Hugging Face's logo

wabu
/
5e-05_AmpGPT2

Text Generation
Transformers
Safetensors
gpt2
Generated from Trainer
Inference Endpoints
text-generation-inference
Model card Files Community
wabu commited on
Commit
408c5e7
•
1 Parent(s): ecdb479

Upload 14 files

Browse files
Files changed (14) hide show
  1. README.md +108 -0
  2. all_results.json +15 -0
  3. config.json +39 -0
  4. eval_results.json +10 -0
  5. generation_config.json +6 -0
  6. merges.txt +0 -0
  7. model.safetensors +3 -0
  8. special_tokens_map.json +23 -0
  9. tokenizer.json +0 -0
  10. tokenizer_config.json +19 -0
  11. train_results.json +8 -0
  12. trainer_state.json +564 -0
  13. training_args.bin +3 -0
  14. vocab.json +0 -0
README.md ADDED
@@ -0,0 +1,108 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ ---
2
+ base_model: nferruz/ProtGPT2
3
+ tags:
4
+ - generated_from_trainer
5
+ metrics:
6
+ - accuracy
7
+ model-index:
8
+ - name: 5e05_output_dir_clean_df_10-100_noX_100_50_epoch_cluster
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
+ # 5e05_output_dir_clean_df_10-100_noX_100_50_epoch_cluster
16
+
17
+ This model is a fine-tuned version of [nferruz/ProtGPT2](https://huggingface.co/nferruz/ProtGPT2) on an unknown dataset.
18
+ It achieves the following results on the evaluation set:
19
+ - Loss: 3.4181
20
+ - Accuracy: 0.5481
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: 5e-05
40
+ - train_batch_size: 32
41
+ - eval_batch_size: 32
42
+ - seed: 42
43
+ - optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
44
+ - lr_scheduler_type: linear
45
+ - num_epochs: 50.0
46
+
47
+ ### Training results
48
+
49
+ | Training Loss | Epoch | Step | Validation Loss | Accuracy |
50
+ |:-------------:|:-----:|:----:|:---------------:|:--------:|
51
+ | No log | 1.0 | 148 | 5.1495 | 0.2912 |
52
+ | No log | 2.0 | 296 | 4.6761 | 0.3383 |
53
+ | No log | 3.0 | 444 | 4.3827 | 0.3712 |
54
+ | 4.9816 | 4.0 | 592 | 4.1762 | 0.3960 |
55
+ | 4.9816 | 5.0 | 740 | 4.0295 | 0.4136 |
56
+ | 4.9816 | 6.0 | 888 | 3.9151 | 0.4276 |
57
+ | 3.9068 | 7.0 | 1036 | 3.8298 | 0.4400 |
58
+ | 3.9068 | 8.0 | 1184 | 3.7551 | 0.4500 |
59
+ | 3.9068 | 9.0 | 1332 | 3.6968 | 0.4586 |
60
+ | 3.9068 | 10.0 | 1480 | 3.6389 | 0.4668 |
61
+ | 3.3777 | 11.0 | 1628 | 3.5960 | 0.4743 |
62
+ | 3.3777 | 12.0 | 1776 | 3.5560 | 0.4803 |
63
+ | 3.3777 | 13.0 | 1924 | 3.5380 | 0.4868 |
64
+ | 3.0075 | 14.0 | 2072 | 3.4846 | 0.4914 |
65
+ | 3.0075 | 15.0 | 2220 | 3.4658 | 0.4969 |
66
+ | 3.0075 | 16.0 | 2368 | 3.4555 | 0.5010 |
67
+ | 2.7329 | 17.0 | 2516 | 3.4300 | 0.5053 |
68
+ | 2.7329 | 18.0 | 2664 | 3.4208 | 0.5080 |
69
+ | 2.7329 | 19.0 | 2812 | 3.4250 | 0.5121 |
70
+ | 2.7329 | 20.0 | 2960 | 3.3964 | 0.5147 |
71
+ | 2.5153 | 21.0 | 3108 | 3.3893 | 0.5181 |
72
+ | 2.5153 | 22.0 | 3256 | 3.3914 | 0.5204 |
73
+ | 2.5153 | 23.0 | 3404 | 3.3819 | 0.5229 |
74
+ | 2.336 | 24.0 | 3552 | 3.3786 | 0.5247 |
75
