Hugging Face's logo

craa
/
100M_high_100_495

Text Generation
Transformers
Safetensors
gpt2
Generated from Trainer
text-generation-inference
Model card Files
xet
 Community
craa commited on
Commit
3529d09
·
verified ·
1 Parent(s): 8f747c5

Training in progress, step 10000

Browse files
Files changed (12) hide show
  1. README.md +152 -0
  2. all_results.json +16 -0
  3. config.json +31 -0
  4. eval_results.json +10 -0
  5. generation_config.json +6 -0
  6. model.safetensors +3 -0
  7. special_tokens_map.json +1 -0
  8. tokenizer.json +0 -0
  9. tokenizer_config.json +0 -0
  10. train_results.json +9 -0
  11. trainer_state.json +0 -0
  12. training_args.bin +3 -0
README.md ADDED
@@ -0,0 +1,152 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ library_name: transformers
3
+ tags:
4
+ - generated_from_trainer
5
+ metrics:
6
+ - accuracy
7
+ model-index:
8
+ - name: 100M_low_500_495
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
+ # 100M_low_500_495
16
+
17
+ This model is a fine-tuned version of [](https://huggingface.co/) on an unknown dataset.
18
+ It achieves the following results on the evaluation set:
19
+ - Loss: 3.3084
20
+ - Accuracy: 0.3936
21
+
22
+ ## Model description
23
+
24
+ More information needed
25
+
26
+ ## Intended uses & limitations
27
+
28
+ More information needed
29
+
30
+ ## Training and evaluation data
31
+
32
+ More information needed
33
+
34
+ ## Training procedure
35
+
36
+ ### Training hyperparameters
37
+
38
+ The following hyperparameters were used during training:
39
+ - learning_rate: 0.0006
40
+ - train_batch_size: 32
41
+ - eval_batch_size: 16
42
+ - seed: 495
43
+ - optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.98) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
44
+ - lr_scheduler_type: linear
45
+ - lr_scheduler_warmup_steps: 100
46
+ - num_epochs: 10
47
+ - mixed_precision_training: Native AMP
48
+
49
+ ### Training results
50
+
51
+ | Training Loss | Epoch | Step | Validation Loss | Accuracy |
52
+ |:-------------:|:------:|:-----:|:---------------:|:--------:|
53
+ | 5.1138 | 0.1076 | 1000 | 5.0431 | 0.2252 |
54
+ | 4.6002 | 0.2153 | 2000 | 4.5449 | 0.2664 |
55
+ | 4.343 | 0.3229 | 3000 | 4.2586 | 0.2958 |
56
+ | 4.1858 | 0.4305 | 4000 | 4.1070 | 0.3103 |
57
+ | 4.0712 | 0.5382 | 5000 | 4.0071 | 0.3201 |
58
+ | 4.0074 | 0.6458 | 6000 | 3.9347 | 0.3266 |
59
+ | 3.9365 | 0.7534 | 7000 | 3.8809 | 0.3319 |
60
+ | 3.8792 | 0.8610 | 8000 | 3.8332 | 0.3362 |
61
+ | 3.8601 | 0.9687 | 9000 | 3.7926 | 0.3401 |
62
+ | 3.7662 | 1.0763 | 10000 | 3.7643 | 0.3427 |
63
+ | 3.7642 | 1.1839 | 11000 | 3.7349 | 0.3458 |
64
+ | 3.7326 | 1.2916 | 12000 | 3.7107 | 0.3487 |
65
+ | 3.7197 | 1.3992 | 13000 | 3.6893 | 0.3499 |
66
+ | 3.7238 | 1.5068 | 14000 | 3.6698 | 0.3522 |
67
+ | 3.695 | 1.6145 | 15000 | 3.6501 | 0.3541 |
68
+ | 3.6879 | 1.7221 | 16000 | 3.6334 | 0.3556 |
69
+ | 3.6613 | 1.8297 | 17000 | 3.6162 | 0.3577 |
70
+ | 3.6738 | 1.9374 | 18000 | 3.6035 | 0.3587 |
