Hugging Face's logo

craa
/
exceptions

Safetensors
Model card Files
xet
Community
craa commited on
Commit
a0241ef
·
verified ·
1 Parent(s): 6fd87e7

Upload folder using huggingface_hub

Browse files
This view is limited to 50 files because it contains too many changes.   See raw diff
Files changed (50) hide show
  1. last_to_push_frequency_5039/README.md +172 -0
  2. last_to_push_frequency_5039/all_results.json +16 -0
  3. last_to_push_frequency_5039/checkpoint-100000/config.json +31 -0
  4. last_to_push_frequency_5039/checkpoint-100000/generation_config.json +6 -0
  5. last_to_push_frequency_5039/checkpoint-100000/merges.txt +0 -0
  6. last_to_push_frequency_5039/checkpoint-100000/model.safetensors +3 -0
  7. last_to_push_frequency_5039/checkpoint-100000/optimizer.pt +3 -0
  8. last_to_push_frequency_5039/checkpoint-100000/rng_state.pth +3 -0
  9. last_to_push_frequency_5039/checkpoint-100000/scaler.pt +3 -0
  10. last_to_push_frequency_5039/checkpoint-100000/scheduler.pt +3 -0
  11. last_to_push_frequency_5039/checkpoint-100000/special_tokens_map.json +5 -0
  12. last_to_push_frequency_5039/checkpoint-100000/tokenizer.json +0 -0
  13. last_to_push_frequency_5039/checkpoint-100000/tokenizer_config.json +20 -0
  14. last_to_push_frequency_5039/checkpoint-100000/trainer_state.json +0 -0
  15. last_to_push_frequency_5039/checkpoint-100000/training_args.bin +3 -0
  16. last_to_push_frequency_5039/checkpoint-100000/vocab.json +0 -0
  17. last_to_push_frequency_5039/checkpoint-40000/config.json +31 -0
  18. last_to_push_frequency_5039/checkpoint-40000/generation_config.json +6 -0
  19. last_to_push_frequency_5039/checkpoint-40000/merges.txt +0 -0
  20. last_to_push_frequency_5039/checkpoint-40000/model.safetensors +3 -0
  21. last_to_push_frequency_5039/checkpoint-40000/optimizer.pt +3 -0
  22. last_to_push_frequency_5039/checkpoint-40000/rng_state.pth +3 -0
  23. last_to_push_frequency_5039/checkpoint-40000/scaler.pt +3 -0
  24. last_to_push_frequency_5039/checkpoint-40000/scheduler.pt +3 -0
  25. last_to_push_frequency_5039/checkpoint-40000/special_tokens_map.json +5 -0
  26. last_to_push_frequency_5039/checkpoint-40000/tokenizer.json +0 -0
  27. last_to_push_frequency_5039/checkpoint-40000/tokenizer_config.json +20 -0
  28. last_to_push_frequency_5039/checkpoint-40000/trainer_state.json +0 -0
  29. last_to_push_frequency_5039/checkpoint-40000/training_args.bin +3 -0
  30. last_to_push_frequency_5039/checkpoint-40000/vocab.json +0 -0
  31. last_to_push_frequency_5039/checkpoint-70000/config.json +31 -0
  32. last_to_push_frequency_5039/checkpoint-70000/generation_config.json +6 -0
  33. last_to_push_frequency_5039/checkpoint-70000/merges.txt +0 -0
  34. last_to_push_frequency_5039/checkpoint-70000/model.safetensors +3 -0
  35. last_to_push_frequency_5039/checkpoint-70000/optimizer.pt +3 -0
  36. last_to_push_frequency_5039/checkpoint-70000/rng_state.pth +3 -0
  37. last_to_push_frequency_5039/checkpoint-70000/scaler.pt +3 -0
  38. last_to_push_frequency_5039/checkpoint-70000/scheduler.pt +3 -0