+ | 2.336 | 25.0 | 3700 | 3.3749 | 0.5267 |
76
+ | 2.336 | 26.0 | 3848 | 3.3774 | 0.5287 |
77
+ | 2.336 | 27.0 | 3996 | 3.3700 | 0.5303 |
78
+ | 2.1918 | 28.0 | 4144 | 3.3722 | 0.5321 |
79
+ | 2.1918 | 29.0 | 4292 | 3.3729 | 0.5340 |
80
+ | 2.1918 | 30.0 | 4440 | 3.3896 | 0.5350 |
81
+ | 2.0717 | 31.0 | 4588 | 3.3776 | 0.5367 |
82
+ | 2.0717 | 32.0 | 4736 | 3.3842 | 0.5385 |
83
+ | 2.0717 | 33.0 | 4884 | 3.3820 | 0.5399 |
84
+ | 1.9814 | 34.0 | 5032 | 3.3933 | 0.5404 |
85
+ | 1.9814 | 35.0 | 5180 | 3.3861 | 0.5411 |
86
+ | 1.9814 | 36.0 | 5328 | 3.3878 | 0.5425 |
87
+ | 1.9814 | 37.0 | 5476 | 3.3903 | 0.5431 |
88
+ | 1.9049 | 38.0 | 5624 | 3.3848 | 0.5440 |
89
+ | 1.9049 | 39.0 | 5772 | 3.3965 | 0.5447 |
90
+ | 1.9049 | 40.0 | 5920 | 3.4033 | 0.5453 |
91
+ | 1.8441 | 41.0 | 6068 | 3.4074 | 0.5456 |
92
+ | 1.8441 | 42.0 | 6216 | 3.4046 | 0.5462 |
93
+ | 1.8441 | 43.0 | 6364 | 3.4120 | 0.5467 |
94
+ | 1.804 | 44.0 | 6512 | 3.4044 | 0.5467 |
95
+ | 1.804 | 45.0 | 6660 | 3.4125 | 0.5472 |
96
+ | 1.804 | 46.0 | 6808 | 3.4115 | 0.5477 |
97
+ | 1.804 | 47.0 | 6956 | 3.4070 | 0.5477 |
98
+ | 1.7744 | 48.0 | 7104 | 3.4203 | 0.5478 |
99
+ | 1.7744 | 49.0 | 7252 | 3.4174 | 0.5479 |
100
+ | 1.7744 | 50.0 | 7400 | 3.4181 | 0.5481 |
101
+
102
+
103
+ ### Framework versions
104
+
105
+ - Transformers 4.38.0.dev0
106
+ - Pytorch 2.2.0+cu121
107
+ - Datasets 2.16.1
108
+ - Tokenizers 0.15.0
all_results.json ADDED
@@ -0,0 +1,15 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 50.0,
3
+ "eval_accuracy": 0.5480813265244403,
4
+ "eval_loss": 3.4181294441223145,
5
+ "eval_runtime": 19.7036,
6
+ "eval_samples": 1169,
7
+ "eval_samples_per_second": 59.329,
8
+ "eval_steps_per_second": 1.878,
9
+ "perplexity": 30.512286669649498,
10
+ "train_loss": 2.556491814690667,
11
+ "train_runtime": 9037.2432,
12
+ "train_samples": 4718,
13
+ "train_samples_per_second": 26.103,
14
+ "train_steps_per_second": 0.819
15
+ }
config.json ADDED
@@ -0,0 +1,39 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "_name_or_path": "nferruz/ProtGPT2",
3
+ "activation_function": "gelu_new",
4
+ "architectures": [
5
+ "GPT2LMHeadModel"
6
+ ],
7
+ "attn_pdrop": 0.1,
8
+ "bos_token_id": 0,
9
+ "embd_pdrop": 0.1,
10
+ "eos_token_id": 0,
11
+ "initializer_range": 0.02,
12
+ "layer_norm_epsilon": 1e-05,
13
+ "model_type": "gpt2",
14
+ "n_ctx": 1024,
15
+ "n_embd": 1280,
16
+ "n_head": 20,
17
+ "n_inner": null,
18
+ "n_layer": 36,
19
+ "n_positions": 1024,
20
+ "reorder_and_upcast_attn": false,
21
+ "resid_pdrop": 0.1,
22
+ "scale_attn_by_inverse_layer_idx": false,
23
+ "scale_attn_weights": true,
24
+ "summary_activation": null,
25
+ "summary_first_dropout": 0.1,
26
+ "summary_proj_to_labels": true,
27
+ "summary_type": "cls_index",
28
+ "summary_use_proj": true,
29
+ "task_specific_params": {
30
+ "text-generation": {
31
+ "do_sample": true,
32
+ "max_length": 50
33
+ }
34
+ },
35
