Upload folder using huggingface_hub
Browse files- checkpoint-1024/config.json +32 -0
- checkpoint-1024/generation_config.json +6 -0
- checkpoint-1024/model.safetensors +3 -0
- checkpoint-1024/optimizer.pt +3 -0
- checkpoint-1024/rng_state.pth +3 -0
- checkpoint-1024/scheduler.pt +3 -0
- checkpoint-1024/trainer_state.json +827 -0
- checkpoint-1024/training_args.bin +3 -0
- config.json +32 -0
- generation_config.json +6 -0
- model.safetensors +3 -0
- training_args.bin +3 -0
checkpoint-1024/config.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "/vol/tmp/koppelmm/pythia160m_dedup_untrained",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"GPTNeoXForCausalLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_bias": true,
|
| 7 |
+
"attention_dropout": 0.0,
|
| 8 |
+
"bos_token_id": 0,
|
| 9 |
+
"classifier_dropout": 0.1,
|
| 10 |
+
"eos_token_id": 0,
|
| 11 |
+
"hidden_act": "gelu",
|
| 12 |
+
"hidden_dropout": 0.0,
|
| 13 |
+
"hidden_size": 768,
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 3072,
|
| 16 |
+
"layer_norm_eps": 1e-05,
|
| 17 |
+
"max_position_embeddings": 2048,
|
| 18 |
+
"model_type": "gpt_neox",
|
| 19 |
+
"num_attention_heads": 12,
|
| 20 |
+
"num_hidden_layers": 12,
|
| 21 |
+
"partial_rotary_factor": 0.25,
|
| 22 |
+
"rope_scaling": null,
|
| 23 |
+
"rope_theta": 10000,
|
| 24 |
+
"rotary_emb_base": 10000,
|
| 25 |
+
"rotary_pct": 0.25,
|
| 26 |
+
"tie_word_embeddings": false,
|
| 27 |
+
"torch_dtype": "float32",
|
| 28 |
+
"transformers_version": "4.46.2",
|
| 29 |
+
"use_cache": true,
|
| 30 |
+
"use_parallel_residual": true,
|
| 31 |
+
"vocab_size": 50304
|
| 32 |
+
}
|
checkpoint-1024/generation_config.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 0,
|
| 4 |
+
"eos_token_id": 0,
|
| 5 |
+
"transformers_version": "4.46.2"
|
| 6 |
+
}
|
checkpoint-1024/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9c5f335ef94c0ce86cfda59e84cdd9bbef1b433fa1974f80e4b56d4498d8c3c6
|
| 3 |
+
size 649308728
|
checkpoint-1024/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:82ea90bec2bbb7210ee07df2db4735d39bd992d4cb6e7f1989788717eedf65f1
|
| 3 |
+
size 1298707258
|
checkpoint-1024/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:726ae76b86900d8abab60df7866887a207be06427ba6369a4451cd2686c4ed71
|
| 3 |
+
size 14244
|
checkpoint-1024/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:8ebd1a7d23da6a4b09d2235e23699a761e868e0b6ad2f2892a33dea66611ced6
|
| 3 |
+
size 1948045434
|
checkpoint-1024/trainer_state.json
ADDED
|
@@ -0,0 +1,827 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": null,
|
| 3 |
+
"best_model_checkpoint": null,
|
| 4 |
+
"epoch": 1.119799451652303,
|
| 5 |
+
"eval_steps": 100,
|
| 6 |
+
"global_step": 1024,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.010932788971549125,
|
| 13 |
+
"grad_norm": 167.07713317871094,
|
| 14 |
+
"learning_rate": 0.0005999985601583006,
|
| 15 |
+
"loss": 9.6698,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.02186557794309825,
|
| 20 |
+
"grad_norm": 83.92709350585938,
|
| 21 |
+
"learning_rate": 0.0005998257958771109,
|
| 22 |
+
"loss": 8.2484,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.032798366914647374,
|
| 27 |
+
"grad_norm": 127.91200256347656,
|
| 28 |
+
"learning_rate": 0.0005993652532642609,
|
| 29 |
+
"loss": 7.6452,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.0437311558861965,
|
| 34 |
+
"grad_norm": 97.33670043945312,
|
| 35 |
+
"learning_rate": 0.0005986173743570491,
|
| 36 |
+
"loss": 7.4548,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.054663944857745624,
|
| 41 |
+
"grad_norm": 127.0005874633789,
|
| 42 |
+
"learning_rate": 0.0005975828769834513,
|
| 43 |
+