  39. last_to_push_frequency_5039/checkpoint-70000/special_tokens_map.json +5 -0
  40. last_to_push_frequency_5039/checkpoint-70000/tokenizer.json +0 -0
  41. last_to_push_frequency_5039/checkpoint-70000/tokenizer_config.json +20 -0
  42. last_to_push_frequency_5039/checkpoint-70000/trainer_state.json +0 -0
  43. last_to_push_frequency_5039/checkpoint-70000/training_args.bin +3 -0
  44. last_to_push_frequency_5039/checkpoint-70000/vocab.json +0 -0
  45. last_to_push_frequency_5039/checkpoint-80000/config.json +31 -0
  46. last_to_push_frequency_5039/checkpoint-80000/generation_config.json +6 -0
  47. last_to_push_frequency_5039/checkpoint-80000/merges.txt +0 -0
  48. last_to_push_frequency_5039/checkpoint-80000/model.safetensors +3 -0
  49. last_to_push_frequency_5039/checkpoint-80000/optimizer.pt +3 -0
  50. last_to_push_frequency_5039/checkpoint-80000/rng_state.pth +3 -0
last_to_push_frequency_5039/README.md ADDED
@@ -0,0 +1,172 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: transformers
3
+ tags:
4
+ - generated_from_trainer
5
+ metrics:
6
+ - accuracy
7
+ model-index:
8
+ - name: exceptions_exp2_last_to_push_frequency_5039
9
+ results: []
10
+ ---
11
+
12
+ <!-- This model card has been generated automatically according to the information the Trainer had access to. You
13
+ should probably proofread and complete it, then remove this comment. -->
14
+
15
+ [<img src="https://raw.githubusercontent.com/wandb/assets/main/wandb-github-badge-28.svg" alt="Visualize in Weights & Biases" width="200" height="32"/>](https://wandb.ai/craaaa/exceptions_exp2/runs/lyyhsseg)
16
+ # exceptions_exp2_last_to_push_frequency_5039
17
+
18
+ This model is a fine-tuned version of [](https://huggingface.co/) on an unknown dataset.
19
+ It achieves the following results on the evaluation set:
20
+ - Loss: 3.5582
21
+ - Accuracy: 0.3697
22
+
23
+ ## Model description
24
+
25
+ More information needed
26
+
27
+ ## Intended uses & limitations
28
+
29
+ More information needed
30
+
31
+ ## Training and evaluation data
32
+
33
+ More information needed
34
+
35
+ ## Training procedure
36
+
37
+ ### Training hyperparameters
38
+
39
+ The following hyperparameters were used during training:
40
+ - learning_rate: 0.0006
41
+ - train_batch_size: 16
42
+ - eval_batch_size: 16
43
+ - seed: 5039
44
+ - gradient_accumulation_steps: 5
45
+ - total_train_batch_size: 80
46
+ - optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.98) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
47
+ - lr_scheduler_type: linear
48
+ - lr_scheduler_warmup_steps: 100
49
+ - num_epochs: 50.0
50
+ - mixed_precision_training: Native AMP
51
+
52
+ ### Training results
53
+
54
+ | Training Loss | Epoch | Step | Accuracy | Validation Loss |
55
+ |:-------------:|:-------:|:------:|:--------:|:---------------:|
56
+ | 4.8399 | 0.2912 | 1000 | 0.2546 | 4.7571 |
57
+ | 4.3272 | 0.5824 | 2000 | 0.3003 | 4.2753 |
58
+ | 4.1337 | 0.8737 | 3000 | 0.3159 | 4.0920 |
59
+ | 3.9915 | 1.1648 | 4000 | 0.3256 | 3.9854 |
60
+ | 3.91 | 1.4561 | 5000 | 0.3322 | 3.9122 |
61
+ | 3.8822 | 1.7473 | 6000 | 0.3370 | 3.8543 |