+ "torch_dtype": "float32",
36
+ "transformers_version": "4.38.0.dev0",
37
+ "use_cache": true,
38
+ "vocab_size": 50257
39
+ }
eval_results.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 50.0,
3
+ "eval_accuracy": 0.5480813265244403,
4
+ "eval_loss": 3.4181294441223145,
5
+ "eval_runtime": 19.7036,
6
+ "eval_samples": 1169,
7
+ "eval_samples_per_second": 59.329,
8
+ "eval_steps_per_second": 1.878,
9
+ "perplexity": 30.512286669649498
10
+ }
generation_config.json ADDED
@@ -0,0 +1,6 @@
 
 
 
 
 
 
 
1
+ {
2
+ "_from_model_config": true,
3
+ "bos_token_id": 0,
4
+ "eos_token_id": 0,
5
+ "transformers_version": "4.38.0.dev0"
6
+ }
merges.txt ADDED
The diff for this file is too large to render. See raw diff
 
model.safetensors ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:51b58b0611318d4b5ec5399b7ff6ce0d9ec173d696231123c355610459eb9ab7
3
+ size 3096165928
special_tokens_map.json ADDED
@@ -0,0 +1,23 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "bos_token": {
3
+ "content": "<|endoftext|>",
4
+ "lstrip": false,
5
+ "normalized": true,
6
+ "rstrip": false,
7
+ "single_word": false
8
+ },
9
+ "eos_token": {
10
+ "content": "<|endoftext|>",
11
+ "lstrip": false,
12
+ "normalized": true,
13
+ "rstrip": false,
14
+ "single_word": false
15
+ },
16
+ "unk_token": {
17
+ "content": "<|endoftext|>",
18
+ "lstrip": false,
19
+ "normalized": true,
20
+ "rstrip": false,
21
+ "single_word": false
22
+ }
23
+ }
tokenizer.json ADDED
The diff for this file is too large to render. See raw diff
 
tokenizer_config.json ADDED
@@ -0,0 +1,19 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "add_prefix_space": false,
3
+ "added_tokens_decoder": {
4
+ "0": {
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": true,
15
+ "eos_token": "<|endoftext|>",
16
+ "model_max_length": 1000000000000000019884624838656,
17
+ "tokenizer_class": "GPT2Tokenizer",
18
+ "unk_token": "<|endoftext|>"
19
+ }
train_results.json ADDED
@@ -0,0 +1,8 @@
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 50.0,
3
+ "train_loss": 2.556491814690667,
4
+ "train_runtime": 9037.2432,
5
+ "train_samples": 4718,
6
+ "train_samples_per_second": 26.103,
7
+ "train_steps_per_second": 0.819
8
+ }
trainer_state.json ADDED
@@ -0,0 +1,564 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "best_metric": null,
3
+ "best_model_checkpoint": null,
4
+ "epoch": 50.0,
5
+ "eval_steps": 500,
6
+ "global_step": 7400,
7
+ "is_hyper_param_search": false,
8
+ "is_local_process_zero": true,
9
+ "is_world_process_zero": true,
10
+ "log_history": [
11
+ {
12
+ "epoch": 1.0,
13
+ "eval_accuracy": 0.2912011474885726,
14
+ "eval_loss": 5.149496078491211,
15
+ "eval_runtime": 19.7373,
16
+ "eval_samples_per_second": 59.228,
17
+ "eval_steps_per_second": 1.875,
18
+ "step": 148
19
+ },
20
+ {
21
+ "epoch": 2.0,
22
+ "eval_accuracy": 0.33825854783938614,
23
+ "eval_loss": 4.6761064529418945,
24
