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
537f1dd
•
1 Parent(s): e1ce73b

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/Oct24_08-39-52_00e58a194ed0/1666601028.9643993/events.out.tfevents.1666601028.00e58a194ed0.23.1 +3 -0
  11. runs/Oct24_08-39-52_00e58a194ed0/1666601044.218391/events.out.tfevents.1666601044.00e58a194ed0.23.2 +3 -0
  12. runs/Oct24_08-39-52_00e58a194ed0/1666601049.091/events.out.tfevents.1666601049.00e58a194ed0.23.3 +3 -0
  13. runs/Oct24_08-39-52_00e58a194ed0/1666601051.1071177/events.out.tfevents.1666601051.00e58a194ed0.23.4 +3 -0
  14. runs/Oct24_08-39-52_00e58a194ed0/1666601052.2182786/events.out.tfevents.1666601052.00e58a194ed0.23.5 +3 -0
  15. runs/Oct24_08-39-52_00e58a194ed0/1666601052.8901308/events.out.tfevents.1666601052.00e58a194ed0.23.6 +3 -0
  16. runs/Oct24_08-39-52_00e58a194ed0/1666601053.5025167/events.out.tfevents.1666601053.00e58a194ed0.23.7 +3 -0
  17. runs/Oct24_08-39-52_00e58a194ed0/1666601053.927081/events.out.tfevents.1666601053.00e58a194ed0.23.8 +3 -0
  18. runs/Oct24_08-39-52_00e58a194ed0/1666601054.8511527/events.out.tfevents.1666601054.00e58a194ed0.23.9 +3 -0
  19. runs/Oct24_08-39-52_00e58a194ed0/1666601055.629219/events.out.tfevents.1666601055.00e58a194ed0.23.10 +3 -0
  20. runs/Oct24_08-39-52_00e58a194ed0/1666601057.0422728/events.out.tfevents.1666601057.00e58a194ed0.23.11 +3 -0
  21. runs/Oct24_08-39-52_00e58a194ed0/events.out.tfevents.1666601028.00e58a194ed0.23.0 +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:b27276104532404454dbdaeef79485d26921a8eec22bd1db60987316f304dd36
3
+ size 1001681601
last-checkpoint/pytorch_model.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:512999908fe1282992a2a78aea4b69ebed6979af7149a65b6809451cf3176ab4
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": 0.024,
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 sha256:0a9be4165819bdb518e6ce2fb6609c1da8c8f0fe6577ac5717e2df558fc930ee
3
+ size 3439
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:8664a18bcf5e21bd68d91625d161367ba18b6c0ec7cc6b8dac4d3430f0ad3e4c
3
  size 551185105
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:512999908fe1282992a2a78aea4b69ebed6979af7149a65b6809451cf3176ab4
3
  size 551185105
runs/Oct24_08-39-52_00e58a194ed0/1666601028.9643993/events.out.tfevents.1666601028.00e58a194ed0.23.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:261bf0a0dff081c28c96831265698dc25334ffe5f96df5480bd651c31ee67abd
3
+ size 5438
runs/Oct24_08-39-52_00e58a194ed0/1666601044.218391/events.out.tfevents.1666601044.00e58a194ed0.23.2 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:98f2a5c08a7a47c7e7b39fbae9188ff9260ad0b7a60e8c1184e2a74e9a5a4df8
3
+ size 5438
runs/Oct24_08-39-52_00e58a194ed0/1666601049.091/events.out.tfevents.1666601049.00e58a194ed0.23.3 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:c2080764fba6cf6fd36472f59a32c52cd40836beec9bede90345939c1c48b07a
3
+ size 5438
runs/Oct24_08-39-52_00e58a194ed0/1666601051.1071177/events.out.tfevents.1666601051.00e58a194ed0.23.4 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e6a6177271426eb705e42aa355570f4f3d43e858f19ad841a78e8ad551233fe2
3
+ size 5438
runs/Oct24_08-39-52_00e58a194ed0/1666601052.2182786/events.out.tfevents.1666601052.00e58a194ed0.23.5 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5f781f07322daf6023344881ad9ff6b8aa2ec2aa6040ae5c7e53ecf9c5c472b4
3
+ size 5438
runs/Oct24_08-39-52_00e58a194ed0/1666601052.8901308/events.out.tfevents.1666601052.00e58a194ed0.23.6 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b5a2bdd28198848bd7c70f301373aa012e69cf9d77f99b0f6db260f128398d7d
3
+ size 5438
runs/Oct24_08-39-52_00e58a194ed0/1666601053.5025167/events.out.tfevents.1666601053.00e58a194ed0.23.7 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1a313090f8374562d81df8bc14e83670de1e46098ef740dcdca796d9b0515d2c
3
+ size 5438
runs/Oct24_08-39-52_00e58a194ed0/1666601053.927081/events.out.tfevents.1666601053.00e58a194ed0.23.8 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:193e39065aece798d62df40cb6e2b8696c55b77798af476f4548ec8458b0a6c5
3
+ size 5438
runs/Oct24_08-39-52_00e58a194ed0/1666601054.8511527/events.out.tfevents.1666601054.00e58a194ed0.23.9 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4e9d94ecd4f782c540c2ad6df8301e8782f683171c6fddb02de112ade0649cad
3
+ size 5438
runs/Oct24_08-39-52_00e58a194ed0/1666601055.629219/events.out.tfevents.1666601055.00e58a194ed0.23.10 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:63f33c87df8dbcbd84957960634efa3eb1855829b2347313bbc57960cc2d0884
3
+ size 5438
runs/Oct24_08-39-52_00e58a194ed0/1666601057.0422728/events.out.tfevents.1666601057.00e58a194ed0.23.11 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:975d8f06e82d299e11891f816a68e0c07e495ac7dd9c1dbe208f34bbc42bbf26
3
+ size 5438
runs/Oct24_08-39-52_00e58a194ed0/events.out.tfevents.1666601028.00e58a194ed0.23.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a33c601a4edbe355148dc912dea4682ea4089e752a21ebd87f7c202781cee530
3
+ size 45803
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:ba5cc09c321a679be90f03709cafea28721cc11a1abc076f191fcf60d47c338b
3
  size 3439
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0a9be4165819bdb518e6ce2fb6609c1da8c8f0fe6577ac5717e2df558fc930ee
3
  size 3439