"loss": 7.3226,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.06559673382929475,
|
| 48 |
+
"grad_norm": 104.47633361816406,
|
| 49 |
+
"learning_rate": 0.0005962627540731365,
|
| 50 |
+
"loss": 7.204,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.07652952280084388,
|
| 55 |
+
"grad_norm": 164.4477081298828,
|
| 56 |
+
"learning_rate": 0.0005946582727044349,
|
| 57 |
+
"loss": 7.1105,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.087462311772393,
|
| 62 |
+
"grad_norm": 126.8350601196289,
|
| 63 |
+
"learning_rate": 0.0005927709728881719,
|
| 64 |
+
"loss": 7.0511,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.09839510074394213,
|
| 69 |
+
"grad_norm": 158.55856323242188,
|
| 70 |
+
"learning_rate": 0.0005906026660895383,
|
| 71 |
+
"loss": 7.0642,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.10932788971549125,
|
| 76 |
+
"grad_norm": 126.1555404663086,
|
| 77 |
+
"learning_rate": 0.0005881554334894116,
|
| 78 |
+
"loss": 7.031,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.10932788971549125,
|
| 83 |
+
"eval_loss": 7.01555061340332,
|
| 84 |
+
"eval_runtime": 79.0984,
|
| 85 |
+
"eval_samples_per_second": 118.409,
|
| 86 |
+
"eval_steps_per_second": 14.804,
|
| 87 |
+
"step": 100
|
| 88 |
+
},
|
| 89 |
+
{
|
| 90 |
+
"epoch": 0.12026067868704038,
|
| 91 |
+
"grad_norm": 108.58393096923828,
|
| 92 |
+
"learning_rate": 0.0005854316239868012,
|
| 93 |
+
"loss": 7.0123,
|
| 94 |
+
"step": 110
|
| 95 |
+
},
|
| 96 |
+
{
|
| 97 |
+
"epoch": 0.1311934676585895,
|
| 98 |
+
"grad_norm": 178.0326690673828,
|
| 99 |
+
"learning_rate": 0.0005824338519443309,
|
| 100 |
+
"loss": 6.9897,
|
| 101 |
+
"step": 120
|
| 102 |
+
},
|
| 103 |
+
{
|
| 104 |
+
"epoch": 0.14212625663013861,
|
| 105 |
+
"grad_norm": 192.8655242919922,
|
| 106 |
+
"learning_rate": 0.0005791649946789259,
|
| 107 |
+
"loss": 7.0117,
|
| 108 |
+
"step": 130
|
| 109 |
+
},
|
| 110 |
+
{
|
| 111 |
+
"epoch": 0.15305904560168776,
|
| 112 |
+
"grad_norm": 143.3759002685547,
|
| 113 |
+
"learning_rate": 0.0005756281897001107,
|
| 114 |
+
"loss": 7.0073,
|
| 115 |
+
"step": 140
|
| 116 |
+
},
|
| 117 |
+
{
|
| 118 |
+
"epoch": 0.16399183457323688,
|
| 119 |
+
"grad_norm": 171.0679168701172,
|
| 120 |
+
"learning_rate": 0.0005718268316985698,
|
| 121 |
+
"loss": 6.9843,
|
| 122 |
+
"step": 150
|
| 123 |
+
},
|
| 124 |
+
{
|
| 125 |
+
"epoch": 0.174924623544786,
|
| 126 |
+
"grad_norm": 164.86534118652344,
|
| 127 |
+
"learning_rate": 0.0005677645692878606,
|
| 128 |
+
"loss": 7.0083,
|
| 129 |
+
"step": 160
|
| 130 |
+
},
|
| 131 |
+
{
|
| 132 |
+
"epoch": 0.1858574125163351,
|
| 133 |
+
"grad_norm": 125.85225677490234,
|
| 134 |
+
"learning_rate": 0.000563445301502407,
|
| 135 |
+
"loss": 7.02,
|
| 136 |
+
"step": 170
|
| 137 |
+
},
|
| 138 |
+
{
|
| 139 |
+
"epoch": 0.19679020148788426,
|
| 140 |
+
"grad_norm": 144.15589904785156,
|
| 141 |
+
"learning_rate": 0.0005588731740551344,
|
| 142 |
+
"loss": 6.9773,
|
| 143 |
+
"step": 180
|
| 144 |
+
},
|
| 145 |
+
{
|
| 146 |
+
"epoch": 0.20772299045943338,
|
| 147 |
+
"grad_norm": 108.05564880371094,
|
| 148 |
+
"learning_rate": 0.0005540525753583378,
|
| 149 |
+
"loss": 6.9632,
|
| 150 |
+
"step": 190
|
| 151 |
+
},
|
| 152 |
+
{
|
| 153 |
+
"epoch": 0.2186557794309825,
|
| 154 |
+
"grad_norm": 146.53924560546875,
|
| 155 |
+
"learning_rate": 0.0005489881323116018,
|
| 156 |
+
"loss": 6.929,
|
| 157 |
+
"step": 200
|
| 158 |
+
},
|
| 159 |
+
{
|
| 160 |
+
"epoch": 0.2186557794309825,
|
| 161 |
+
"eval_loss": 6.925621509552002,
|
| 162 |
+
"eval_runtime": 78.9467,
|
| 163 |
+
"eval_samples_per_second": 118.637,
|
| 164 |
+
"eval_steps_per_second": 14.833,
|
| 165 |
+
"step": 200
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 0.2295885684025316,