+ "eval_runtime": 19.7373,
25
+ "eval_samples_per_second": 59.228,
26
+ "eval_steps_per_second": 1.875,
27
+ "step": 296
28
+ },
29
+ {
30
+ "epoch": 3.0,
31
+ "eval_accuracy": 0.37122292212112573,
32
+ "eval_loss": 4.382660388946533,
33
+ "eval_runtime": 19.7317,
34
+ "eval_samples_per_second": 59.245,
35
+ "eval_steps_per_second": 1.875,
36
+ "step": 444
37
+ },
38
+ {
39
+ "epoch": 3.38,
40
+ "learning_rate": 4.662162162162162e-05,
41
+ "loss": 4.9816,
42
+ "step": 500
43
+ },
44
+ {
45
+ "epoch": 4.0,
46
+ "eval_accuracy": 0.3959613241050367,
47
+ "eval_loss": 4.176214694976807,
48
+ "eval_runtime": 19.7386,
49
+ "eval_samples_per_second": 59.224,
50
+ "eval_steps_per_second": 1.874,
51
+ "step": 592
52
+ },
53
+ {
54
+ "epoch": 5.0,
55
+ "eval_accuracy": 0.41364025196360527,
56
+ "eval_loss": 4.029497146606445,
57
+ "eval_runtime": 19.7378,
58
+ "eval_samples_per_second": 59.226,
59
+ "eval_steps_per_second": 1.875,
60
+ "step": 740
61
+ },
62
+ {
63
+ "epoch": 6.0,
64
+ "eval_accuracy": 0.4276382300334396,
65
+ "eval_loss": 3.915071725845337,
66
+ "eval_runtime": 19.7391,
67
+ "eval_samples_per_second": 59.223,
68
+ "eval_steps_per_second": 1.874,
69
+ "step": 888
70
+ },
71
+ {
72
+ "epoch": 6.76,
73
+ "learning_rate": 4.324324324324325e-05,
74
+ "loss": 3.9068,
75
+ "step": 1000
76
+ },
77
+ {
78
+ "epoch": 7.0,
79
+ "eval_accuracy": 0.43997718848018247,
80
+ "eval_loss": 3.829800844192505,
81
+ "eval_runtime": 19.73,
82
+ "eval_samples_per_second": 59.25,
83
+ "eval_steps_per_second": 1.875,
84
+ "step": 1036
85
+ },
86
+ {
87
+ "epoch": 8.0,
88
+ "eval_accuracy": 0.450026354217971,
89
+ "eval_loss": 3.7551472187042236,
90
+ "eval_runtime": 19.7353,
91
+ "eval_samples_per_second": 59.234,
92
+ "eval_steps_per_second": 1.875,
93
+ "step": 1184
94
+ },
95
+ {
96
+ "epoch": 9.0,
97
+ "eval_accuracy": 0.45856339269512925,
98
+ "eval_loss": 3.6967544555664062,
99
+ "eval_runtime": 19.735,
100
+ "eval_samples_per_second": 59.235,
101
+ "eval_steps_per_second": 1.875,
102
+ "step": 1332
103
+ },
104
+ {
105
+ "epoch": 10.0,
106
+ "eval_accuracy": 0.4668152871745686,
107
+ "eval_loss": 3.6389007568359375,
108
+ "eval_runtime": 19.7369,
109
+ "eval_samples_per_second": 59.229,
110
+ "eval_steps_per_second": 1.875,
111
+ "step": 1480
112
+ },
113
+ {
114
+ "epoch": 10.14,
115
+ "learning_rate": 3.986486486486487e-05,
116
+ "loss": 3.3777,
117
+ "step": 1500
118
+ },
119
+ {
120
+ "epoch": 11.0,
121
+ "eval_accuracy": 0.4742895162056839,
122
+ "eval_loss": 3.5959553718566895,
123
+ "eval_runtime": 19.7345,
124
+ "eval_samples_per_second": 59.236,
125
+ "eval_steps_per_second": 1.875,
126
+ "step": 1628
127
+ },
128
+ {
129
+ "epoch": 12.0,
130
+ "eval_accuracy": 0.4803034623393905,
131
+ "eval_loss": 3.5559630393981934,
132
+ "eval_runtime": 19.7359,
133
+ "eval_samples_per_second": 59.232,
134
+ "eval_steps_per_second": 1.875,
135
+ "step": 1776
136
+ },