62
+ | 3.7549 | 2.0384 | 7000 | 0.3418 | 3.8109 |
63
+ | 3.751 | 2.3297 | 8000 | 0.3446 | 3.7817 |
64
+ | 3.7396 | 2.6209 | 9000 | 0.3475 | 3.7505 |
65
+ | 3.7105 | 2.9121 | 10000 | 0.3496 | 3.7247 |
66
+ | 3.6418 | 3.2033 | 11000 | 0.3519 | 3.7117 |
67
+ | 3.6293 | 3.4945 | 12000 | 0.3533 | 3.6937 |
68
+ | 3.642 | 3.7857 | 13000 | 0.3556 | 3.6763 |
69
+ | 3.543 | 4.0769 | 14000 | 0.3560 | 3.6722 |
70
+ | 3.5577 | 4.3681 | 15000 | 0.3575 | 3.6585 |
71
+ | 3.5824 | 4.6593 | 16000 | 0.3588 | 3.6462 |
72
+ | 3.5754 | 4.9506 | 17000 | 0.3599 | 3.6312 |
73
+ | 3.5 | 5.2417 | 18000 | 0.3602 | 3.6348 |
74
+ | 3.5244 | 5.5329 | 19000 | 0.3612 | 3.6237 |
75
+ | 3.5255 | 5.8242 | 20000 | 0.3623 | 3.6134 |
76
+ | 3.4384 | 6.1153 | 21000 | 0.3628 | 3.6150 |
77
+ | 3.4611 | 6.4065 | 22000 | 0.3634 | 3.6078 |
78
+ | 3.5017 | 6.6978 | 23000 | 0.3638 | 3.6003 |
79
+ | 3.4861 | 6.9890 | 24000 | 0.3650 | 3.5895 |
80
+ | 3.4213 | 7.2802 | 25000 | 0.3646 | 3.5996 |
81
+ | 3.4574 | 7.5714 | 26000 | 0.3649 | 3.5899 |
82
+ | 3.4488 | 7.8626 | 27000 | 0.3660 | 3.5817 |
83
+ | 3.3653 | 8.1538 | 28000 | 0.3656 | 3.5892 |
84
+ | 3.4157 | 8.4450 | 29000 | 0.3666 | 3.5810 |
85
+ | 3.4305 | 8.7362 | 30000 | 0.3670 | 3.5749 |
86
+ | 3.3188 | 9.0274 | 31000 | 0.3675 | 3.5799 |
87
+ | 3.367 | 9.3186 | 32000 | 0.3672 | 3.5794 |
88
+ | 3.4041 | 9.6098 | 33000 | 0.3676 | 3.5733 |
89
+ | 3.4055 | 9.9010 | 34000 | 0.3685 | 3.5640 |
90
+ | 3.3366 | 10.1922 | 35000 | 0.3681 | 3.5721 |
91
+ | 3.3638 | 10.4834 | 36000 | 0.3686 | 3.5680 |
92
+ | 3.3768 | 10.7747 | 37000 | 0.3688 | 3.5606 |
93
+ | 3.2775 | 11.0658 | 38000 | 0.3689 | 3.5706 |
94
+ | 3.331 | 11.3570 | 39000 | 0.3696 | 3.5672 |
95
+ | 3.3565 | 11.6483 | 40000 | 0.3697 | 3.5582 |
96
+ | 3.3535 | 11.9395 | 41000 | 0.3703 | 3.5503 |
97
+ | 3.3005 | 12.2306 | 42000 | 0.3701 | 3.5637 |
98
+ | 3.3365 | 12.5219 | 43000 | 0.3700 | 3.5570 |
99
+ | 3.3409 | 12.8131 | 44000 | 0.3706 | 3.5486 |
100
+ | 3.2575 | 13.1043 | 45000 | 0.3700 | 3.5608 |
101
+ | 3.2947 | 13.3955 | 46000 | 0.3707 | 3.5595 |
102
+ | 3.316 | 13.6867 | 47000 | 0.3710 | 3.5519 |
103
+ | 3.3273 | 13.9779 | 48000 | 0.3715 | 3.5409 |
104
+ | 3.2788 | 14.2691 | 49000 | 0.3708 | 3.5570 |
105
+ | 3.3039 | 14.5603 | 50000 | 0.3712 | 3.5518 |
106
+ | 3.3262 | 14.8515 | 51000 | 0.3718 | 3.5432 |
107
+ | 3.2326 | 15.1427 | 52000 | 0.3708 | 3.5588 |
108
+ | 3.2782 | 15.4339 | 53000 | 0.3716 | 3.5535 |
109
+ | 3.3013 | 15.7251 | 54000 | 0.3721 | 3.5432 |
110
+ | 3.2179 | 16.0163 | 55000 | 0.3715 | 3.5536 |
111
+ | 3.2511 | 16.3075 | 56000 | 0.3715 | 3.5548 |
112
+ | 3.2776 | 16.5988 | 57000 | 0.3724 | 3.5451 |
113
+ | 3.3012 | 16.8900 | 58000 | 0.3724 | 3.5374 |
114
+ | 3.2162 | 17.1811 | 59000 | 0.3720 | 3.5568 |
115
+ | 3.2594 | 17.4724 | 60000 | 0.3723 | 3.5469 |
116
+ | 3.262 | 17.7636 | 61000 | 0.3729 | 3.5397 |
117
+ | 3.1853 | 18.0547 | 62000 | 0.3727 | 3.5497 |
118
+ | 3.239 | 18.3460 | 63000 | 0.3726 | 3.5486 |
119
+ | 3.2576 | 18.6372 | 64000 | 0.3728 | 3.5436 |
120
+ | 3.2736 | 18.9284 | 65000 | 0.3734 | 3.5318 |
121