|
| 169 |
+
"grad_norm": 204.57968139648438,
|
| 170 |
+
"learning_rate": 0.0005436847058608189,
|
| 171 |
+
"loss": 6.9631,
|
| 172 |
+
"step": 210
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 0.24052135737408076,
|
| 176 |
+
"grad_norm": 171.31556701660156,
|
| 177 |
+
"learning_rate": 0.0005381473863325621,
|
| 178 |
+
"loss": 7.0389,
|
| 179 |
+
"step": 220
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 0.25145414634562985,
|
| 183 |
+
"grad_norm": 142.57449340820312,
|
| 184 |
+
"learning_rate": 0.0005323814885482963,
|
| 185 |
+
"loss": 6.967,
|
| 186 |
+
"step": 230
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 0.262386935317179,
|
| 190 |
+
"grad_norm": 119.19646453857422,
|
| 191 |
+
"learning_rate": 0.000526392546723115,
|
| 192 |
+
"loss": 6.9456,
|
| 193 |
+
"step": 240
|
| 194 |
+
},
|
| 195 |
+
{
|
| 196 |
+
"epoch": 0.27331972428872814,
|
| 197 |
+
"grad_norm": 153.62359619140625,
|
| 198 |
+
"learning_rate": 0.0005201863091538979,
|
| 199 |
+
"loss": 6.9686,
|
| 200 |
+
"step": 250
|
| 201 |
+
},
|
| 202 |
+
{
|
| 203 |
+
"epoch": 0.28425251326027723,
|
| 204 |
+
"grad_norm": 150.35699462890625,
|
| 205 |
+
"learning_rate": 0.000513768732701989,
|
| 206 |
+
"loss": 6.9846,
|
| 207 |
+
"step": 260
|
| 208 |
+
},
|
| 209 |
+
{
|
| 210 |
+
"epoch": 0.2951853022318264,
|
| 211 |
+
"grad_norm": 215.55368041992188,
|
| 212 |
+
"learning_rate": 0.0005071459770756929,
|
| 213 |
+
"loss": 6.9968,
|
| 214 |
+
"step": 270
|
| 215 |
+
},
|
| 216 |
+
{
|
| 217 |
+
"epoch": 0.3061180912033755,
|
| 218 |
+
"grad_norm": 107.55154418945312,
|
| 219 |
+
"learning_rate": 0.0005003243989180711,
|
| 220 |
+
"loss": 7.0033,
|
| 221 |
+
"step": 280
|
| 222 |
+
},
|
| 223 |
+
{
|
| 224 |
+
"epoch": 0.3170508801749246,
|
| 225 |
+
"grad_norm": 190.4154052734375,
|
| 226 |
+
"learning_rate": 0.0004933105457057203,
|
| 227 |
+
"loss": 6.9816,
|
| 228 |
+
"step": 290
|
| 229 |
+
},
|
| 230 |
+
{
|
| 231 |
+
"epoch": 0.32798366914647376,
|
| 232 |
+
"grad_norm": 159.7703094482422,
|
| 233 |
+
"learning_rate": 0.0004861111494643821,
|
| 234 |
+
"loss": 7.0486,
|
| 235 |
+
"step": 300
|
| 236 |
+
},
|
| 237 |
+
{
|
| 238 |
+
"epoch": 0.32798366914647376,
|
| 239 |
+
"eval_loss": 7.4869384765625,
|
| 240 |
+
"eval_runtime": 79.1717,
|
| 241 |
+
"eval_samples_per_second": 118.3,
|
| 242 |
+
"eval_steps_per_second": 14.791,
|
| 243 |
+
"step": 300
|
| 244 |
+
},
|
| 245 |
+
{
|
| 246 |
+
"epoch": 0.3389164581180229,
|
| 247 |
+
"grad_norm": 218.22604370117188,
|
| 248 |
+
"learning_rate": 0.0004794787611927562,
|
| 249 |
+
"loss": 7.2679,
|
| 250 |
+
"step": 310
|
| 251 |
+
},
|
| 252 |
+
{
|
| 253 |
+
"epoch": 0.349849247089572,
|
| 254 |
+
"grad_norm": 182.51431274414062,
|
| 255 |
+
"learning_rate": 0.0004719460124060748,
|
| 256 |
+
"loss": 7.1809,
|
| 257 |
+
"step": 320
|
| 258 |
+
},
|
| 259 |
+
{
|
| 260 |
+
"epoch": 0.36078203606112114,
|
| 261 |
+
"grad_norm": 137.0953826904297,
|
| 262 |
+
"learning_rate": 0.0004642482266637136,
|
| 263 |
+
"loss": 7.0417,
|
| 264 |
+
"step": 330
|
| 265 |
+
},
|
| 266 |
+
{
|
| 267 |
+
"epoch": 0.3717148250326702,
|
| 268 |
+
"grad_norm": 92.07840728759766,
|
| 269 |
+
"learning_rate": 0.0004563927924424775,
|
| 270 |
+
"loss": 6.9309,
|
| 271 |
+
"step": 340
|
| 272 |
+
},
|
| 273 |
+
{
|
| 274 |
+
"epoch": 0.38264761400421937,
|
| 275 |
+
"grad_norm": 147.35975646972656,
|
| 276 |
+
"learning_rate": 0.00044838724953309093,
|
| 277 |
+
"loss": 6.8844,
|
| 278 |
+
"step": 350
|
| 279 |
+
},
|
| 280 |
+
{
|
| 281 |
+
"epoch": 0.3935804029757685,
|
| 282 |
+
"grad_norm": 262.996337890625,
|
| 283 |
+
"learning_rate": 0.0004402392818033671,
|
| 284 |
+
"loss": 6.966,
|
| 285 |
+
