Hugging Face's logo

cedpsam
/
EleutherAI_gpt-neo-125M-stablediffionprompts

Text Generation
Transformers
PyTorch
TensorBoard
Safetensors
gpt_neo
Generated from Trainer
Inference Endpoints
Model card Files Metrics Community
2
cedpsam commited on
Commit
7e4f678
•
1 Parent(s): 46144fe

Training in progress, step 500

Browse files
Files changed (22) hide show
  1. last-checkpoint/config.json +54 -0
  2. last-checkpoint/optimizer.pt +3 -0
  3. last-checkpoint/pytorch_model.bin +3 -0
  4. last-checkpoint/rng_state.pth +3 -0
  5. last-checkpoint/scaler.pt +3 -0
  6. last-checkpoint/scheduler.pt +3 -0
  7. last-checkpoint/trainer_state.json +22 -0
  8. last-checkpoint/training_args.bin +3 -0
  9. pytorch_model.bin +1 -1
  10. runs/Nov05_07-57-57_0b25abb50174/1667635297.5829215/events.out.tfevents.1667635297.0b25abb50174.23.1 +3 -0
  11. runs/Nov05_07-57-57_0b25abb50174/1667635311.999316/events.out.tfevents.1667635311.0b25abb50174.23.2 +3 -0
  12. runs/Nov05_07-57-57_0b25abb50174/1667635316.2924867/events.out.tfevents.1667635316.0b25abb50174.23.3 +3 -0
  13. runs/Nov05_07-57-57_0b25abb50174/1667635318.4582205/events.out.tfevents.1667635318.0b25abb50174.23.4 +3 -0
  14. runs/Nov05_07-57-57_0b25abb50174/1667635319.7992213/events.out.tfevents.1667635319.0b25abb50174.23.5 +3 -0
  15. runs/Nov05_07-57-57_0b25abb50174/1667635320.7718089/events.out.tfevents.1667635320.0b25abb50174.23.6 +3 -0
  16. runs/Nov05_07-57-57_0b25abb50174/1667635321.6110454/events.out.tfevents.1667635321.0b25abb50174.23.7 +3 -0
  17. runs/Nov05_07-57-57_0b25abb50174/1667635322.041799/events.out.tfevents.1667635322.0b25abb50174.23.8 +3 -0
  18. runs/Nov05_07-57-57_0b25abb50174/1667635322.8277125/events.out.tfevents.1667635322.0b25abb50174.23.9 +3 -0
  19. runs/Nov05_07-57-57_0b25abb50174/1667635323.5899823/events.out.tfevents.1667635323.0b25abb50174.23.10 +3 -0
  20. runs/Nov05_07-57-57_0b25abb50174/1667635325.0169144/events.out.tfevents.1667635325.0b25abb50174.23.11 +3 -0
  21. runs/Nov05_07-57-57_0b25abb50174/events.out.tfevents.1667635297.0b25abb50174.23.0 +3 -0
  22. training_args.bin +1 -1
last-checkpoint/config.json ADDED
@@ -0,0 +1,54 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "cedpsam/EleutherAI_gpt-neo-125M-stablediffionprompts",
3
+ "activation_function": "gelu_new",
4
+ "architectures": [
5
+ "GPTNeoForCausalLM"
6
+ ],
7
+ "attention_dropout": 0,
8
+ "attention_layers": [
9
+ "global",
10
+ "local",
11
+ "global",
12
+ "local",
13
+ "global",
14
+ "local",
15
+ "global",
16
+ "local",
17
+ "global",
18
+ "local",
19
+ "global",
20
+ "local"
21
+ ],
22
+ "attention_types": [
23
+ [
24
+ [
25
+ "global",
26
+ "local"
27
+ ],
28
+ 6
29
+ ]
30
+ ],
31
+ "bos_token_id": 50256,
32
+ "embed_dropout": 0,
33
+ "eos_token_id": 50256,
34
+ "gradient_checkpointing": false,
35
+ "hidden_size": 768,
36
+ "initializer_range": 0.02,
37
+ "intermediate_size": null,
38
+ "layer_norm_epsilon": 1e-05,
39
+ "max_position_embeddings": 2048,
40
+ "model_type": "gpt_neo",
41
+ "num_heads": 12,
42
+ "num_layers": 12,
43
+ "resid_dropout": 0,
44
+ "summary_activation": null,
45
+ "summary_first_dropout": 0.1,
46
+ "summary_proj_to_labels": true,
47
+ "summary_type": "cls_index",
48
+ "summary_use_proj": true,
49
+ "torch_dtype": "float32",
50
+ "transformers_version": "4.20.1",
51
+ "use_cache": true,
52
+ "vocab_size": 50257,
53
+ "window_size": 256
54
+ }
last-checkpoint/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e7b4238dbcc86ddd648d40b1c597c6fea79f3cc7777622003f0442b5dc6e5bdf
3
+ size 1001681601
last-checkpoint/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a00135f4f5bb0ff45565a049b47e543e84f59d9399362908abf186b6a8ba93c9
3
+ size 551185105
last-checkpoint/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:62851f63d26f31f0098aaf22bfc590da9d5dd492f2a149b0c502c13be7100a3d
3
+ size 14503
last-checkpoint/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3978f33a9eccdea52f77a26979e6a0077a50cd239694fba6bc25cd0891e8edf1
3
+ size 559
last-checkpoint/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4af5659a4ffafd6e5e4486d3da6802b3d89fc41fd3462038b6768c0dd81c78f7