+ | 3.1832 | 19.2196 | 66000 | 0.3727 | 3.5538 |
122
+ | 3.2362 | 19.5108 | 67000 | 0.3728 | 3.5470 |
123
+ | 3.2708 | 19.8020 | 68000 | 0.3732 | 3.5365 |
124
+ | 3.1756 | 20.0932 | 69000 | 0.3731 | 3.5479 |
125
+ | 3.2256 | 20.3844 | 70000 | 0.3727 | 3.5475 |
126
+ | 3.2426 | 20.6756 | 71000 | 0.3733 | 3.5405 |
127
+ | 3.2544 | 20.9669 | 72000 | 0.3740 | 3.5309 |
128
+ | 3.1981 | 21.2580 | 73000 | 0.3729 | 3.5506 |
129
+ | 3.2153 | 21.5492 | 74000 | 0.3736 | 3.5443 |
130
+ | 3.2287 | 21.8405 | 75000 | 0.3742 | 3.5345 |
131
+ | 3.1662 | 22.1316 | 76000 | 0.3728 | 3.5559 |
132
+ | 3.1907 | 22.4229 | 77000 | 0.3737 | 3.5441 |
133
+ | 3.2248 | 22.7141 | 78000 | 0.3740 | 3.5370 |
134
+ | 3.1964 | 23.0052 | 79000 | 0.3738 | 3.5501 |
135
+ | 3.1882 | 23.2965 | 80000 | 0.3736 | 3.5463 |
136
+ | 3.1855 | 23.5877 | 81000 | 3.5508 | 0.3735 |
137
+ | 3.1989 | 23.8789 | 82000 | 3.5459 | 0.3738 |
138
+ | 3.1544 | 24.1704 | 83000 | 3.5584 | 0.3731 |
139
+ | 3.1822 | 24.4616 | 84000 | 3.5505 | 0.3738 |
140
+ | 3.2122 | 24.7528 | 85000 | 3.5393 | 0.3741 |
141
+ | 3.1227 | 25.0440 | 86000 | 3.5507 | 0.3736 |
142
+ | 3.1596 | 25.3352 | 87000 | 3.5509 | 0.3737 |
143
+ | 3.1876 | 25.6264 | 88000 | 3.5418 | 0.3743 |
144
+ | 3.1966 | 25.9176 | 89000 | 3.5306 | 0.3750 |
145
+ | 3.1457 | 26.2088 | 90000 | 3.5505 | 0.3740 |
146
+ | 3.1694 | 26.5000 | 91000 | 3.5417 | 0.3742 |
147
+ | 3.1857 | 26.7913 | 92000 | 3.5379 | 0.3748 |
148
+ | 3.1146 | 27.0824 | 93000 | 3.5564 | 0.3737 |
149
+ | 3.1478 | 27.3736 | 94000 | 3.5478 | 0.3741 |
150
+ | 3.1677 | 27.6649 | 95000 | 3.5378 | 0.3749 |
151
+ | 3.1917 | 27.9561 | 96000 | 3.5335 | 0.3749 |
152
+ | 3.1284 | 28.2472 | 97000 | 3.5506 | 0.3742 |
153
+ | 3.1637 | 28.5385 | 98000 | 3.5436 | 0.3749 |
154
+ | 3.1684 | 28.8297 | 99000 | 3.5397 | 0.3746 |
155
+ | 3.1085 | 29.1209 | 100000 | 3.5619 | 0.3739 |
156
+ | 3.1363 | 29.4121 | 101000 | 3.5465 | 0.3745 |
157
+ | 3.1494 | 29.7033 | 102000 | 3.5423 | 0.3749 |
158
+ | 3.1697 | 29.9945 | 103000 | 3.5358 | 0.3751 |
159
+ | 3.1182 | 30.2857 | 104000 | 3.5521 | 0.3742 |
160
+ | 3.123 | 30.5769 | 105000 | 3.5467 | 0.3749 |
161
+ | 3.1642 | 30.8681 | 106000 | 3.5389 | 0.3753 |
162
+ | 3.1068 | 31.1593 | 107000 | 3.5552 | 0.3747 |
163
+ | 3.1252 | 31.4505 | 108000 | 3.5485 | 0.3746 |
164
+ | 3.1427 | 31.7417 | 109000 | 3.5421 | 0.3751 |
165
+
166
+
167
+ ### Framework versions
168
+
169
+ - Transformers 4.55.2
170
+ - Pytorch 2.8.0+cu128
171
+ - Datasets 4.0.0
172
+ - Tokenizers 0.21.4
last_to_push_frequency_5039/all_results.json ADDED
@@ -0,0 +1,16 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 31.741743840643018,
3
+ "eval_accuracy": 0.36973110845315116,
4
+ "eval_loss": 3.5581958293914795,
5
+ "eval_runtime": 180.7942,
6
+ "eval_samples": 16646,
7
+ "eval_samples_per_second": 92.072,
8
+ "eval_steps_per_second": 5.758,
9
+ "perplexity": 35.09981393257323,
10
+ "total_flos": 2.27832542134272e+18,
11
+ "train_loss": 0.8390047937795656,
12
+ "train_runtime": 58351.8889,
13
+ "train_samples": 274700,
14
+ "train_samples_per_second": 235.382,
15
+ "train_steps_per_second": 2.942