"step": 360
|
| 286 |
+
},
|
| 287 |
+
{
|
| 288 |
+
"epoch": 0.4045131919473176,
|
| 289 |
+
"grad_norm": 155.3452606201172,
|
| 290 |
+
"learning_rate": 0.00043195670982308984,
|
| 291 |
+
"loss": 7.0715,
|
| 292 |
+
"step": 370
|
| 293 |
+
},
|
| 294 |
+
{
|
| 295 |
+
"epoch": 0.41544598091886675,
|
| 296 |
+
"grad_norm": 129.5069580078125,
|
| 297 |
+
"learning_rate": 0.00042354748335768664,
|
| 298 |
+
"loss": 7.0806,
|
| 299 |
+
"step": 380
|
| 300 |
+
},
|
| 301 |
+
{
|
| 302 |
+
"epoch": 0.4263787698904159,
|
| 303 |
+
"grad_norm": 92.96502685546875,
|
| 304 |
+
"learning_rate": 0.0004150196737378971,
|
| 305 |
+
"loss": 6.9999,
|
| 306 |
+
"step": 390
|
| 307 |
+
},
|
| 308 |
+
{
|
| 309 |
+
"epoch": 0.437311558861965,
|
| 310 |
+
"grad_norm": 120.41193389892578,
|
| 311 |
+
"learning_rate": 0.0004063814661127606,
|
| 312 |
+
"loss": 6.9339,
|
| 313 |
+
"step": 400
|
| 314 |
+
},
|
| 315 |
+
{
|
| 316 |
+
"epoch": 0.437311558861965,
|
| 317 |
+
"eval_loss": 6.931961536407471,
|
| 318 |
+
"eval_runtime": 78.8373,
|
| 319 |
+
"eval_samples_per_second": 118.802,
|
| 320 |
+
"eval_steps_per_second": 14.853,
|
| 321 |
+
"step": 400
|
| 322 |
+
},
|
| 323 |
+
{
|
| 324 |
+
"epoch": 0.44824434783351413,
|
| 325 |
+
"grad_norm": 188.7049560546875,
|
| 326 |
+
"learning_rate": 0.00039764115159335935,
|
| 327 |
+
"loss": 6.9242,
|
| 328 |
+
"step": 410
|
| 329 |
+
},
|
| 330 |
+
{
|
| 331 |
+
"epoch": 0.4591771368050632,
|
| 332 |
+
"grad_norm": 131.7518768310547,
|
| 333 |
+
"learning_rate": 0.0003888071192948565,
|
| 334 |
+
"loss": 6.9815,
|
| 335 |
+
"step": 420
|
| 336 |
+
},
|
| 337 |
+
{
|
| 338 |
+
"epoch": 0.47010992577661237,
|
| 339 |
+
"grad_norm": 247.91549682617188,
|
| 340 |
+
"learning_rate": 0.0003798878482844695,
|
| 341 |
+
"loss": 7.0838,
|
| 342 |
+
"step": 430
|
| 343 |
+
},
|
| 344 |
+
{
|
| 345 |
+
"epoch": 0.4810427147481615,
|
| 346 |
+
"grad_norm": 135.4517364501953,
|
| 347 |
+
"learning_rate": 0.000370891899443104,
|
| 348 |
+
"loss": 7.1813,
|
| 349 |
+
"step": 440
|
| 350 |
+
},
|
| 351 |
+
{
|
| 352 |
+
"epoch": 0.4919755037197106,
|
| 353 |
+
"grad_norm": 99.5172119140625,
|
| 354 |
+
"learning_rate": 0.00036182790724846315,
|
| 355 |
+
"loss": 7.1557,
|
| 356 |
+
"step": 450
|
| 357 |
+
},
|
| 358 |
+
{
|
| 359 |
+
"epoch": 0.5029082926912597,
|
| 360 |
+
"grad_norm": 165.1914825439453,
|
| 361 |
+
"learning_rate": 0.00035270457148751575,
|
| 362 |
+
"loss": 7.0382,
|
| 363 |
+
"step": 460
|
| 364 |
+
},
|
| 365 |
+
{
|
| 366 |
+
"epoch": 0.5138410816628088,
|
| 367 |
+
"grad_norm": 128.59959411621094,
|
| 368 |
+
"learning_rate": 0.00034353064890628107,
|
| 369 |
+
"loss": 7.0597,
|
| 370 |
+
"step": 470
|
| 371 |
+
},
|
| 372 |
+
{
|
| 373 |
+
"epoch": 0.524773870634358,
|
| 374 |
+
"grad_norm": 142.37147521972656,
|
| 375 |
+
"learning_rate": 0.00033431494480494175,
|
| 376 |
+
"loss": 7.092,
|
| 377 |
+
"step": 480
|
| 378 |
+
},
|
| 379 |
+
{
|
| 380 |
+
"epoch": 0.5357066596059071,
|
| 381 |
+
"grad_norm": 217.4059295654297,
|
| 382 |
+
"learning_rate": 0.0003250663045863544,
|
| 383 |
+
"loss": 7.0457,
|
| 384 |
+
"step": 490
|
| 385 |
+
},
|
| 386 |
+
{
|
| 387 |
+
"epoch": 0.5466394485774563,
|
| 388 |
+
"grad_norm": 125.81988525390625,
|
| 389 |
+
"learning_rate": 0.0003157936052660688,
|
| 390 |
+
"loss": 7.0112,
|
| 391 |
+
"step": 500
|
| 392 |
+
},
|
| 393 |
+
{
|
| 394 |
+
"epoch": 0.5466394485774563,
|
| 395 |
+
"eval_loss": 7.004736423492432,
|
| 396 |
+
"eval_runtime": 78.8041,
|
| 397 |
+
"eval_samples_per_second": 118.852,
|
| 398 |
+
"eval_steps_per_second": 14.86,
|
| 399 |
+
"step": 500
|
| 400 |
+
},
|
| 401 |
+
{
|
| 402 |
+
"epoch": 0.5575722375490054,
|
| 403 |
+
"grad_norm": 170.00523376464844,