137
+ {
138
+ "epoch": 13.0,
139
+ "eval_accuracy": 0.48679264846929515,
140
+ "eval_loss": 3.537980794906616,
141
+ "eval_runtime": 19.7353,
142
+ "eval_samples_per_second": 59.234,
143
+ "eval_steps_per_second": 1.875,
144
+ "step": 1924
145
+ },
146
+ {
147
+ "epoch": 13.51,
148
+ "learning_rate": 3.648648648648649e-05,
149
+ "loss": 3.0075,
150
+ "step": 2000
151
+ },
152
+ {
153
+ "epoch": 14.0,
154
+ "eval_accuracy": 0.49136359316000033,
155
+ "eval_loss": 3.484649181365967,
156
+ "eval_runtime": 20.0887,
157
+ "eval_samples_per_second": 58.192,
158
+ "eval_steps_per_second": 1.842,
159
+ "step": 2072
160
+ },
161
+ {
162
+ "epoch": 15.0,
163
+ "eval_accuracy": 0.49689365857030526,
164
+ "eval_loss": 3.4657890796661377,
165
+ "eval_runtime": 19.7322,
166
+ "eval_samples_per_second": 59.243,
167
+ "eval_steps_per_second": 1.875,
168
+ "step": 2220
169
+ },
170
+ {
171
+ "epoch": 16.0,
172
+ "eval_accuracy": 0.501041207628034,
173
+ "eval_loss": 3.4555246829986572,
174
+ "eval_runtime": 19.7332,
175
+ "eval_samples_per_second": 59.24,
176
+ "eval_steps_per_second": 1.875,
177
+ "step": 2368
178
+ },
179
+ {
180
+ "epoch": 16.89,
181
+ "learning_rate": 3.310810810810811e-05,
182
+ "loss": 2.7329,
183
+ "step": 2500
184
+ },
185
+ {
186
+ "epoch": 17.0,
187
+ "eval_accuracy": 0.5052838046850023,
188
+ "eval_loss": 3.4299707412719727,
189
+ "eval_runtime": 19.7326,
190
+ "eval_samples_per_second": 59.242,
191
+ "eval_steps_per_second": 1.875,
192
+ "step": 2516
193
+ },
194
+ {
195
+ "epoch": 18.0,
196
+ "eval_accuracy": 0.5079710708453223,
197
+ "eval_loss": 3.420835494995117,
198
+ "eval_runtime": 19.737,
199
+ "eval_samples_per_second": 59.229,
200
+ "eval_steps_per_second": 1.875,
201
+ "step": 2664
202
+ },
203
+ {
204
+ "epoch": 19.0,
205
+ "eval_accuracy": 0.5120667755398294,
206
+ "eval_loss": 3.425023078918457,
207
+ "eval_runtime": 19.7397,
208
+ "eval_samples_per_second": 59.221,
209
+ "eval_steps_per_second": 1.874,
210
+ "step": 2812
211
+ },
212
+ {
213
+ "epoch": 20.0,
214
+ "eval_accuracy": 0.5146849158825206,
215
+ "eval_loss": 3.3964016437530518,
216
+ "eval_runtime": 19.7401,
217
+ "eval_samples_per_second": 59.22,
218
+ "eval_steps_per_second": 1.874,
219
+ "step": 2960
220
+ },
221
+ {
222
+ "epoch": 20.27,
223
+ "learning_rate": 2.9729729729729733e-05,
224
+ "loss": 2.5153,
225
+ "step": 3000
226
+ },
227
+ {
228
+ "epoch": 21.0,
229
+ "eval_accuracy": 0.5181498474911649,
230
+ "eval_loss": 3.3892605304718018,
231
+ "eval_runtime": 19.7415,
232
+ "eval_samples_per_second": 59.215,
233
+ "eval_steps_per_second": 1.874,
234
+ "step": 3108
235
+ },
236
+ {
237
+ "epoch": 22.0,
238
+ "eval_accuracy": 0.5204396402001192,
239
+ "eval_loss": 3.3914339542388916,
240
+ "eval_runtime": 19.7345,
241
+ "eval_samples_per_second": 59.236,
242
+ "eval_steps_per_second": 1.875,
243
+ "step": 3256
244
+ },
245
+ {
246
+ "epoch": 23.0,
247