16
+ }
last_to_push_frequency_5039/checkpoint-100000/config.json ADDED
@@ -0,0 +1,31 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation_function": "gelu_new",
3
+ "architectures": [
4
+ "GPT2LMHeadModel"
5
+ ],
6
+ "attn_pdrop": 0.1,
7
+ "bos_token_id": 50256,
8
+ "embd_pdrop": 0.1,
9
+ "eos_token_id": 50256,
10
+ "initializer_range": 0.02,
11
+ "layer_norm_epsilon": 1e-05,
12
+ "model_type": "gpt2",
13
+ "n_embd": 768,
14
+ "n_head": 12,
15
+ "n_inner": null,
16
+ "n_layer": 12,
17
+ "n_positions": 1024,
18
+ "reorder_and_upcast_attn": false,
19
+ "resid_pdrop": 0.1,
20
+ "scale_attn_by_inverse_layer_idx": false,
21
+ "scale_attn_weights": true,
22
+ "summary_activation": null,
23
+ "summary_first_dropout": 0.1,
24
+ "summary_proj_to_labels": true,
25
+ "summary_type": "cls_index",
26
+ "summary_use_proj": true,
27
+ "torch_dtype": "float32",
28
+ "transformers_version": "4.55.2",
29
+ "use_cache": true,
30
+ "vocab_size": 50257
31
+ }
last_to_push_frequency_5039/checkpoint-100000/generation_config.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 50256,
4
+ "eos_token_id": 50256,
5
+ "transformers_version": "4.55.2"
6
+ }
last_to_push_frequency_5039/checkpoint-100000/merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
last_to_push_frequency_5039/checkpoint-100000/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9398a9f11ce3271dcd4f94baf30ea6ef3242906dc660ed61093c5629eb6e5b3f
3
+ size 497774208
last_to_push_frequency_5039/checkpoint-100000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:1fd478d36dd37ecfe08fdd51730061204d75107eba3e4076c1efdeb6ec764cde
3
+ size 995644811
last_to_push_frequency_5039/checkpoint-100000/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0aa5a7fd577dc8625fb8ad34cffad95babe9fcf80c1a340f935ea216f5513f78
3
+ size 14709
last_to_push_frequency_5039/checkpoint-100000/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:73f19613fd6692afe114f08ff8861db08f6ca3202db5f5c26bd6b69e3bc0503b
3
+ size 1383
last_to_push_frequency_5039/checkpoint-100000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9da9e2053f96dfaa4f9443ff249d73b13ba7633481bc9e611e7cb72583f49ae5
3
+ size 1465
last_to_push_frequency_5039/checkpoint-100000/special_tokens_map.json ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<|endoftext|>",
3
+ "eos_token": "<|endoftext|>",
4
+ "unk_token": "<|endoftext|>"
5
+ }
last_to_push_frequency_5039/checkpoint-100000/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
last_to_push_frequency_5039/checkpoint-100000/tokenizer_config.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "added_tokens_decoder": {
4
+ "50256": {
5
+ "content": "<|endoftext|>",
6
+ "lstrip": false,
7
+ "normalized": true,
8
+ "rstrip": false,
9
+ "single_word": false,
10
+ "special": true
11
+ }
12
+ },
13
+ "bos_token": "<|endoftext|>",
14
+ "clean_up_tokenization_spaces": false,
15
+ "eos_token": "<|endoftext|>",
16
+ "extra_special_tokens": {},
17
+ "model_max_length": 1024,
18
+ "tokenizer_class": "GPT2Tokenizer",
19
+ "unk_token": "<|endoftext|>"
20
+ }