|
| 404 |
+
"learning_rate": 0.0003065057469520046,
|
| 405 |
+
"loss": 7.0162,
|
| 406 |
+
"step": 510
|
| 407 |
+
},
|
| 408 |
+
{
|
| 409 |
+
"epoch": 0.5685050265205545,
|
| 410 |
+
"grad_norm": 216.81466674804688,
|
| 411 |
+
"learning_rate": 0.0002972116443019633,
|
| 412 |
+
"loss": 7.0584,
|
| 413 |
+
"step": 520
|
| 414 |
+
},
|
| 415 |
+
{
|
| 416 |
+
"epoch": 0.5794378154921036,
|
| 417 |
+
"grad_norm": 239.21087646484375,
|
| 418 |
+
"learning_rate": 0.0002879202179671755,
|
| 419 |
+
"loss": 7.1254,
|
| 420 |
+
"step": 530
|
| 421 |
+
},
|
| 422 |
+
{
|
| 423 |
+
"epoch": 0.5903706044636527,
|
| 424 |
+
"grad_norm": 190.0070343017578,
|
| 425 |
+
"learning_rate": 0.00027864038603009453,
|
| 426 |
+
"loss": 7.1717,
|
| 427 |
+
"step": 540
|
| 428 |
+
},
|
| 429 |
+
{
|
| 430 |
+
"epoch": 0.6013033934352019,
|
| 431 |
+
"grad_norm": 179.18785095214844,
|
| 432 |
+
"learning_rate": 0.00026938105544465745,
|
| 433 |
+
"loss": 7.1185,
|
| 434 |
+
"step": 550
|
| 435 |
+
},
|
| 436 |
+
{
|
| 437 |
+
"epoch": 0.612236182406751,
|
| 438 |
+
"grad_norm": 279.44781494140625,
|
| 439 |
+
"learning_rate": 0.0002601511134872255,
|
| 440 |
+
"loss": 7.0727,
|
| 441 |
+
"step": 560
|
| 442 |
+
},
|
| 443 |
+
{
|
| 444 |
+
"epoch": 0.6231689713783001,
|
| 445 |
+
"grad_norm": 227.90072631835938,
|
| 446 |
+
"learning_rate": 0.0002509594192264121,
|
| 447 |
+
"loss": 7.1088,
|
| 448 |
+
"step": 570
|
| 449 |
+
},
|
| 450 |
+
{
|
| 451 |
+
"epoch": 0.6341017603498492,
|
| 452 |
+
"grad_norm": 173.11819458007812,
|
| 453 |
+
"learning_rate": 0.0002418147950199862,
|
| 454 |
+
"loss": 7.0927,
|
| 455 |
+
"step": 580
|
| 456 |
+
},
|
| 457 |
+
{
|
| 458 |
+
"epoch": 0.6450345493213984,
|
| 459 |
+
"grad_norm": 164.40736389160156,
|
| 460 |
+
"learning_rate": 0.00023272601804700946,
|
| 461 |
+
"loss": 7.0701,
|
| 462 |
+
"step": 590
|
| 463 |
+
},
|
| 464 |
+
{
|
| 465 |
+
"epoch": 0.6559673382929475,
|
| 466 |
+
"grad_norm": 123.35533142089844,
|
| 467 |
+
"learning_rate": 0.0002237018118833387,
|
| 468 |
+
"loss": 7.0496,
|
| 469 |
+
"step": 600
|
| 470 |
+
},
|
| 471 |
+
{
|
| 472 |
+
"epoch": 0.6559673382929475,
|
| 473 |
+
"eval_loss": 7.052866458892822,
|
| 474 |
+
"eval_runtime": 78.8887,
|
| 475 |
+
"eval_samples_per_second": 118.724,
|
| 476 |
+
"eval_steps_per_second": 14.844,
|
| 477 |
+
"step": 600
|
| 478 |
+
},
|
| 479 |
+
{
|
| 480 |
+
"epoch": 0.6669001272644967,
|
| 481 |
+
"grad_norm": 225.67015075683594,
|
| 482 |
+
"learning_rate": 0.0002147508381285762,
|
| 483 |
+
"loss": 7.04,
|
| 484 |
+
"step": 610
|
| 485 |
+
},
|
| 486 |
+
{
|
| 487 |
+
"epoch": 0.6778329162360458,
|
| 488 |
+
"grad_norm": 140.2364501953125,
|
| 489 |
+
"learning_rate": 0.00020588168809250687,
|
| 490 |
+
"loss": 7.0902,
|
| 491 |
+
"step": 620
|
| 492 |
+
},
|
| 493 |
+
{
|
| 494 |
+
"epoch": 0.6887657052075948,
|
| 495 |
+
"grad_norm": 262.8550720214844,
|
| 496 |
+
"learning_rate": 0.00019710287454900033,
|
| 497 |
+
"loss": 7.1224,
|
| 498 |
+
"step": 630
|
| 499 |
+
},
|
| 500 |
+
{
|
| 501 |
+
"epoch": 0.699698494179144,
|
| 502 |
+
"grad_norm": 150.97813415527344,
|
| 503 |
+
"learning_rate": 0.00018842282356529402,
|
| 504 |
+
"loss": 7.1802,
|
| 505 |
+
"step": 640
|
| 506 |
+
},
|
| 507 |
+
{
|
| 508 |
+
"epoch": 0.7106312831506931,
|
| 509 |
+
"grad_norm": 452.73431396484375,
|
| 510 |
+
"learning_rate": 0.00017984986641449754,
|
| 511 |
+
"loss": 7.1497,
|
| 512 |
+
"step": 650
|
| 513 |
+
},
|
| 514 |
+
{
|
| 515 |
+
"epoch": 0.7215640721222423,
|
| 516 |
+
"grad_norm": 138.37220764160156,
|
| 517 |
+
"learning_rate": 0.00017139223157908368,
|
| 518 |
+
"loss": 7.1715,
|
| 519 |
+
"step": 660
|
| 520 |
+
},
|
| 521 |
+
{
|
| 522 |