+ "eval_accuracy": 0.5228590438171277,
248
+ "eval_loss": 3.3818881511688232,
249
+ "eval_runtime": 19.7369,
250
+ "eval_samples_per_second": 59.229,
251
+ "eval_steps_per_second": 1.875,
252
+ "step": 3404
253
+ },
254
+ {
255
+ "epoch": 23.65,
256
+ "learning_rate": 2.635135135135135e-05,
257
+ "loss": 2.336,
258
+ "step": 3500
259
+ },
260
+ {
261
+ "epoch": 24.0,
262
+ "eval_accuracy": 0.5246649558026804,
263
+ "eval_loss": 3.3786017894744873,
264
+ "eval_runtime": 19.7359,
265
+ "eval_samples_per_second": 59.232,
266
+ "eval_steps_per_second": 1.875,
267
+ "step": 3552
268
+ },
269
+ {
270
+ "epoch": 25.0,
271
+ "eval_accuracy": 0.5267128081499339,
272
+ "eval_loss": 3.374890089035034,
273
+ "eval_runtime": 19.7329,
274
+ "eval_samples_per_second": 59.241,
275
+ "eval_steps_per_second": 1.875,
276
+ "step": 3700
277
+ },
278
+ {
279
+ "epoch": 26.0,
280
+ "eval_accuracy": 0.5287001754067623,
281
+ "eval_loss": 3.377443552017212,
282
+ "eval_runtime": 19.7328,
283
+ "eval_samples_per_second": 59.241,
284
+ "eval_steps_per_second": 1.875,
285
+ "step": 3848
286
+ },
287
+ {
288
+ "epoch": 27.0,
289
+ "eval_accuracy": 0.5302555063034105,
290
+ "eval_loss": 3.3700437545776367,
291
+ "eval_runtime": 19.7375,
292
+ "eval_samples_per_second": 59.227,
293
+ "eval_steps_per_second": 1.875,
294
+ "step": 3996
295
+ },
296
+ {
297
+ "epoch": 27.03,
298
+ "learning_rate": 2.2972972972972976e-05,
299
+ "loss": 2.1918,
300
+ "step": 4000
301
+ },
302
+ {
303
+ "epoch": 28.0,
304
+ "eval_accuracy": 0.5320614182889633,
305
+ "eval_loss": 3.372175693511963,
306
+ "eval_runtime": 19.7333,
307
+ "eval_samples_per_second": 59.24,
308
+ "eval_steps_per_second": 1.875,
309
+ "step": 4144
310
+ },
311
+ {
312
+ "epoch": 29.0,
313
+ "eval_accuracy": 0.5339623782737555,
314
+ "eval_loss": 3.3728740215301514,
315
+ "eval_runtime": 19.7294,
316
+ "eval_samples_per_second": 59.252,
317
+ "eval_steps_per_second": 1.875,
318
+ "step": 4292
319
+ },
320
+ {
321
+ "epoch": 30.0,
322
+ "eval_accuracy": 0.5349819840837805,
323
+ "eval_loss": 3.3896090984344482,
324
+ "eval_runtime": 19.7317,
325
+ "eval_samples_per_second": 59.245,
326
+ "eval_steps_per_second": 1.875,
327
+ "step": 4440
328
+ },
329
+ {
330
+ "epoch": 30.41,
331
+ "learning_rate": 1.9594594594594595e-05,
332
+ "loss": 2.0717,
333
+ "step": 4500
334
+ },
335
+ {
336
+ "epoch": 31.0,
337
+ "eval_accuracy": 0.5367446924333152,
338
+ "eval_loss": 3.37764048576355,
339
+ "eval_runtime": 19.7293,
340
+ "eval_samples_per_second": 59.252,
341
+ "eval_steps_per_second": 1.875,
342
+ "step": 4588
343
+ },
344
+ {
345
+ "epoch": 32.0,
346
+ "eval_accuracy": 0.5384901193284427,
347
+ "eval_loss": 3.384190797805786,
348
+ "eval_runtime": 19.7297,
349
+ "eval_samples_per_second": 59.251,
350
+ "eval_steps_per_second": 1.875,
351
+ "step": 4736
352
+ },
353
+ {
354
+ "epoch": 33.0,
355
+ "eval_accuracy": 0.5399244800442405,
