   if inputs.grad is not None:
53: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
53:   if inputs.grad is not None:
55: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
55:   if inputs.grad is not None:
55: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
55:   if inputs.grad is not None:
55: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
55:   if inputs.grad is not None:
55: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
55:   if inputs.grad is not None:
55: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
55:   if inputs.grad is not None:
61: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
61:   if inputs.grad is not None:
61: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
61:   if inputs.grad is not None:
61: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
61:   if inputs.grad is not None:
58: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
58:   if inputs.grad is not None:
61: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
61:   if inputs.grad is not None:
61: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
61:   if inputs.grad is not None:
58: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
58:   if inputs.grad is not None:
61: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
61:   if inputs.grad is not None:
58: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
58:   if inputs.grad is not None:
55: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
55:   if inputs.grad is not None:
58: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
58:   if inputs.grad is not None:
58: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
58:   if inputs.grad is not None:
58: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
58:   if inputs.grad is not None:
44: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
44:   if inputs.grad is not None:
44: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
44:   if inputs.grad is not None:
44: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
44:   if inputs.grad is not None:
44: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
44:   if inputs.grad is not None:
44: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
44:   if inputs.grad is not None:
44: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
44:   if inputs.grad is not None:
51: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
51:   if inputs.grad is not None:
51: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
51:   if inputs.grad is not None:
51: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
51:   if inputs.grad is not None:
51: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
51:   if inputs.grad is not None:
51: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
51:   if inputs.grad is not None:
51: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
51:   if inputs.grad is not None:
47: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
47:   if inputs.grad is not None:
47: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
47:   if inputs.grad is not None:
47: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
47:   if inputs.grad is not None:
47: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
47:   if inputs.grad is not None:
47: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
47:   if inputs.grad is not None:
47: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
47:   if inputs.grad is not None:
58: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
58:   if inputs.grad is not None:
58: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
58:   if inputs.grad is not None:
37: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
37:   if inputs.grad is not None:
37: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
37:   if inputs.grad is not None:
37: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
37:   if inputs.grad is not None:
37: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
37:   if inputs.grad is not None:
35: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
35:   if inputs.grad is not None:
35: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
35:   if inputs.grad is not None:
35: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
35:   if inputs.grad is not None:
35: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
35:   if inputs.grad is not None:
35: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
35:   if inputs.grad is not None:
55: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
55:   if inputs.grad is not None:
35: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
35:   if inputs.grad is not None:
55: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
55:   if inputs.grad is not None:
35: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
35:   if inputs.grad is not None:
46: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
46:   if inputs.grad is not None:
46: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
46:   if inputs.grad is not None:
46: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
46:   if inputs.grad is not None:
46: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
46:   if inputs.grad is not None:
46: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
46:   if inputs.grad is not None:
37: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
37:   if inputs.grad is not None:
46: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
46:   if inputs.grad is not None:
37: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
37:   if inputs.grad is not None:
35: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
35:   if inputs.grad is not None:
49: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
49:   if inputs.grad is not None:
49: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
49:   if inputs.grad is not None:
49: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
49:   if inputs.grad is not None:
49: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
49:   if inputs.grad is not None:
49: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
49:   if inputs.grad is not None:
49: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
49:   if inputs.grad is not None:
44: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
44:   if inputs.grad is not None:
57: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
57:   if inputs.grad is not None:
57: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
57:   if inputs.grad is not None:
52: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
52:   if inputs.grad is not None:
57: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
57:   if inputs.grad is not None:
52: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
52:   if inputs.grad is not None:
57: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
57:   if inputs.grad is not None:
52: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
52:   if inputs.grad is not None:
57: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
57:   if inputs.grad is not None:
52: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
52:   if inputs.grad is not None:
57: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
57:   if inputs.grad is not None:
52: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
52:   if inputs.grad is not None:
57: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