+
"epoch": 0.7324968610937914,
|
| 523 |
+
"grad_norm": 144.21133422851562,
|
| 524 |
+
"learning_rate": 0.00016305803685303906,
|
| 525 |
+
"loss": 7.1458,
|
| 526 |
+
"step": 670
|
| 527 |
+
},
|
| 528 |
+
{
|
| 529 |
+
"epoch": 0.7434296500653405,
|
| 530 |
+
"grad_norm": 142.4859161376953,
|
| 531 |
+
"learning_rate": 0.00015485528155025473,
|
| 532 |
+
"loss": 7.1041,
|
| 533 |
+
"step": 680
|
| 534 |
+
},
|
| 535 |
+
{
|
| 536 |
+
"epoch": 0.7543624390368896,
|
| 537 |
+
"grad_norm": 190.189208984375,
|
| 538 |
+
"learning_rate": 0.00014679183882663872,
|
| 539 |
+
"loss": 7.0798,
|
| 540 |
+
"step": 690
|
| 541 |
+
},
|
| 542 |
+
{
|
| 543 |
+
"epoch": 0.7652952280084387,
|
| 544 |
+
"grad_norm": 160.14442443847656,
|
| 545 |
+
"learning_rate": 0.0001388754481233139,
|
| 546 |
+
"loss": 7.074,
|
| 547 |
+
"step": 700
|
| 548 |
+
},
|
| 549 |
+
{
|
| 550 |
+
"epoch": 0.7652952280084387,
|
| 551 |
+
"eval_loss": 7.0790934562683105,
|
| 552 |
+
"eval_runtime": 79.0053,
|
| 553 |
+
"eval_samples_per_second": 118.549,
|
| 554 |
+
"eval_steps_per_second": 14.822,
|
| 555 |
+
"step": 700
|
| 556 |
+
},
|
| 557 |
+
{
|
| 558 |
+
"epoch": 0.7762280169799879,
|
| 559 |
+
"grad_norm": 173.01499938964844,
|
| 560 |
+
"learning_rate": 0.0001311137077381614,
|
| 561 |
+
"loss": 7.0821,
|
| 562 |
+
"step": 710
|
| 563 |
+
},
|
| 564 |
+
{
|
| 565 |
+
"epoch": 0.787160805951537,
|
| 566 |
+
"grad_norm": 156.1138458251953,
|
| 567 |
+
"learning_rate": 0.00012351406753283216,
|
| 568 |
+
"loss": 7.0838,
|
| 569 |
+
"step": 720
|
| 570 |
+
},
|
| 571 |
+
{
|
| 572 |
+
"epoch": 0.7980935949230861,
|
| 573 |
+
"grad_norm": 161.9981689453125,
|
| 574 |
+
"learning_rate": 0.00011681901904809884,
|
| 575 |
+
"loss": 7.0639,
|
| 576 |
+
"step": 730
|
| 577 |
+
},
|
| 578 |
+
{
|
| 579 |
+
"epoch": 0.8090263838946352,
|
| 580 |
+
"grad_norm": 174.0237579345703,
|
| 581 |
+
"learning_rate": 0.00010954733067505213,
|
| 582 |
+
"loss": 7.0604,
|
| 583 |
+
"step": 740
|
| 584 |
+
},
|
| 585 |
+
{
|
| 586 |
+
"epoch": 0.8199591728661844,
|
| 587 |
+
"grad_norm": 141.823974609375,
|
| 588 |
+
"learning_rate": 0.0001024584422885053,
|
| 589 |
+
"loss": 7.0508,
|
| 590 |
+
"step": 750
|
| 591 |
+
},
|
| 592 |
+
{
|
| 593 |
+
"epoch": 0.8308919618377335,
|
| 594 |
+
"grad_norm": 121.39106750488281,
|
| 595 |
+
"learning_rate": 9.555915793434476e-05,
|
| 596 |
+
"loss": 7.0568,
|
| 597 |
+
"step": 760
|
| 598 |
+
},
|
| 599 |
+
{
|
| 600 |
+
"epoch": 0.8418247508092827,
|
| 601 |
+
"grad_norm": 178.37924194335938,
|
| 602 |
+
"learning_rate": 8.885609967300851e-05,
|
| 603 |
+
"loss": 7.0589,
|
| 604 |
+
"step": 770
|
| 605 |
+
},
|
| 606 |
+
{
|
| 607 |
+
"epoch": 0.8527575397808318,
|
| 608 |
+
"grad_norm": 304.8969421386719,
|
| 609 |
+
"learning_rate": 8.235570122350937e-05,
|
| 610 |
+
"loss": 7.0582,
|
| 611 |
+
"step": 780
|
| 612 |
+
},
|
| 613 |
+
{
|
| 614 |
+
"epoch": 0.8636903287523808,
|
| 615 |
+
"grad_norm": 128.75843811035156,
|
| 616 |
+
"learning_rate": 7.606420178823293e-05,
|
| 617 |
+
"loss": 7.0622,
|
| 618 |
+
"step": 790
|
| 619 |
+
},
|
| 620 |
+
{
|
| 621 |
+
"epoch": 0.87462311772393,
|
| 622 |
+
"grad_norm": 88.88775634765625,
|
| 623 |
+
"learning_rate": 6.998764006443615e-05,
|
| 624 |
+
"loss": 7.0664,
|
| 625 |
+
"step": 800
|
| 626 |
+
},
|
| 627 |
+
{
|
| 628 |
+
"epoch": 0.87462311772393,
|
| 629 |
+
"eval_loss": 7.048069477081299,
|
| 630 |
+
"eval_runtime": 78.7086,
|
| 631 |
+
"eval_samples_per_second": 118.996,
|
| 632 |
+
"eval_steps_per_second": 14.878,
|
| 633 |
+
"step": 800
|
| 634 |
+
},
|
| 635 |
+
{
|
| 636 |
+
"epoch": 0.8855559066954791,
|
| 637 |
+
"grad_norm": 131.33584594726562,
|
| 638 |
+
"learning_rate": 6.413184844819423e-05,