last_to_push_frequency_5039/checkpoint-100000/trainer_state.json ADDED
The diff for this file is too large to render. See raw diff
 
last_to_push_frequency_5039/checkpoint-100000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6d61e9bff56f18bbda54b485d0fbd51152411788b5de9a16b2286392b9b4dc5a
3
+ size 5969
last_to_push_frequency_5039/checkpoint-100000/vocab.json ADDED
The diff for this file is too large to render. See raw diff
 
last_to_push_frequency_5039/checkpoint-40000/config.json ADDED
@@ -0,0 +1,31 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation_function": "gelu_new",
3
+ "architectures": [
4
+ "GPT2LMHeadModel"
5
+ ],
6
+ "attn_pdrop": 0.1,
7
+ "bos_token_id": 50256,
8
+ "embd_pdrop": 0.1,
9
+ "eos_token_id": 50256,
10
+ "initializer_range": 0.02,
11
+ "layer_norm_epsilon": 1e-05,
12
+ "model_type": "gpt2",
13
+ "n_embd": 768,
14
+ "n_head": 12,
15
+ "n_inner": null,
16
+ "n_layer": 12,
17
+ "n_positions": 1024,
18
+ "reorder_and_upcast_attn": false,
19
+ "resid_pdrop": 0.1,
20
+ "scale_attn_by_inverse_layer_idx": false,
21
+ "scale_attn_weights": true,
22
+ "summary_activation": null,
23
+ "summary_first_dropout": 0.1,
24
+ "summary_proj_to_labels": true,
25
+ "summary_type": "cls_index",
26
+ "summary_use_proj": true,
27
+ "torch_dtype": "float32",
28
+ "transformers_version": "4.55.2",
29
+ "use_cache": true,
30
+ "vocab_size": 50257
31
+ }
last_to_push_frequency_5039/checkpoint-40000/generation_config.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 50256,
4
+ "eos_token_id": 50256,
5
+ "transformers_version": "4.55.2"
6
+ }
last_to_push_frequency_5039/checkpoint-40000/merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
last_to_push_frequency_5039/checkpoint-40000/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:343fee108dbf10adef885cfe3bd0c608cfd211b01758562e7fc8d8e06599c85e
3
+ size 497774208
last_to_push_frequency_5039/checkpoint-40000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:289fce982fb9641c1376f61973fde53541497e973f64fed715715165f9221fb0
3
+ size 995644811
last_to_push_frequency_5039/checkpoint-40000/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f0ab65b3b742c5834f9569ec000f3b495057154ab5add41e29f714b7bc941863
3
+ size 14709
last_to_push_frequency_5039/checkpoint-40000/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:be7c9bcd82835b41da69e38def32d6d379ffec3d7ba1284db366058fe070bedc
3
+ size 1383
last_to_push_frequency_5039/checkpoint-40000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:f2358aa5657b1e21f344557abc28c2385065d2ed2f82389cf3ff94ad7b8bd44f
3
+ size 1465
last_to_push_frequency_5039/checkpoint-40000/special_tokens_map.json ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<|endoftext|>",
3
+ "eos_token": "<|endoftext|>",
4
+ "unk_token": "<|endoftext|>"
5
+ }
last_to_push_frequency_5039/checkpoint-40000/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
last_to_push_frequency_5039/checkpoint-40000/tokenizer_config.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "added_tokens_decoder": {
4
+ "50256": {
5
+ "content": "<|endoftext|>",
6
+ "lstrip": false,
