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
a01a3d8
•
1 Parent(s): 445b59c

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/Oct19_12-48-30_40567d3bb726/1666183950.8252308/events.out.tfevents.1666183950.40567d3bb726.23.1 +3 -0
  11. runs/Oct19_12-48-30_40567d3bb726/1666183965.7741637/events.out.tfevents.1666183965.40567d3bb726.23.2 +3 -0
  12. runs/Oct19_12-48-30_40567d3bb726/1666183970.0390635/events.out.tfevents.1666183970.40567d3bb726.23.3 +3 -0
  13. runs/Oct19_12-48-30_40567d3bb726/1666183972.2567992/events.out.tfevents.1666183972.40567d3bb726.23.4 +3 -0
  14. runs/Oct19_12-48-30_40567d3bb726/1666183973.615987/events.out.tfevents.1666183973.40567d3bb726.23.5 +3 -0
  15. runs/Oct19_12-48-30_40567d3bb726/1666183974.3226142/events.out.tfevents.1666183974.40567d3bb726.23.6 +3 -0
  16. runs/Oct19_12-48-30_40567d3bb726/1666183974.883771/events.out.tfevents.1666183974.40567d3bb726.23.7 +3 -0
  17. runs/Oct19_12-48-30_40567d3bb726/1666183975.2863002/events.out.tfevents.1666183975.40567d3bb726.23.8 +3 -0
  18. runs/Oct19_12-48-30_40567d3bb726/1666183976.1415691/events.out.tfevents.1666183976.40567d3bb726.23.9 +3 -0
  19. runs/Oct19_12-48-30_40567d3bb726/1666183976.9398482/events.out.tfevents.1666183976.40567d3bb726.23.10 +3 -0
  20. runs/Oct19_12-48-30_40567d3bb726/1666183978.3474762/events.out.tfevents.1666183978.40567d3bb726.23.11 +3 -0
  21. runs/Oct19_12-48-30_40567d3bb726/events.out.tfevents.1666183950.40567d3bb726.23.0 +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 sha256:07343b487b5a01c7eb9386565f3b71dd54f1fcfe3b05bdfec318d4d089f3f8bb
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:ff6b06f9126ce505ee89f47f0dcc8cb18027fb2b618c0919423d0c30a7a827c1
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:a2333e6dc66e5bc5128ed4a58628cc6055af26dd6ec990ae8cd891f0a5ea3088
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": 4.943068181818182e-05,
13
+ "loss": 0.0316,
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:e3c08e55cf4fa548ab6ad1d0976187b3df878db7839f630f4040317081083107
3
+ size 3439
pytorch_model.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:cc6f0fe8cbc4df9dafd302bcd33ce88ceafe99d520ebe8b294e6213b8ff529ed
3
  size 551185105
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ff6b06f9126ce505ee89f47f0dcc8cb18027fb2b618c0919423d0c30a7a827c1
3
  size 551185105
runs/Oct19_12-48-30_40567d3bb726/1666183950.8252308/events.out.tfevents.1666183950.40567d3bb726.23.1 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f1152b5d835fb455dabe5b328b8f26a657ea0a004e67622f73a05c6757044e06
3
+ size 5438
runs/Oct19_12-48-30_40567d3bb726/1666183965.7741637/events.out.tfevents.1666183965.40567d3bb726.23.2 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e788d1a42c2119f471a700ac4540385f56ff89311586654b3d9a75946b817a09
3
+ size 5438
runs/Oct19_12-48-30_40567d3bb726/1666183970.0390635/events.out.tfevents.1666183970.40567d3bb726.23.3 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f982f68ad36d21c827c53219785c8d4bc0f192abd81606bac7439334886cef14
3
+ size 5438
runs/Oct19_12-48-30_40567d3bb726/1666183972.2567992/events.out.tfevents.1666183972.40567d3bb726.23.4 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a5510e6f0c36cc737194821bce46a60034d336ac844400eff1f2e192bdd7aa53
3
+ size 5438
runs/Oct19_12-48-30_40567d3bb726/1666183973.615987/events.out.tfevents.1666183973.40567d3bb726.23.5 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8de750967f5481efce001b4839833cb8c16b760d5c35cc690661185d8e035567
3
+ size 5438
runs/Oct19_12-48-30_40567d3bb726/1666183974.3226142/events.out.tfevents.1666183974.40567d3bb726.23.6 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:85024c486789c53f0fa710da71e94ccf1263db52376667e3807f351501818ac3
3
+ size 5438
runs/Oct19_12-48-30_40567d3bb726/1666183974.883771/events.out.tfevents.1666183974.40567d3bb726.23.7 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:286ceff42699685fae6b6e48e08916735c151dc011e369b4d88c8726b8251761
3
+ size 5438
runs/Oct19_12-48-30_40567d3bb726/1666183975.2863002/events.out.tfevents.1666183975.40567d3bb726.23.8 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:54aae06634932845c5d76db9d1a0898abb5ed0afe13c1fd23dd76b4bf906f3e4
3
+ size 5438
runs/Oct19_12-48-30_40567d3bb726/1666183976.1415691/events.out.tfevents.1666183976.40567d3bb726.23.9 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b85943923be0da0b6d1437e4598d35c5694b7de21f90675ec2cf107045dbc431
3
+ size 5438
runs/Oct19_12-48-30_40567d3bb726/1666183976.9398482/events.out.tfevents.1666183976.40567d3bb726.23.10 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7d6314aa06344ba028041813289f75edf13c6dc4d574956be023b089db86549d
3
+ size 5438
runs/Oct19_12-48-30_40567d3bb726/1666183978.3474762/events.out.tfevents.1666183978.40567d3bb726.23.11 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:4c7f2117df09af6d507ef311b61065bad1c65407979cd495963140f7ef7b44d2
3
+ size 5438
runs/Oct19_12-48-30_40567d3bb726/events.out.tfevents.1666183950.40567d3bb726.23.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1e34471e473c2efdcd9cf42bb7d66d143477d65cfc1c43303e0abec06e4ac58a
3
+ size 45803
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:a320bfa9bffaa64280e8626313b80750a7b77117261fb20ff9a0e50028843be1
3
  size 3439
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:e3c08e55cf4fa548ab6ad1d0976187b3df878db7839f630f4040317081083107
3
  size 3439