|
| 639 |
+
"loss": 7.0381,
|
| 640 |
+
"step": 810
|
| 641 |
+
},
|
| 642 |
+
{
|
| 643 |
+
"epoch": 0.8964886956670283,
|
| 644 |
+
"grad_norm": 176.8515625,
|
| 645 |
+
"learning_rate": 6e-05,
|
| 646 |
+
"loss": 7.0461,
|
| 647 |
+
"step": 820
|
| 648 |
+
},
|
| 649 |
+
{
|
| 650 |
+
"epoch": 0.9074214846385774,
|
| 651 |
+
"grad_norm": 128.32069396972656,
|
| 652 |
+
"learning_rate": 6e-05,
|
| 653 |
+
"loss": 7.0597,
|
| 654 |
+
"step": 830
|
| 655 |
+
},
|
| 656 |
+
{
|
| 657 |
+
"epoch": 0.9183542736101264,
|
| 658 |
+
"grad_norm": 150.107421875,
|
| 659 |
+
"learning_rate": 6e-05,
|
| 660 |
+
"loss": 7.0582,
|
| 661 |
+
"step": 840
|
| 662 |
+
},
|
| 663 |
+
{
|
| 664 |
+
"epoch": 0.9292870625816756,
|
| 665 |
+
"grad_norm": 174.95352172851562,
|
| 666 |
+
"learning_rate": 6e-05,
|
| 667 |
+
"loss": 7.0729,
|
| 668 |
+
"step": 850
|
| 669 |
+
},
|
| 670 |
+
{
|
| 671 |
+
"epoch": 0.9402198515532247,
|
| 672 |
+
"grad_norm": 209.878173828125,
|
| 673 |
+
"learning_rate": 6e-05,
|
| 674 |
+
"loss": 7.0949,
|
| 675 |
+
"step": 860
|
| 676 |
+
},
|
| 677 |
+
{
|
| 678 |
+
"epoch": 0.9511526405247739,
|
| 679 |
+
"grad_norm": 181.1326904296875,
|
| 680 |
+
"learning_rate": 6e-05,
|
| 681 |
+
"loss": 7.109,
|
| 682 |
+
"step": 870
|
| 683 |
+
},
|
| 684 |
+
{
|
| 685 |
+
"epoch": 0.962085429496323,
|
| 686 |
+
"grad_norm": 197.11639404296875,
|
| 687 |
+
"learning_rate": 6e-05,
|
| 688 |
+
"loss": 7.1132,
|
| 689 |
+
"step": 880
|
| 690 |
+
},
|
| 691 |
+
{
|
| 692 |
+
"epoch": 0.9730182184678722,
|
| 693 |
+
"grad_norm": 197.16473388671875,
|
| 694 |
+
"learning_rate": 6e-05,
|
| 695 |
+
"loss": 7.1008,
|
| 696 |
+
"step": 890
|
| 697 |
+
},
|
| 698 |
+
{
|
| 699 |
+
"epoch": 0.9839510074394212,
|
| 700 |
+
"grad_norm": 224.1211395263672,
|
| 701 |
+
"learning_rate": 6e-05,
|
| 702 |
+
"loss": 7.1024,
|
| 703 |
+
"step": 900
|
| 704 |
+
},
|
| 705 |
+
{
|
| 706 |
+
"epoch": 0.9839510074394212,
|
| 707 |
+
"eval_loss": 7.119234561920166,
|
| 708 |
+
"eval_runtime": 78.4745,
|
| 709 |
+
"eval_samples_per_second": 119.351,
|
| 710 |
+
"eval_steps_per_second": 14.922,
|
| 711 |
+
"step": 900
|
| 712 |
+
},
|
| 713 |
+
{
|
| 714 |
+
"epoch": 0.9948837964109704,
|
| 715 |
+
"grad_norm": 161.86753845214844,
|
| 716 |
+
"learning_rate": 6e-05,
|
| 717 |
+
"loss": 7.1127,
|
| 718 |
+
"step": 910
|
| 719 |
+
},
|
| 720 |
+
{
|
| 721 |
+
"epoch": 1.0060984463481923,
|
| 722 |
+
"grad_norm": 247.6467742919922,
|
| 723 |
+
"learning_rate": 6e-05,
|
| 724 |
+
"loss": 7.1115,
|
| 725 |
+
"step": 920
|
| 726 |
+
},
|
| 727 |
+
{
|
| 728 |
+
"epoch": 1.0170312353197413,
|
| 729 |
+
"grad_norm": 228.1467742919922,
|
| 730 |
+
"learning_rate": 6e-05,
|
| 731 |
+
"loss": 7.1172,
|
| 732 |
+
"step": 930
|
| 733 |
+
},
|
| 734 |
+
{
|
| 735 |
+
"epoch": 1.0279640242912904,
|
| 736 |
+
"grad_norm": 400.675537109375,
|
| 737 |
+
"learning_rate": 6e-05,
|
| 738 |
+
"loss": 7.1351,
|
| 739 |
+
"step": 940
|
| 740 |
+
},
|
| 741 |
+
{
|
| 742 |
+
"epoch": 1.0388968132628396,
|
| 743 |
+
"grad_norm": 293.3075866699219,
|
| 744 |
+
"learning_rate": 6e-05,
|
| 745 |
+
"loss": 7.1747,
|
| 746 |
+
"step": 950
|
| 747 |
+
},
|
| 748 |
+
{
|
| 749 |
+
"epoch": 1.0498296022343887,
|
| 750 |
+
"grad_norm": 439.60760498046875,
|
| 751 |
+
"learning_rate": 6e-05,
|
| 752 |
+
"loss": 7.1955,
|
| 753 |
+
"step": 960
|
| 754 |
+
},
|
| 755 |
+
{
|
| 756 |
+
"epoch": 1.0607623912059378,
|
| 757 |
+
"grad_norm": 336.15521240234375,
|
| 758 |
+
"learning_rate": 6e-05,
|
| 759 |
+
"loss": 7.2134,
|
| 760 |
+
"step": 970
|
| 761 |
+
},
|
| 762 |
+
{
|
| 763 |
+
"epoch": 1.071695180177487,
|
| 764 |
+
"grad_norm": 232.90606689453125,
|
| 765 |
+
"learning_rate": 6e-05,
|
| 766 |
+
"loss": 7.2589,
|
| 767 |
+
"step": 980
|
| 768 |
+
},
|
| 769 |
+
{
|
| 770 |
+
"epoch": 1.0826279691490361,
|
| 771 |
+
"grad_norm": 453.7010803222656,
|
| 772 |
+
"learning_rate": 6e-05,
|
| 773 |
+
"loss": 7.2537,
|
| 774 |
+
"step": 990
|
| 775 |
+
},
|
| 776 |
+
{
|
| 777 |
+
"epoch": 1.0935607581205853,
|
| 778 |
+
"grad_norm": 156.7413330078125,
|
| 779 |
+
"learning_rate": 6e-05,
|
| 780 |
+
"loss": 7.2678,
|
| 781 |
+
"step": 1000
|
| 782 |
+
},
|
| 783 |
+
{
|
| 784 |
+
"epoch": 1.0935607581205853,
|
| 785 |
+
"eval_loss": 7.271553993225098,
|
| 786 |
+
"eval_runtime": 78.5452,
|
| 787 |
+
"eval_samples_per_second": 119.243,
|
| 788 |
+
"eval_steps_per_second": 14.909,
|
| 789 |
+
"step": 1000
|
| 790 |
+
},
|
| 791 |
+
{
|
| 792 |
+
"epoch": 1.1044935470921344,
|
| 793 |
+
"grad_norm": 225.9600067138672,
|
| 794 |
+
"learning_rate": 6e-05,
|
| 795 |
+
"loss": 7.2489,
|
| 796 |
+
"step": 1010
|
| 797 |
+
},
|
| 798 |
+
{
|
| 799 |
+
"epoch": 1.1154263360636836,
|
| 800 |
+
"grad_norm": 258.6958312988281,
|
| 801 |
+
"learning_rate": 6e-05,
|
| 802 |
+
"loss": 7.2224,
|
| 803 |
+
"step": 1020
|
| 804 |
+
}
|
| 805 |
+
],
|
| 806 |
+
"logging_steps": 10,
|
| 807 |
+
"max_steps": 1024,
|
| 808 |
+
"num_input_tokens_seen": 0,
|
| 809 |
+
"num_train_epochs": 2,
|
| 810 |
+
"save_steps": 1024,
|
| 811 |
+
"stateful_callbacks": {
|
| 812 |
+
"TrainerControl": {
|
| 813 |
+
"args": {
|
| 814 |
+
"should_epoch_stop": false,
|
| 815 |
+
"should_evaluate": false,
|
| 816 |
+
"should_log": false,
|
| 817 |
+
"should_save": true,
|
| 818 |
+
"should_training_stop": true
|
| 819 |
+
},
|
| 820 |
+
"attributes": {}
|
| 821 |
+
}
|
| 822 |
+
},
|
| 823 |
+
"total_flos": 1.383804151351214e+18,
|
| 824 |
+
"train_batch_size": 8,
|
| 825 |
+
"trial_name": null,
|
| 826 |
+
"trial_params": null
|
| 827 |
+
}
|
checkpoint-1024/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:95492cc580971074153fb0dee7b60aedfda648c5f8a7cb99bd7fbc9464d590d2
|
| 3 |
+
size 5304
|
config.json
ADDED
|
@@ -0,0 +1,32 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "/vol/tmp/koppelmm/pythia160m_dedup_untrained",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"GPTNeoXForCausalLM"
|
| 5 |
+
],
|
| 6 |
+
"attention_bias": true,
|
| 7 |
+
"attention_dropout": 0.0,
|
| 8 |
+
"bos_token_id": 0,
|
| 9 |
+
"classifier_dropout": 0.1,
|
| 10 |
+
"eos_token_id": 0,
|
| 11 |
+
"hidden_act": "gelu",
|
| 12 |
+
"hidden_dropout": 0.0,
|
| 13 |
+
"hidden_size": 768,
|
| 14 |
+
"initializer_range": 0.02,
|
| 15 |
+
"intermediate_size": 3072,
|
| 16 |
+
"layer_norm_eps": 1e-05,
|
| 17 |
+
"max_position_embeddings": 2048,
|
| 18 |
+
"model_type": "gpt_neox",
|
| 19 |
+
"num_attention_heads": 12,
|
| 20 |
+
"num_hidden_layers": 12,
|
| 21 |
+
"partial_rotary_factor": 0.25,
|
| 22 |
+
"rope_scaling": null,
|
| 23 |
+
"rope_theta": 10000,
|
| 24 |
+
"rotary_emb_base": 10000,
|
| 25 |
+
"rotary_pct": 0.25,
|
| 26 |
+
"tie_word_embeddings": false,
|
| 27 |
+
"torch_dtype": "float32",
|
| 28 |
+
"transformers_version": "4.46.2",
|
| 29 |
+
"use_cache": true,
|
| 30 |
+
"use_parallel_residual": true,
|
| 31 |
+
"vocab_size": 50304
|
| 32 |
+
}
|
generation_config.json
ADDED
|
@@ -0,0 +1,6 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_from_model_config": true,
|
| 3 |
+
"bos_token_id": 0,
|
| 4 |
+
"eos_token_id": 0,
|
| 5 |
+
"transformers_version": "4.46.2"
|
| 6 |
+
}
|
model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:9c5f335ef94c0ce86cfda59e84cdd9bbef1b433fa1974f80e4b56d4498d8c3c6
|
| 3 |
+
size 649308728
|
training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:95492cc580971074153fb0dee7b60aedfda648c5f8a7cb99bd7fbc9464d590d2
|
| 3 |
+
size 5304
|