7
+ "normalized": true,
8
+ "rstrip": false,
9
+ "single_word": false,
10
+ "special": true
11
+ }
12
+ },
13
+ "bos_token": "<|endoftext|>",
14
+ "clean_up_tokenization_spaces": false,
15
+ "eos_token": "<|endoftext|>",
16
+ "extra_special_tokens": {},
17
+ "model_max_length": 1024,
18
+ "tokenizer_class": "GPT2Tokenizer",
19
+ "unk_token": "<|endoftext|>"
20
+ }
last_to_push_frequency_5039/checkpoint-40000/trainer_state.json ADDED
The diff for this file is too large to render. See raw diff
 
last_to_push_frequency_5039/checkpoint-40000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5fb688434a35edfe6202cfe9d613ed57f19275e6e0f1b8b850608df645a9b2fe
3
+ size 5969
last_to_push_frequency_5039/checkpoint-40000/vocab.json ADDED
The diff for this file is too large to render. See raw diff
 
last_to_push_frequency_5039/checkpoint-70000/config.json ADDED
@@ -0,0 +1,31 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation_function": "gelu_new",
3
+ "architectures": [
4
+ "GPT2LMHeadModel"
5
+ ],
6
+ "attn_pdrop": 0.1,
7
+ "bos_token_id": 50256,
8
+ "embd_pdrop": 0.1,
9
+ "eos_token_id": 50256,
10
+ "initializer_range": 0.02,
11
+ "layer_norm_epsilon": 1e-05,
12
+ "model_type": "gpt2",
13
+ "n_embd": 768,
14
+ "n_head": 12,
15
+ "n_inner": null,
16
+ "n_layer": 12,
17
+ "n_positions": 1024,
18
+ "reorder_and_upcast_attn": false,
19
+ "resid_pdrop": 0.1,
20
+ "scale_attn_by_inverse_layer_idx": false,
21
+ "scale_attn_weights": true,
22
+ "summary_activation": null,
23
+ "summary_first_dropout": 0.1,
24
+ "summary_proj_to_labels": true,
25
+ "summary_type": "cls_index",
26
+ "summary_use_proj": true,
27
+ "torch_dtype": "float32",
28
+ "transformers_version": "4.55.2",
29
+ "use_cache": true,
30
+ "vocab_size": 50257
31
+ }
last_to_push_frequency_5039/checkpoint-70000/generation_config.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 50256,
4
+ "eos_token_id": 50256,
5
+ "transformers_version": "4.55.2"
6
+ }
last_to_push_frequency_5039/checkpoint-70000/merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
last_to_push_frequency_5039/checkpoint-70000/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:73b85ef80681ef8ff9907145b1cd44e215f6e9e1f61a6891ff619c88f7333b72
3
+ size 497774208
last_to_push_frequency_5039/checkpoint-70000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:9e6888fc4c73669cd9394cb21e88feedebe7631863b2784ac9baadb976a09b0f
3
+ size 995644811
last_to_push_frequency_5039/checkpoint-70000/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:37f9b0a76bc863399b310fcd77f8d1f8877fbc1246a463278bbcb84ff07ed8e5
3
+ size 14709
last_to_push_frequency_5039/checkpoint-70000/scaler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:af767f77c6f0eb12cbfd8325ae57292eafa7caebaed3146fdabb7e366bb1c6b7
3
+ size 1383
last_to_push_frequency_5039/checkpoint-70000/scheduler.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:b5d9147f6c552b97d557708553469c436b903ee5755fef452f0f9cdd97c2e62e
3
+ size 1465
last_to_push_frequency_5039/checkpoint-70000/special_tokens_map.json ADDED
@@ -0,0 +1,5 @@
 
 
 
 
 
 
1
+ {
2