57:   if inputs.grad is not None:
52: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
52:   if inputs.grad is not None:
57: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
57:   if inputs.grad is not None:
32: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
32:   if inputs.grad is not None:
32: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
32:   if inputs.grad is not None:
32: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
32:   if inputs.grad is not None:
32: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
32:   if inputs.grad is not None:
32: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
32:   if inputs.grad is not None:
32: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
32:   if inputs.grad is not None:
32: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
32:   if inputs.grad is not None:
32: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
32:   if inputs.grad is not None:
42: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
42:   if inputs.grad is not None:
43: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
43:   if inputs.grad is not None:
43: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
43:   if inputs.grad is not None:
51: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
51:   if inputs.grad is not None:
51: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
51:   if inputs.grad is not None:
59: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
59:   if inputs.grad is not None:
59: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
59:   if inputs.grad is not None:
59: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
59:   if inputs.grad is not None:
59: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
59:   if inputs.grad is not None:
59: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
59:   if inputs.grad is not None:
59: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
59:   if inputs.grad is not None:
34: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
34:   if inputs.grad is not None:
34: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
34:   if inputs.grad is not None:
50: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
50:   if inputs.grad is not None:
34: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
34:   if inputs.grad is not None:
34: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
34:   if inputs.grad is not None:
50: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
50:   if inputs.grad is not None:
34: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
34:   if inputs.grad is not None:
50: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
50:   if inputs.grad is not None:
50: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
50:   if inputs.grad is not None:
50: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
50:   if inputs.grad is not None:
50: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
50:   if inputs.grad is not None:
50: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
50:   if inputs.grad is not None:
34: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
34:   if inputs.grad is not None:
34: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
34:   if inputs.grad is not None:
36: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
36:   if inputs.grad is not None:
36: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
36:   if inputs.grad is not None:
36: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
36:   if inputs.grad is not None:
36: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
36:   if inputs.grad is not None:
36: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
36:   if inputs.grad is not None:
36: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
36:   if inputs.grad is not None:
36: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
36:   if inputs.grad is not None:
36: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
36:   if inputs.grad is not None:
49: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
49:   if inputs.grad is not None:
49: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
49:   if inputs.grad is not None:
52: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
52:   if inputs.grad is not None:
52: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
52:   if inputs.grad is not None:
63: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
63:   if inputs.grad is not None:
45: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
45:   if inputs.grad is not None:
63: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
63:   if inputs.grad is not None:
63: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
63:   if inputs.grad is not None:
45: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
45:   if inputs.grad is not None:
63: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
63:   if inputs.grad is not None:
45: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
45:   if inputs.grad is not None:
63: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
63:   if inputs.grad is not None:
45: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
45:   if inputs.grad is not None:
62: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
62:   if inputs.grad is not None:
63: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
63:   if inputs.grad is not None:
45: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
45:   if inputs.grad is not None:
62: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
62:   if inputs.grad is not None:
63: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
63:   if inputs.grad is not None:
45: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
45:   if inputs.grad is not None:
62: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
62:   if inputs.grad is not None:
63: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
63:   if inputs.grad is not None:
62: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
62:   if inputs.grad is not None:
62: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
62:   if inputs.grad is not None:
62: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
62:   if inputs.grad is not None:
62: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
62:   if inputs.grad is not None:
41: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
41:   if inputs.grad is not None:
41: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
41:   if inputs.grad is not None:
41: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
41:   if inputs.grad is not None:
41: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
41:   if inputs.grad is not None:
41: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
41:   if inputs.grad is not None:
45: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
45:   if inputs.grad is not None:
39: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
39:   if inputs.grad is not None:
39: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
39:   if inputs.grad is not None:
45: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
45:   if inputs.grad is not None:
39: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
39:   if inputs.grad is not None:
39: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
39:   if inputs.grad is not None:
39: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