71
+ | 3.5709 | 2.0450 | 19000 | 3.5942 | 0.3600 |
72
+ | 3.576 | 2.1526 | 20000 | 3.5813 | 0.3613 |
73
+ | 3.5863 | 2.2603 | 21000 | 3.5722 | 0.3625 |
74
+ | 3.574 | 2.3679 | 22000 | 3.5611 | 0.3636 |
75
+ | 3.571 | 2.4755 | 23000 | 3.5508 | 0.3644 |
76
+ | 3.5674 | 2.5831 | 24000 | 3.5432 | 0.3658 |
77
+ | 3.5657 | 2.6908 | 25000 | 3.5351 | 0.3665 |
78
+ | 3.541 | 2.7984 | 26000 | 3.5254 | 0.3672 |
79
+ | 3.5422 | 2.9060 | 27000 | 3.5163 | 0.3684 |
80
+ | 3.4324 | 3.0137 | 28000 | 3.5107 | 0.3694 |
81
+ | 3.4761 | 3.1213 | 29000 | 3.5078 | 0.3698 |
82
+ | 3.4682 | 3.2289 | 30000 | 3.5009 | 0.3706 |
83
+ | 3.4774 | 3.3366 | 31000 | 3.4929 | 0.3708 |
84
+ | 3.4843 | 3.4442 | 32000 | 3.4871 | 0.3719 |
85
+ | 3.4749 | 3.5518 | 33000 | 3.4815 | 0.3721 |
86
+ | 3.4807 | 3.6595 | 34000 | 3.4726 | 0.3729 |
87
+ | 3.4845 | 3.7671 | 35000 | 3.4679 | 0.3736 |
88
+ | 3.4522 | 3.8747 | 36000 | 3.4624 | 0.3742 |
89
+ | 3.4692 | 3.9823 | 37000 | 3.4549 | 0.3751 |
90
+ | 3.3819 | 4.0900 | 38000 | 3.4570 | 0.3755 |
91
+ | 3.3902 | 4.1976 | 39000 | 3.4555 | 0.3761 |
92
+ | 3.3962 | 4.3052 | 40000 | 3.4490 | 0.3768 |
93
+ | 3.3919 | 4.4129 | 41000 | 3.4425 | 0.3770 |
94
+ | 3.4006 | 4.5205 | 42000 | 3.4389 | 0.3774 |
95
+ | 3.3908 | 4.6281 | 43000 | 3.4347 | 0.3773 |
96
+ | 3.4225 | 4.7358 | 44000 | 3.4284 | 0.3784 |
97
+ | 3.3928 | 4.8434 | 45000 | 3.4239 | 0.3791 |
98
+ | 3.3996 | 4.9510 | 46000 | 3.4188 | 0.3795 |
99
+ | 3.304 | 5.0587 | 47000 | 3.4196 | 0.3799 |
100
+ | 3.3463 | 5.1663 | 48000 | 3.4181 | 0.3800 |
101
+ | 3.3574 | 5.2739 | 49000 | 3.4146 | 0.3803 |
102
+ | 3.3465 | 5.3816 | 50000 | 3.4093 | 0.3808 |
103
+ | 3.3419 | 5.4892 | 51000 | 3.4084 | 0.3809 |
104
+ | 3.3549 | 5.5968 | 52000 | 3.4009 | 0.3819 |
105
+ | 3.3589 | 5.7044 | 53000 | 3.3958 | 0.3822 |
106
+ | 3.3318 | 5.8121 | 54000 | 3.3923 | 0.3826 |
107
+ | 3.3547 | 5.9197 | 55000 | 3.3893 | 0.3828 |
108
+ | 3.2872 | 6.0273 | 56000 | 3.3929 | 0.3834 |
109
+ | 3.2871 | 6.1350 | 57000 | 3.3935 | 0.3833 |
110
+ | 3.2995 | 6.2426 | 58000 | 3.3883 | 0.3840 |
111
+ | 3.2715 | 6.3502 | 59000 | 3.3835 | 0.3839 |
112
+ | 3.2983 | 6.4579 | 60000 | 3.3799 | 0.3847 |
113
+ | 3.3111 | 6.5655 | 61000 | 3.3768 | 0.3850 |
114
+ | 3.2886 | 6.6731 | 62000 | 3.3724 | 0.3856 |
115
+ | 3.2883 | 6.7808 | 63000 | 3.3686 | 0.3857 |
116
+ | 3.2944 | 6.8884 | 64000 | 3.3644 | 0.3859 |
117
+ | 3.2865 | 6.9960 | 65000 | 3.3587 | 0.3867 |
118
+ | 3.2471 | 7.1036 | 66000 | 3.3655 | 0.3863 |
119
+ | 3.2356 | 7.2113 | 67000 | 3.3650 | 0.3866 |
120
+ | 3.237 | 7.3189 | 68000 | 3.3602 | 0.3870 |
121
+ | 3.247 | 7.4265 | 69000 | 3.3561 | 0.3875 |
122
+ | 3.2441 | 7.5342 | 70000 | 3.3530 | 0.3879 |
123
+ | 3.2557 | 7.6418 | 71000 | 3.3489 | 0.3883 |
124
+ | 3.2571 | 7.7494 | 72000 | 3.3464 | 0.3886 |
125
+ | 3.246 | 7.8571 | 73000 | 3.3413 | 0.3891 |
126
+ | 3.2583 | 7.9647 | 74000 | 3.3392 | 0.3896 |
127
+ | 3.1864 | 8.0723 | 75000 | 3.3438 | 0.3893 |
128
+ | 3.1953 | 8.1800 | 76000 | 3.3420 | 0.3895 |
129
+ | 3.1767 | 8.2876 | 77000 | 3.3377 | 0.3902 |
130
+ | 3.2186 | 8.3952 | 78000 | 3.3354 | 0.3900 |
131
+ | 3.2188 | 8.5029 | 79000 | 3.3315 | 0.3906 |
132
+ | 3.203 | 8.6105 | 80000 | 3.3296 | 0.3911 |
133
+ | 3.1951 | 8.7181 | 81000 | 3.3254 | 0.3912 |
134
+ | 3.2144 | 8.8257 | 82000 | 3.3213 | 0.3916 |
135
+ | 3.2158 | 8.9334 | 83000 | 3.3179 | 0.3920 |
136
+ | 3.134 | 9.0410 | 84000 | 3.3217 | 0.3921 |
137
+ | 3.1403 | 9.1486 | 85000 | 3.3201 | 0.3922 |
138
+ | 3.1444 | 9.2563 | 86000 | 3.3181 | 0.3926 |
139
+ | 3.1524 | 9.3639 | 87000 | 3.3152 | 0.3928 |
140
+ | 3.1443 | 9.4715 | 88000 | 3.3128 | 0.3930 |
141
+ | 3.1415 | 9.5792 | 89000 | 3.3095 | 0.3933 |
142
+ | 3.1312 | 9.6868 | 90000 | 3.3084 | 0.3936 |
143
+ | 3.1554 | 9.7944 | 91000 | 3.3063 | 0.3938 |
144
+ | 3.1524 | 9.9021 | 92000 | 3.3044 | 0.3941 |
145
+
146
+
147
+ ### Framework versions
148
+
149
+ - Transformers 4.47.0.dev0
150
+ - Pytorch 2.5.0+cu124
151
+ - Datasets 3.0.2
152
+ - Tokenizers 0.20.1
all_results.json ADDED
@@ -0,0 +1,16 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 10.0,
3
+ "eval_accuracy": 0.3935781362574578,
4
+ "eval_loss": 3.308422088623047,
5
+ "eval_runtime": 188.968,
6
+ "eval_samples": 18011,
7
+ "eval_samples_per_second": 95.312,
8
+ "eval_steps_per_second": 5.959,
9
+ "perplexity": 27.341948244739367,
10
+ "total_flos": 7.7681598529536e+17,
11
+ "train_loss": 3.472916846254287,
12
+ "train_runtime": 81600.0966,
13
+ "train_samples": 297298,
14
+ "train_samples_per_second": 36.434,
15
+ "train_steps_per_second": 1.139
16
+ }
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": "4.47.0.dev0",
29
+ "use_cache": true,
30
+ "vocab_size": 52000
31
+ }
eval_results.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 10.0,
3
+ "eval_accuracy": 0.3935781362574578,
4
+ "eval_loss": 3.308422088623047,
5
+ "eval_runtime": 188.968,
6
+ "eval_samples": 18011,
7
+ "eval_samples_per_second": 95.312,
8
+ "eval_steps_per_second": 5.959,
9
+ "perplexity": 27.341948244739367
10
+ }
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": "4.47.0.dev0"
6
+ }
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:0f573415a31a6e5e51de63c04347402fa6cc4b78a86595a2c6839881eece7e5a
3
+ size 503128704
special_tokens_map.json ADDED
@@ -0,0 +1 @@
 
 
1
+ {}
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
The diff for this file is too large to render. See raw diff
 
train_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 10.0,
3
+ "total_flos": 7.7681598529536e+17,
4
+ "train_loss": 3.472916846254287,
5
+ "train_runtime": 81600.0966,
6
+ "train_samples": 297298,
7
+ "train_samples_per_second": 36.434,
8
+ "train_steps_per_second": 1.139
9
+ }
trainer_state.json ADDED
The diff for this file is too large to render. See raw diff
 
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:df0910feec2c6b0f91a8c136538b32a570b9c17dd79f334444184194ebab2b39
3
+ size 5304