3
+ size 623
last-checkpoint/trainer_state.json ADDED
@@ -0,0 +1,22 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 0.006782603977318972,
5
+ "global_step": 500,
6
+ "is_hyper_param_search": false,
7
+ "is_local_process_zero": true,
8
+ "is_world_process_zero": true,
9
+ "log_history": [
10
+ {
11
+ "epoch": 0.01,
12
+ "learning_rate": 9.886136363636363e-08,
13
+ "loss": 0.0254,
14
+ "step": 500
15
+ }
16
+ ],
17
+ "max_steps": 44000,
18
+ "num_train_epochs": 1,
19
+ "total_flos": 524503851466752.0,
20
+ "trial_name": null,
21
+ "trial_params": null
22
+ }
last-checkpoint/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:635866ae1eed60031c299fe67aead1ce34854e11198992998523b4a37d002974
3
+ size 3439
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:cecc6aeb658ce5cddba59472bfbea3e76852b232a433f21fe1213ccd1ba34907
3
  size 551185105
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a00135f4f5bb0ff45565a049b47e543e84f59d9399362908abf186b6a8ba93c9
3
  size 551185105
runs/Nov05_07-57-57_0b25abb50174/1667635297.5829215/events.out.tfevents.1667635297.0b25abb50174.23.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6a362fec0acf0068f0db968fcf54df8c38694d531952e1385cad18a8a47a4a8a
3
+ size 5438
runs/Nov05_07-57-57_0b25abb50174/1667635311.999316/events.out.tfevents.1667635311.0b25abb50174.23.2 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1a189df51830bace7dae4cf2c22d3e89fef59e57565366c5e46bab5790ae3b30
3
+ size 5438
runs/Nov05_07-57-57_0b25abb50174/1667635316.2924867/events.out.tfevents.1667635316.0b25abb50174.23.3 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:49f69d76979a2dee3c886666db5625f2a06b79761369db22d3e0d4f47352489f
3
+ size 5438
runs/Nov05_07-57-57_0b25abb50174/1667635318.4582205/events.out.tfevents.1667635318.0b25abb50174.23.4 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4ec2811bbe74fe07a63ea593197ed2d5925ff2ea069b1368f9187053599b16df
3
+ size 5438
runs/Nov05_07-57-57_0b25abb50174/1667635319.7992213/events.out.tfevents.1667635319.0b25abb50174.23.5 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e90fbc7149d830375468cb3f6f701677ea75fcc593c93b6ca5474a314cd044cd
3
+ size 5438
runs/Nov05_07-57-57_0b25abb50174/1667635320.7718089/events.out.tfevents.1667635320.0b25abb50174.23.6 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9e2612878a0feb2bd91ae93ff9d250bea8b451e63c4789151f68645a5860b274
3
+ size 5438
runs/Nov05_07-57-57_0b25abb50174/1667635321.6110454/events.out.tfevents.1667635321.0b25abb50174.23.7 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a0d9c8d2482867899999ea6de97eda4d0aff9ef2343621156a82cb6cb5dbe519
3
+ size 5438
runs/Nov05_07-57-57_0b25abb50174/1667635322.041799/events.out.tfevents.1667635322.0b25abb50174.23.8 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a04ca6e9575defad90e87bc9a5c5d09aca27acee1deb12842bc44fd9411eb1c4
3
+ size 5438
runs/Nov05_07-57-57_0b25abb50174/1667635322.8277125/events.out.tfevents.1667635322.0b25abb50174.23.9 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d3c170be1d098d570c9b1b4260db20d781ff0a01e0718c5f411ef9160351a31a
3
+ size 5438
runs/Nov05_07-57-57_0b25abb50174/1667635323.5899823/events.out.tfevents.1667635323.0b25abb50174.23.10 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:cb1a797b6719a17c2c4c11c681fea8dc01fe7647f4a607183cb10cf11fef985c
3
+ size 5438
runs/Nov05_07-57-57_0b25abb50174/1667635325.0169144/events.out.tfevents.1667635325.0b25abb50174.23.11 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:d3dcfa837a83fe459bc5f0328be3050d912d00db708d796e57e4921887a074e4
3
+ size 5438
runs/Nov05_07-57-57_0b25abb50174/events.out.tfevents.1667635297.0b25abb50174.23.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:349d49a2e780996cc73c66f1f400966f1be92733df20a402c652a719b77ad078
3
+ size 45803
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:0a9be4165819bdb518e6ce2fb6609c1da8c8f0fe6577ac5717e2df558fc930ee
3
  size 3439
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:635866ae1eed60031c299fe67aead1ce34854e11198992998523b4a37d002974
3
  size 3439