+ "bos_token": "<|endoftext|>",
3
+ "eos_token": "<|endoftext|>",
4
+ "unk_token": "<|endoftext|>"
5
+ }
last_to_push_frequency_5039/checkpoint-70000/tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
last_to_push_frequency_5039/checkpoint-70000/tokenizer_config.json ADDED
@@ -0,0 +1,20 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "added_tokens_decoder": {
4
+ "50256": {
5
+ "content": "<|endoftext|>",
6
+ "lstrip": false,
7
+ "normalized": true,
8
+ "rstrip": false,
9
+ "single_word": false,
10
+ "special": true
11
+ }
12
+ },
13
+ "bos_token": "<|endoftext|>",
14
+ "clean_up_tokenization_spaces": false,
15
+ "eos_token": "<|endoftext|>",
16
+ "extra_special_tokens": {},
17
+ "model_max_length": 1024,
18
+ "tokenizer_class": "GPT2Tokenizer",
19
+ "unk_token": "<|endoftext|>"
20
+ }
last_to_push_frequency_5039/checkpoint-70000/trainer_state.json ADDED
The diff for this file is too large to render. See raw diff
 
last_to_push_frequency_5039/checkpoint-70000/training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:5fb688434a35edfe6202cfe9d613ed57f19275e6e0f1b8b850608df645a9b2fe
3
+ size 5969
last_to_push_frequency_5039/checkpoint-70000/vocab.json ADDED
The diff for this file is too large to render. See raw diff
 
last_to_push_frequency_5039/checkpoint-80000/config.json ADDED
@@ -0,0 +1,31 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "activation_function": "gelu_new",
3
+ "architectures": [
4
+ "GPT2LMHeadModel"
5
+ ],
6
+ "attn_pdrop": 0.1,
7
+ "bos_token_id": 50256,
8
+ "embd_pdrop": 0.1,
9
+ "eos_token_id": 50256,
10
+ "initializer_range": 0.02,
11
+ "layer_norm_epsilon": 1e-05,
12
+ "model_type": "gpt2",
13
+ "n_embd": 768,
14
+ "n_head": 12,
15
+ "n_inner": null,
16
+ "n_layer": 12,
17
+ "n_positions": 1024,
18
+ "reorder_and_upcast_attn": false,
19
+ "resid_pdrop": 0.1,
20
+ "scale_attn_by_inverse_layer_idx": false,
21
+ "scale_attn_weights": true,
22
+ "summary_activation": null,
23
+ "summary_first_dropout": 0.1,
24
+ "summary_proj_to_labels": true,
25
+ "summary_type": "cls_index",
26
+ "summary_use_proj": true,
27
+ "torch_dtype": "float32",
28
+ "transformers_version": "4.55.2",
29
+ "use_cache": true,
30
+ "vocab_size": 50257
31
+ }
last_to_push_frequency_5039/checkpoint-80000/generation_config.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 50256,
4
+ "eos_token_id": 50256,
5
+ "transformers_version": "4.55.2"
6
+ }
last_to_push_frequency_5039/checkpoint-80000/merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
last_to_push_frequency_5039/checkpoint-80000/model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a14785ccbff802b675fff12a4e3ae2a4e23d3905b71afcddef7ea42bb46ac987
3
+ size 497774208
last_to_push_frequency_5039/checkpoint-80000/optimizer.pt ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:dec277fe1bb6c490d5278349a8c5f74e37f4abb45583c13b204803d0d6f96e3c
3
+ size 995644811
last_to_push_frequency_5039/checkpoint-80000/rng_state.pth ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:eec635d3e3f0e38707801205742298c9b60727032b3f987121e284a29a694f47
3
+ size 14709