356
+ "eval_loss": 3.38201642036438,
357
+ "eval_runtime": 19.7316,
358
+ "eval_samples_per_second": 59.245,
359
+ "eval_steps_per_second": 1.875,
360
+ "step": 4884
361
+ },
362
+ {
363
+ "epoch": 33.78,
364
+ "learning_rate": 1.6216216216216218e-05,
365
+ "loss": 1.9814,
366
+ "step": 5000
367
+ },
368
+ {
369
+ "epoch": 34.0,
370
+ "eval_accuracy": 0.540391079313235,
371
+ "eval_loss": 3.393296241760254,
372
+ "eval_runtime": 19.7393,
373
+ "eval_samples_per_second": 59.222,
374
+ "eval_steps_per_second": 1.874,
375
+ "step": 5032
376
+ },
377
+ {
378
+ "epoch": 35.0,
379
+ "eval_accuracy": 0.5410564153079123,
380
+ "eval_loss": 3.386063814163208,
381
+ "eval_runtime": 19.7351,
382
+ "eval_samples_per_second": 59.234,
383
+ "eval_steps_per_second": 1.875,
384
+ "step": 5180
385
+ },
386
+ {
387
+ "epoch": 36.0,
388
+ "eval_accuracy": 0.5425253389325245,
389
+ "eval_loss": 3.3877809047698975,
390
+ "eval_runtime": 19.7289,
391
+ "eval_samples_per_second": 59.253,
392
+ "eval_steps_per_second": 1.875,
393
+ "step": 5328
394
+ },
395
+ {
396
+ "epoch": 37.0,
397
+ "eval_accuracy": 0.5431129083823695,
398
+ "eval_loss": 3.3903427124023438,
399
+ "eval_runtime": 19.7301,
400
+ "eval_samples_per_second": 59.25,
401
+ "eval_steps_per_second": 1.875,
402
+ "step": 5476
403
+ },
404
+ {
405
+ "epoch": 37.16,
406
+ "learning_rate": 1.2837837837837838e-05,
407
+ "loss": 1.9049,
408
+ "step": 5500
409
+ },
410
+ {
411
+ "epoch": 38.0,
412
+ "eval_accuracy": 0.5439856218299332,
413
+ "eval_loss": 3.3847618103027344,
414
+ "eval_runtime": 19.731,
415
+ "eval_samples_per_second": 59.247,
416
+ "eval_steps_per_second": 1.875,
417
+ "step": 5624
418
+ },
419
+ {
420
+ "epoch": 39.0,
421
+ "eval_accuracy": 0.5446855207334249,
422
+ "eval_loss": 3.396543502807617,
423
+ "eval_runtime": 19.7337,
424
+ "eval_samples_per_second": 59.239,
425
+ "eval_steps_per_second": 1.875,
426
+ "step": 5772
427
+ },
428
+ {
429
+ "epoch": 40.0,
430
+ "eval_accuracy": 0.5453249345464914,
431
+ "eval_loss": 3.403301477432251,
432
+ "eval_runtime": 19.7292,
433
+ "eval_samples_per_second": 59.252,
434
+ "eval_steps_per_second": 1.875,
435
+ "step": 5920
436
+ },
437
+ {
438
+ "epoch": 40.54,
439
+ "learning_rate": 9.45945945945946e-06,
440
+ "loss": 1.8441,
441
+ "step": 6000
442
+ },
443
+ {
444
+ "epoch": 41.0,
445
+ "eval_accuracy": 0.5456187192714139,
446
+ "eval_loss": 3.4074459075927734,
447
+ "eval_runtime": 19.7354,
448
+ "eval_samples_per_second": 59.234,
449
+ "eval_steps_per_second": 1.875,
450
+ "step": 6068
451
+ },
452
+ {
453
+ "epoch": 42.0,
454
+ "eval_accuracy": 0.5462322109028696,
455
+ "eval_loss": 3.4045963287353516,
456
+ "eval_runtime": 19.7291,
457
+ "eval_samples_per_second": 59.253,
458
+ "eval_steps_per_second": 1.875,
459
+ "step": 6216
460
+ },
461
+ {
462
+ "epoch": 43.0,
463
+ "eval_accuracy": 0.5467333730806785,
464
+ "eval_loss": 3.4119956493377686,