39:   if inputs.grad is not None:
39: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
39:   if inputs.grad is not None:
39: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
39:   if inputs.grad is not None:
41: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
41:   if inputs.grad is not None:
38: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
38:   if inputs.grad is not None:
38: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
38:   if inputs.grad is not None:
33: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
33:   if inputs.grad is not None:
33: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
33:   if inputs.grad is not None:
33: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
33:   if inputs.grad is not None:
33: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
33:   if inputs.grad is not None:
33: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
33:   if inputs.grad is not None:
33: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
33:   if inputs.grad is not None:
33: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
33:   if inputs.grad is not None:
54: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
54:   if inputs.grad is not None:
54: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
54:   if inputs.grad is not None:
54: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
54:   if inputs.grad is not None:
54: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
54:   if inputs.grad is not None:
54: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
54:   if inputs.grad is not None:
54: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
54:   if inputs.grad is not None:
54: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
54:   if inputs.grad is not None:
54: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
54:   if inputs.grad is not None:
34: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
34:   if inputs.grad is not None:
48: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
48:   if inputs.grad is not None:
48: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
48:   if inputs.grad is not None:
48: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
48:   if inputs.grad is not None:
48: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
48:   if inputs.grad is not None:
48: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
48:   if inputs.grad is not None:
48: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
48:   if inputs.grad is not None:
48: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
48:   if inputs.grad is not None:
60: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
60:   if inputs.grad is not None:
60: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
60:   if inputs.grad is not None:
60: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
60:   if inputs.grad is not None:
60: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
60:   if inputs.grad is not None:
60: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
60:   if inputs.grad is not None:
60: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
60:   if inputs.grad is not None:
48: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
48:   if inputs.grad is not None:
33: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
33:   if inputs.grad is not None:
40: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
40:   if inputs.grad is not None:
40: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
40:   if inputs.grad is not None:
40: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
40:   if inputs.grad is not None:
40: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
40:   if inputs.grad is not None:
40: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
40:   if inputs.grad is not None:
40: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
40:   if inputs.grad is not None:
40: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
40:   if inputs.grad is not None:
40: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
40:   if inputs.grad is not None:
59: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
59:   if inputs.grad is not None:
50: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
50:   if inputs.grad is not None:
60: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
60:   if inputs.grad is not None:
60: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
60:   if inputs.grad is not None:
37: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
37:   if inputs.grad is not None:
37: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
37:   if inputs.grad is not None:
61: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
61:   if inputs.grad is not None:
42: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
42:   if inputs.grad is not None:
59: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
59:   if inputs.grad is not None:
39: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
39:   if inputs.grad is not None:
44: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
44:   if inputs.grad is not None:
47: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
47:   if inputs.grad is not None:
61: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
61:   if inputs.grad is not None:
47: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
47:   if inputs.grad is not None:
53: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
53:   if inputs.grad is not None:
53: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
53:   if inputs.grad is not None:
62: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
62:   if inputs.grad is not None:
56: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
56:   if inputs.grad is not None:
56: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
56:   if inputs.grad is not None:
56: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
56:   if inputs.grad is not None:
56: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
56:   if inputs.grad is not None:
56: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
56:   if inputs.grad is not None:
56: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
56:   if inputs.grad is not None:
46: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
46:   if inputs.grad is not None:
46: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
46:   if inputs.grad is not None:
56: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
56:   if inputs.grad is not None:
56: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
56:   if inputs.grad is not None:
41: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
41:   if inputs.grad is not None:
41: /pfs/lustrep4/scratch/project_462000119/muennighoff/nov-2022-bettercom/venv/lib/python3.9/site-packages/deepspeed/runtime/pipe/engine.py:1200: UserWarning: The .grad attribute of a Tensor that is not a leaf Tensor is being accessed. Its .grad attribute won't be populated during autograd.backward(). If you indeed want the .grad field to be populated for a non-leaf Tensor, use .retain_grad() on the non-leaf Tensor. If you access the non-leaf Tensor by mistake, make sure you access the leaf Tensor instead. See github.com/pytorch/pytorch/pull/30531 for more informations. (Triggered internally at aten/src/ATen/core/TensorBody.h:485.)
41:   if inputs.grad is not None: