Hugging Face's logo

unionai
/
pythia-1B-deduped-wikipedia-8bit

Text Generation
Transformers
PyTorch
English
gpt_neox
causal-lm
pythia
fine-tuning
wikipedia
text-generation-inference
Inference Endpoints
Model card Files Community
1
nielsbantilan commited on
Commit
891d8d5
·
1 Parent(s): 51ad877

Upload folder using huggingface_hub

Browse files
Files changed (49) hide show
  1. .gitattributes +9 -0
  2. config.json +26 -0
  3. flyte_training_config.json +1 -0
  4. flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/config.json +26 -0
  5. flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/debug-cli.root.log +0 -0
  6. flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/flyte-ez1nahht/sandbox/local_flytekit/ede012531e15ed10d52fdd2ae3481a91/00000 +3 -0
  7. flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/flyte-fvb0g199/sandbox/local_flytekit/80ad53ec846c80e861bcc631bf298149/00000 +3 -0
  8. flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/flyte-km4yz5vs/sandbox/local_flytekit/ee9bb692f1a49ac4a87c207c9b935fd3/00000 +3 -0
  9. flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/flyte-mbnq5p0f/sandbox/local_flytekit/fd378b2fbca9eab3c7673e2c4117d552/00000 +3 -0
  10. flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/flyte-mzuqx55l/sandbox/local_flytekit/2731c7daa7ff8d2e77c90c793b3059dd/00000 +3 -0
  11. flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/flyte-t4w0b1ci/sandbox/local_flytekit/2bd86a648a2e35e5e40f4396f3e12b35/00000 +3 -0
  12. flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/flyte-vink1pbs/sandbox/local_flytekit/5c0420d4c4b3f4d23493c53a092687d0/00000 +3 -0
  13. flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/flyte-wap6j530/sandbox/local_flytekit/inputs.pb +3 -0
  14. flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/flyte-zlncfcoi/sandbox/local_flytekit/6aee1fb3e9250ccde375fc50d36bc745/00000 +3 -0
  15. flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/flyte_training_config.json +1 -0
  16. flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/generation_config.json +6 -0
  17. flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/models--EleutherAI--pythia-1B-deduped/.no_exist/46811f6b6cf475f7d242d310bb7073ad20a54d6b/added_tokens.json +0 -0
  18. flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/models--EleutherAI--pythia-1B-deduped/.no_exist/46811f6b6cf475f7d242d310bb7073ad20a54d6b/generation_config.json +0 -0
  19. flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/models--EleutherAI--pythia-1B-deduped/.no_exist/46811f6b6cf475f7d242d310bb7073ad20a54d6b/merges.txt +0 -0
  20. flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/models--EleutherAI--pythia-1B-deduped/.no_exist/46811f6b6cf475f7d242d310bb7073ad20a54d6b/vocab.json +0 -0
  21. flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/models--EleutherAI--pythia-1B-deduped/blobs/0204ed10c186a4c7c68f55dff8f26087a45898d6 +5 -0
  22. flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/models--EleutherAI--pythia-1B-deduped/blobs/0204ed10c186a4c7c68f55dff8f26087a45898d6.lock +0 -0
  23. flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/models--EleutherAI--pythia-1B-deduped/blobs/0677b767d0e8c7dc792f2848d0fb74db3a03419e89b7f05c7421fe066f1906eb +3 -0
  24. flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/models--EleutherAI--pythia-1B-deduped/blobs/0677b767d0e8c7dc792f2848d0fb74db3a03419e89b7f05c7421fe066f1906eb.lock +0 -0
  25. flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/models--EleutherAI--pythia-1B-deduped/blobs/631d0e859247ad69414b72e657e7471a07403090 +24 -0
  26. flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/models--EleutherAI--pythia-1B-deduped/blobs/631d0e859247ad69414b72e657e7471a07403090.lock +0 -0
  27. flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/models--EleutherAI--pythia-1B-deduped/blobs/f1860edb10f80bcaf7b023fce47c68a23b724c23 +9 -0
  28. flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/models--EleutherAI--pythia-1B-deduped/blobs/f1860edb10f80bcaf7b023fce47c68a23b724c23.lock +0 -0
  29. flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/models--EleutherAI--pythia-1B-deduped/blobs/f74dfbfab8f97770a87769c739fb080c21c8bacc +0 -0
  30. flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/models--EleutherAI--pythia-1B-deduped/blobs/f74dfbfab8f97770a87769c739fb080c21c8bacc.lock +0 -0
  31. flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/models--EleutherAI--pythia-1B-deduped/refs/main +1 -0
  32. flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/models--EleutherAI--pythia-1B-deduped/snapshots/46811f6b6cf475f7d242d310bb7073ad20a54d6b/config.json +24 -0
  33. flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/models--EleutherAI--pythia-1B-deduped/snapshots/46811f6b6cf475f7d242d310bb7073ad20a54d6b/pytorch_model.bin +3 -0
  34. flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/models--EleutherAI--pythia-1B-deduped/snapshots/46811f6b6cf475f7d242d310bb7073ad20a54d6b/special_tokens_map.json +5 -0
  35. flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/models--EleutherAI--pythia-1B-deduped/snapshots/46811f6b6cf475f7d242d310bb7073ad20a54d6b/tokenizer.json +0 -0
  36. flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/models--EleutherAI--pythia-1B-deduped/snapshots/46811f6b6cf475f7d242d310bb7073ad20a54d6b/tokenizer_config.json +9 -0
  37. flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/pytorch_model.bin +3 -0
  38. flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/special_tokens_map.json +6 -0
  39. flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/tokenizer.json +0 -0
  40. flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/tokenizer_config.json +11 -0
  41. flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/trainer_state.json +31 -0
  42. flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/training_args.bin +3 -0
  43. flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/version.txt +1 -0
  44. generation_config.json +6 -0
  45. pytorch_model.bin +3 -0
  46. special_tokens_map.json +6 -0
  47. tokenizer.json +0 -0
  48. tokenizer_config.json +11 -0
  49. training_args.bin +3 -0
.gitattributes CHANGED
@@ -32,3 +32,12 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
32
  *.zip filter=lfs diff=lfs merge=lfs -text
33
  *.zst filter=lfs diff=lfs merge=lfs -text
34
  *tfevents* filter=lfs diff=lfs merge=lfs -text
 
 
 
 
 
 
 
 
 
 
32
  *.zip filter=lfs diff=lfs merge=lfs -text
33
  *.zst filter=lfs diff=lfs merge=lfs -text
34
  *tfevents* filter=lfs diff=lfs merge=lfs -text
35
+ flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/flyte-ez1nahht/sandbox/local_flytekit/ede012531e15ed10d52fdd2ae3481a91/00000 filter=lfs diff=lfs merge=lfs -text
36
+ flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/flyte-fvb0g199/sandbox/local_flytekit/80ad53ec846c80e861bcc631bf298149/00000 filter=lfs diff=lfs merge=lfs -text
37
+ flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/flyte-km4yz5vs/sandbox/local_flytekit/ee9bb692f1a49ac4a87c207c9b935fd3/00000 filter=lfs diff=lfs merge=lfs -text
38
+ flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/flyte-mbnq5p0f/sandbox/local_flytekit/fd378b2fbca9eab3c7673e2c4117d552/00000 filter=lfs diff=lfs merge=lfs -text
39
+ flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/flyte-mzuqx55l/sandbox/local_flytekit/2731c7daa7ff8d2e77c90c793b3059dd/00000 filter=lfs diff=lfs merge=lfs -text
40
+ flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/flyte-t4w0b1ci/sandbox/local_flytekit/2bd86a648a2e35e5e40f4396f3e12b35/00000 filter=lfs diff=lfs merge=lfs -text
41
+ flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/flyte-vink1pbs/sandbox/local_flytekit/5c0420d4c4b3f4d23493c53a092687d0/00000 filter=lfs diff=lfs merge=lfs -text
42
+ flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/flyte-zlncfcoi/sandbox/local_flytekit/6aee1fb3e9250ccde375fc50d36bc745/00000 filter=lfs diff=lfs merge=lfs -text
43
+ flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/models--EleutherAI--pythia-1B-deduped/blobs/0677b767d0e8c7dc792f2848d0fb74db3a03419e89b7f05c7421fe066f1906eb filter=lfs diff=lfs merge=lfs -text
config.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "EleutherAI/pythia-1B-deduped",
3
+ "architectures": [
4
+ "GPTNeoXForCausalLM"
5
+ ],
6
+ "bos_token_id": 0,
7
+ "classifier_dropout": 0.1,
8
+ "eos_token_id": 0,
9
+ "hidden_act": "gelu",
10
+ "hidden_size": 2048,
11
+ "initializer_range": 0.02,
12
+ "intermediate_size": 8192,
13
+ "layer_norm_eps": 1e-05,
14
+ "max_position_embeddings": 2048,
15
+ "model_type": "gpt_neox",
16
+ "num_attention_heads": 8,
17
+ "num_hidden_layers": 16,
18
+ "rotary_emb_base": 10000,
19
+ "rotary_pct": 0.25,
20
+ "tie_word_embeddings": false,
21
+ "torch_dtype": "float16",
22
+ "transformers_version": "4.29.2",
23
+ "use_cache": true,
24
+ "use_parallel_residual": true,
25
+ "vocab_size": 50304
26
+ }
flyte_training_config.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"base_model": "EleutherAI/pythia-1B-deduped", "data_path": "wikipedia", "data_name": "20220301.simple", "num_epochs": 1, "max_steps": 10, "learning_rate": 2e-05, "weight_decay": 0.02, "warmup_ratio": 0.03, "lr_scheduler_type": "cosine", "batch_size": 8, "micro_batch_size": 1, "val_set_size": 0, "group_by_length": false, "instruction_key": "instruction", "input_key": "input", "output_key": "output", "device_map": "auto", "cache_dir": null, "optim": "adamw_torch", "model_max_length": 512, "debug_mode": false, "debug_train_data_size": 1024, "wandb_project": ""}
flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/config.json ADDED
@@ -0,0 +1,26 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "EleutherAI/pythia-1B-deduped",
3
+ "architectures": [
4
+ "GPTNeoXForCausalLM"
5
+ ],
6
+ "bos_token_id": 0,
7
+ "classifier_dropout": 0.1,
8
+ "eos_token_id": 0,
9
+ "hidden_act": "gelu",
10
+ "hidden_size": 2048,
11
+ "initializer_range": 0.02,
12
+ "intermediate_size": 8192,
13
+ "layer_norm_eps": 1e-05,
14
+ "max_position_embeddings": 2048,
15
+ "model_type": "gpt_neox",
16
+ "num_attention_heads": 8,
17
+ "num_hidden_layers": 16,
18
+ "rotary_emb_base": 10000,
19
+ "rotary_pct": 0.25,
20
+ "tie_word_embeddings": false,
21
+ "torch_dtype": "float16",
22
+ "transformers_version": "4.29.2",
23
+ "use_cache": true,
24
+ "use_parallel_residual": true,
25
+ "vocab_size": 50304
26
+ }
flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/debug-cli.root.log ADDED
File without changes
flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/flyte-ez1nahht/sandbox/local_flytekit/ede012531e15ed10d52fdd2ae3481a91/00000 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:067772915d011157436dc1ea88cb38756555e25be2d07616d1ee97dfac6e6535
3
+ size 133886409
flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/flyte-fvb0g199/sandbox/local_flytekit/80ad53ec846c80e861bcc631bf298149/00000 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:067772915d011157436dc1ea88cb38756555e25be2d07616d1ee97dfac6e6535
3
+ size 133886409
flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/flyte-km4yz5vs/sandbox/local_flytekit/ee9bb692f1a49ac4a87c207c9b935fd3/00000 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:067772915d011157436dc1ea88cb38756555e25be2d07616d1ee97dfac6e6535
3
+ size 133886409
flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/flyte-mbnq5p0f/sandbox/local_flytekit/fd378b2fbca9eab3c7673e2c4117d552/00000 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:067772915d011157436dc1ea88cb38756555e25be2d07616d1ee97dfac6e6535
3
+ size 133886409
flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/flyte-mzuqx55l/sandbox/local_flytekit/2731c7daa7ff8d2e77c90c793b3059dd/00000 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:067772915d011157436dc1ea88cb38756555e25be2d07616d1ee97dfac6e6535
3
+ size 133886409
flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/flyte-t4w0b1ci/sandbox/local_flytekit/2bd86a648a2e35e5e40f4396f3e12b35/00000 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:067772915d011157436dc1ea88cb38756555e25be2d07616d1ee97dfac6e6535
3
+ size 133886409
flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/flyte-vink1pbs/sandbox/local_flytekit/5c0420d4c4b3f4d23493c53a092687d0/00000 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:067772915d011157436dc1ea88cb38756555e25be2d07616d1ee97dfac6e6535
3
+ size 133886409
flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/flyte-wap6j530/sandbox/local_flytekit/inputs.pb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:82b8805a426e9819e3184337fdc598080d62922b8f09653a7afdde1bad73b548
3
+ size 1823
flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/flyte-zlncfcoi/sandbox/local_flytekit/6aee1fb3e9250ccde375fc50d36bc745/00000 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:067772915d011157436dc1ea88cb38756555e25be2d07616d1ee97dfac6e6535
3
+ size 133886409
flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/flyte_training_config.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {"base_model": "EleutherAI/pythia-1B-deduped", "data_path": "wikipedia", "data_name": "20220301.simple", "num_epochs": 1, "max_steps": 10, "learning_rate": 2e-05, "weight_decay": 0.02, "warmup_ratio": 0.03, "lr_scheduler_type": "cosine", "batch_size": 8, "micro_batch_size": 1, "val_set_size": 0, "group_by_length": false, "instruction_key": "instruction", "input_key": "input", "output_key": "output", "device_map": "auto", "cache_dir": null, "optim": "adamw_torch", "model_max_length": 512, "debug_mode": false, "debug_train_data_size": 1024, "wandb_project": ""}
flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/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": "4.29.2"
6
+ }
flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/models--EleutherAI--pythia-1B-deduped/.no_exist/46811f6b6cf475f7d242d310bb7073ad20a54d6b/added_tokens.json ADDED
File without changes
flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/models--EleutherAI--pythia-1B-deduped/.no_exist/46811f6b6cf475f7d242d310bb7073ad20a54d6b/generation_config.json ADDED
File without changes
flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/models--EleutherAI--pythia-1B-deduped/.no_exist/46811f6b6cf475f7d242d310bb7073ad20a54d6b/merges.txt ADDED
File without changes
flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/models--EleutherAI--pythia-1B-deduped/.no_exist/46811f6b6cf475f7d242d310bb7073ad20a54d6b/vocab.json ADDED
File without changes
flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/models--EleutherAI--pythia-1B-deduped/blobs/0204ed10c186a4c7c68f55dff8f26087a45898d6 ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<|endoftext|>",
3
+ "eos_token": "<|endoftext|>",
4
+ "unk_token": "<|endoftext|>"
5
+ }
flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/models--EleutherAI--pythia-1B-deduped/blobs/0204ed10c186a4c7c68f55dff8f26087a45898d6.lock ADDED
File without changes
flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/models--EleutherAI--pythia-1B-deduped/blobs/0677b767d0e8c7dc792f2848d0fb74db3a03419e89b7f05c7421fe066f1906eb ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0677b767d0e8c7dc792f2848d0fb74db3a03419e89b7f05c7421fe066f1906eb
3
+ size 2090751965
flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/models--EleutherAI--pythia-1B-deduped/blobs/0677b767d0e8c7dc792f2848d0fb74db3a03419e89b7f05c7421fe066f1906eb.lock ADDED
File without changes
flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/models--EleutherAI--pythia-1B-deduped/blobs/631d0e859247ad69414b72e657e7471a07403090 ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "GPTNeoXForCausalLM"
4
+ ],
5
+ "bos_token_id": 0,
6
+ "eos_token_id": 0,
7
+ "hidden_act": "gelu",
8
+ "hidden_size": 2048,
9
+ "initializer_range": 0.02,
10
+ "intermediate_size": 8192,
11
+ "layer_norm_eps": 1e-05,
12
+ "max_position_embeddings": 2048,
13
+ "model_type": "gpt_neox",
14
+ "num_attention_heads": 8,
15
+ "num_hidden_layers": 16,
16
+ "rotary_emb_base": 10000,
17
+ "rotary_pct": 0.25,
18
+ "tie_word_embeddings": false,
19
+ "torch_dtype": "float16",
20
+ "transformers_version": "4.24.0",
21
+ "use_cache": true,
22
+ "use_parallel_residual": true,
23
+ "vocab_size": 50304
24
+ }
flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/models--EleutherAI--pythia-1B-deduped/blobs/631d0e859247ad69414b72e657e7471a07403090.lock ADDED
File without changes
flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/models--EleutherAI--pythia-1B-deduped/blobs/f1860edb10f80bcaf7b023fce47c68a23b724c23 ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "bos_token": "<|endoftext|>",
4
+ "eos_token": "<|endoftext|>",
5
+ "name_or_path": "EleutherAI/gpt-neox-20b",
6
+ "special_tokens_map_file": "/admin/home-hailey/.cache/huggingface/hub/models--EleutherAI--gpt-neox-20b/snapshots/4e49eadb5d14bd22f314ec3f45b69a87b88c7691/special_tokens_map.json",
7
+ "tokenizer_class": "GPTNeoXTokenizer",
8
+ "unk_token": "<|endoftext|>"
9
+ }
flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/models--EleutherAI--pythia-1B-deduped/blobs/f1860edb10f80bcaf7b023fce47c68a23b724c23.lock ADDED
File without changes
flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/models--EleutherAI--pythia-1B-deduped/blobs/f74dfbfab8f97770a87769c739fb080c21c8bacc ADDED
The diff for this file is too large to render. See raw diff
 
flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/models--EleutherAI--pythia-1B-deduped/blobs/f74dfbfab8f97770a87769c739fb080c21c8bacc.lock ADDED
File without changes
flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/models--EleutherAI--pythia-1B-deduped/refs/main ADDED
@@ -0,0 +1 @@
 
 
1
+ 46811f6b6cf475f7d242d310bb7073ad20a54d6b
flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/models--EleutherAI--pythia-1B-deduped/snapshots/46811f6b6cf475f7d242d310bb7073ad20a54d6b/config.json ADDED
@@ -0,0 +1,24 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "architectures": [
3
+ "GPTNeoXForCausalLM"
4
+ ],
5
+ "bos_token_id": 0,
6
+ "eos_token_id": 0,
7
+ "hidden_act": "gelu",
8
+ "hidden_size": 2048,
9
+ "initializer_range": 0.02,
10
+ "intermediate_size": 8192,
11
+ "layer_norm_eps": 1e-05,
12
+ "max_position_embeddings": 2048,
13
+ "model_type": "gpt_neox",
14
+ "num_attention_heads": 8,
15
+ "num_hidden_layers": 16,
16
+ "rotary_emb_base": 10000,
17
+ "rotary_pct": 0.25,
18
+ "tie_word_embeddings": false,
19
+ "torch_dtype": "float16",
20
+ "transformers_version": "4.24.0",
21
+ "use_cache": true,
22
+ "use_parallel_residual": true,
23
+ "vocab_size": 50304
24
+ }
flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/models--EleutherAI--pythia-1B-deduped/snapshots/46811f6b6cf475f7d242d310bb7073ad20a54d6b/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0677b767d0e8c7dc792f2848d0fb74db3a03419e89b7f05c7421fe066f1906eb
3
+ size 2090751965
flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/models--EleutherAI--pythia-1B-deduped/snapshots/46811f6b6cf475f7d242d310bb7073ad20a54d6b/special_tokens_map.json ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<|endoftext|>",
3
+ "eos_token": "<|endoftext|>",
4
+ "unk_token": "<|endoftext|>"
5
+ }
flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/models--EleutherAI--pythia-1B-deduped/snapshots/46811f6b6cf475f7d242d310bb7073ad20a54d6b/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/models--EleutherAI--pythia-1B-deduped/snapshots/46811f6b6cf475f7d242d310bb7073ad20a54d6b/tokenizer_config.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "bos_token": "<|endoftext|>",
4
+ "eos_token": "<|endoftext|>",
5
+ "name_or_path": "EleutherAI/gpt-neox-20b",
6
+ "special_tokens_map_file": "/admin/home-hailey/.cache/huggingface/hub/models--EleutherAI--gpt-neox-20b/snapshots/4e49eadb5d14bd22f314ec3f45b69a87b88c7691/special_tokens_map.json",
7
+ "tokenizer_class": "GPTNeoXTokenizer",
8
+ "unk_token": "<|endoftext|>"
9
+ }
flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:247c2c8d4e70d963e87cfcd9c5ff7c32aac07fc279083a9c02af4c141ad296fc
3
+ size 2090751965
flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/special_tokens_map.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<|endoftext|>",
3
+ "eos_token": "<|endoftext|>",
4
+ "pad_token": "[PAD]",
5
+ "unk_token": "<|endoftext|>"
6
+ }
flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/tokenizer_config.json ADDED
@@ -0,0 +1,11 @@
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "bos_token": "<|endoftext|>",
4
+ "clean_up_tokenization_spaces": true,
5
+ "eos_token": "<|endoftext|>",
6
+ "model_max_length": 512,
7
+ "pad_token": "[PAD]",
8
+ "padding_side": "right",
9
+ "tokenizer_class": "GPTNeoXTokenizer",
10
+ "unk_token": "<|endoftext|>"
11
+ }
flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/trainer_state.json ADDED
@@ -0,0 +1,31 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 3.3333333333333335,
5
+ "global_step": 10,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 3.33,
12
+ "learning_rate": 2e-05,
13
+ "loss": 2.1837,
14
+ "step": 10
15
+ },
16
+ {
17
+ "epoch": 3.33,
18
+ "step": 10,
19
+ "total_flos": 835002040320.0,
20
+ "train_loss": 2.1836517333984373,
21
+ "train_runtime": 74.2827,
22
+ "train_samples_per_second": 8.616,
23
+ "train_steps_per_second": 0.135
24
+ }
25
+ ],
26
+ "max_steps": 10,
27
+ "num_train_epochs": 4,
28
+ "total_flos": 835002040320.0,
29
+ "trial_name": null,
30
+ "trial_params": null
31
+ }
flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dccd10791eeca2765d0dab04a01995f27d811561179ad9f79d8ca91bbc1bae49
3
+ size 5499
flyteva2a9_9t/local_flytekit/df86818aaa4263bdf5800c2d18124919/version.txt ADDED
@@ -0,0 +1 @@
 
 
1
+ 1
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": "4.29.2"
6
+ }
pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3815356a26467dbb22ff4925dad4b33727d4c7c3d9816e9872711058d69016c5
3
+ size 1285449053
special_tokens_map.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<|endoftext|>",
3
+ "eos_token": "<|endoftext|>",
4
+ "pad_token": "[PAD]",
5
+ "unk_token": "<|endoftext|>"
6
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,11 @@
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "bos_token": "<|endoftext|>",
4
+ "clean_up_tokenization_spaces": true,
5
+ "eos_token": "<|endoftext|>",
6
+ "model_max_length": 512,
7
+ "pad_token": "[PAD]",
8
+ "padding_side": "right",
9
+ "tokenizer_class": "GPTNeoXTokenizer",
10
+ "unk_token": "<|endoftext|>"
11
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c85c668f7c6177cf1bc693b5aa289624c794d7783baa149a857a3634c259c98f
3
+ size 3899