465
+ "eval_runtime": 19.7304,
466
+ "eval_samples_per_second": 59.249,
467
+ "eval_steps_per_second": 1.875,
468
+ "step": 6364
469
+ },
470
+ {
471
+ "epoch": 43.92,
472
+ "learning_rate": 6.081081081081082e-06,
473
+ "loss": 1.804,
474
+ "step": 6500
475
+ },
476
+ {
477
+ "epoch": 44.0,
478
+ "eval_accuracy": 0.5466901694446604,
479
+ "eval_loss": 3.4043545722961426,
480
+ "eval_runtime": 19.7301,
481
+ "eval_samples_per_second": 59.25,
482
+ "eval_steps_per_second": 1.875,
483
+ "step": 6512
484
+ },
485
+ {
486
+ "epoch": 45.0,
487
+ "eval_accuracy": 0.5471913316224694,
488
+ "eval_loss": 3.4124903678894043,
489
+ "eval_runtime": 19.7367,
490
+ "eval_samples_per_second": 59.23,
491
+ "eval_steps_per_second": 1.875,
492
+ "step": 6660
493
+ },
494
+ {
495
+ "epoch": 46.0,
496
+ "eval_accuracy": 0.5476665716186674,
497
+ "eval_loss": 3.4115209579467773,
498
+ "eval_runtime": 19.7299,
499
+ "eval_samples_per_second": 59.25,
500
+ "eval_steps_per_second": 1.875,
501
+ "step": 6808
502
+ },
503
+ {
504
+ "epoch": 47.0,
505
+ "eval_accuracy": 0.547675212345871,
506
+ "eval_loss": 3.4069876670837402,
507
+ "eval_runtime": 19.7339,
508
+ "eval_samples_per_second": 59.238,
509
+ "eval_steps_per_second": 1.875,
510
+ "step": 6956
511
+ },
512
+ {
513
+ "epoch": 47.3,
514
+ "learning_rate": 2.702702702702703e-06,
515
+ "loss": 1.7744,
516
+ "step": 7000
517
+ },
518
+ {
519
+ "epoch": 48.0,
520
+ "eval_accuracy": 0.5477875417995178,
521
+ "eval_loss": 3.420271396636963,
522
+ "eval_runtime": 19.7291,
523
+ "eval_samples_per_second": 59.253,
524
+ "eval_steps_per_second": 1.875,
525
+ "step": 7104
526
+ },
527
+ {
528
+ "epoch": 49.0,
529
+ "eval_accuracy": 0.5479430748891827,
530
+ "eval_loss": 3.4173614978790283,
531
+ "eval_runtime": 19.7304,
532
+ "eval_samples_per_second": 59.249,
533
+ "eval_steps_per_second": 1.875,
534
+ "step": 7252
535
+ },
536
+ {
537
+ "epoch": 50.0,
538
+ "eval_accuracy": 0.5480813265244403,
539
+ "eval_loss": 3.4181294441223145,
540
+ "eval_runtime": 19.7276,
541
+ "eval_samples_per_second": 59.257,
542
+ "eval_steps_per_second": 1.876,
543
+ "step": 7400
544
+ },
545
+ {
546
+ "epoch": 50.0,
547
+ "step": 7400,
548
+ "total_flos": 1.00265577216e+17,
549
+ "train_loss": 2.556491814690667,
550
+ "train_runtime": 9037.2432,
551
+ "train_samples_per_second": 26.103,
552
+ "train_steps_per_second": 0.819
553
+ }
554
+ ],
555
+ "logging_steps": 500,
556
+ "max_steps": 7400,
557
+ "num_input_tokens_seen": 0,
558
+ "num_train_epochs": 50,
559
+ "save_steps": 500,
560
+ "total_flos": 1.00265577216e+17,
561
+ "train_batch_size": 32,
562
+ "trial_name": null,
563
+ "trial_params": null
564
+ }
training_args.bin ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:3206938b60ad9765d2dafcb643c6cdb4aa635c631499599b3b946b1ab274f820
3
+ size 4856
vocab.json ADDED
The diff for this file is too large to render. See raw diff