Hugging Face's logo

craa
/
100M__634

Text Generation
Transformers
Safetensors
gpt2
Generated from Trainer
text-generation-inference
Model card Files
xet
Community
craa commited on
Commit
d942429
·
verified ·
1 Parent(s): f8f5f58

Model save

Browse files
Files changed (6) hide show
  1. README.md +100 -97
  2. all_results.json +16 -0
  3. eval_results.json +10 -0
  4. train_results.json +9 -0
  5. trainer_state.json +0 -0
  6. training_args.bin +1 -1
README.md CHANGED
@@ -16,8 +16,8 @@ should probably proofread and complete it, then remove this comment. -->
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: 3.3004
20
- - Accuracy: 0.3946
21
 
22
  ## Model description
23
 
@@ -43,105 +43,108 @@ The following hyperparameters were used during training:
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
- | 5.0778 | 0.1078 | 1000 | 5.0244 | 0.2270 |
54
- | 4.5738 | 0.2156 | 2000 | 4.5008 | 0.2719 |
55
- | 4.309 | 0.3235 | 3000 | 4.2284 | 0.2995 |
56
- | 4.1524 | 0.4313 | 4000 | 4.0876 | 0.3122 |
57
- | 4.0714 | 0.5391 | 5000 | 3.9901 | 0.3216 |
58
- | 3.9631 | 0.6469 | 6000 | 3.9180 | 0.3281 |
59
- | 3.9285 | 0.7547 | 7000 | 3.8616 | 0.3334 |
60
- | 3.8676 | 0.8625 | 8000 | 3.8167 | 0.3375 |
61
- | 3.8648 | 0.9704 | 9000 | 3.7798 | 0.3409 |
62
- | 3.7661 | 1.0782 | 10000 | 3.7474 | 0.3448 |
63
- | 3.7415 | 1.1860 | 11000 | 3.7234 | 0.3470 |
64
- | 3.7241 | 1.2938 | 12000 | 3.7014 | 0.3486 |
65
- | 3.715 | 1.4016 | 13000 | 3.6799 | 0.3512 |
66
- | 3.6982 | 1.5094 | 14000 | 3.6556 | 0.3534 |
67
- | 3.6829 | 1.6173 | 15000 | 3.6376 | 0.3553 |
68
- | 3.6554 | 1.7251 | 16000 | 3.6238 | 0.3569 |
69
- | 3.6345 | 1.8329 | 17000 | 3.6062 | 0.3583 |
70
- | 3.6469 | 1.9407 | 18000 | 3.5914 | 0.3599 |
71
- | 3.5639 | 2.0485 | 19000 | 3.5848 | 0.3609 |
72
- | 3.5606 | 2.1563 | 20000 | 3.5729 | 0.3622 |
73
- | 3.5497 | 2.2642 | 21000 | 3.5615 | 0.3636 |
74
- | 3.5624 | 2.3720 | 22000 | 3.5510 | 0.3644 |
75
- | 3.5459 | 2.4798 | 23000 | 3.5396 | 0.3655 |
76
- | 3.5482 | 2.5876 | 24000 | 3.5318 | 0.3666 |
77
- | 3.5542 | 2.6954 | 25000 | 3.5221 | 0.3674 |
78
- | 3.5342 | 2.8032 | 26000 | 3.5146 | 0.3682 |
79
- | 3.5381 | 2.9111 | 27000 | 3.5050 | 0.3693 |
80
- | 3.4278 | 3.0189 | 28000 | 3.5020 | 0.3700 |
81
- | 3.4513 | 3.1267 | 29000 | 3.4942 | 0.3708 |
82
- | 3.4627 | 3.2345 | 30000 | 3.4898 | 0.3713 |
83
- | 3.4463 | 3.3423 | 31000 | 3.4832 | 0.3720 |
84
- | 3.461 | 3.4501 | 32000 | 3.4776 | 0.3727 |
85
- | 3.4454 | 3.5580 | 33000 | 3.4710 | 0.3733 |
86
- | 3.4627 | 3.6658 | 34000 | 3.4632 | 0.3741 |
87
- | 3.444 | 3.7736 | 35000 | 3.4588 | 0.3743 |
88
- | 3.4513 | 3.8814 | 36000 | 3.4522 | 0.3752 |
89
- | 3.4411 | 3.9892 | 37000 | 3.4460 | 0.3758 |
90
- | 3.3781 | 4.0970 | 38000 | 3.4480 | 0.3764 |
91
- | 3.3767 | 4.2049 | 39000 | 3.4423 | 0.3767 |
92
- | 3.3947 | 4.3127 | 40000 | 3.4385 | 0.3773 |
93
- | 3.389 | 4.4205 | 41000 | 3.4339 | 0.3776 |
94
- | 3.3849 | 4.5283 | 42000 | 3.4289 | 0.3786 |
95
- | 3.4041 | 4.6361 | 43000 | 3.4244 | 0.3791 |
96
- | 3.3693 | 4.7439 | 44000 | 3.4180 | 0.3796 |
97
- | 3.3953 | 4.8518 | 45000 | 3.4133 | 0.3799 |
98
- | 3.3841 | 4.9596 | 46000 | 3.4098 | 0.3805 |
99
- | 3.3078 | 5.0674 | 47000 | 3.4134 | 0.3803 |
100
- | 3.315 | 5.1752 | 48000 | 3.4103 | 0.3810 |
101
- | 3.3108 | 5.2830 | 49000 | 3.4087 | 0.3809 |
102
- | 3.3356 | 5.3908 | 50000 | 3.4030 | 0.3818 |
103
- | 3.3193 | 5.4987 | 51000 | 3.3966 | 0.3824 |
104
- | 3.3407 | 5.6065 | 52000 | 3.3946 | 0.3826 |
105
- | 3.348 | 5.7143 | 53000 | 3.3898 | 0.3831 |
106
- | 3.3288 | 5.8221 | 54000 | 3.3838 | 0.3834 |
107
- | 3.3283 | 5.9299 | 55000 | 3.3800 | 0.3838 |
108
- | 3.2613 | 6.0377 | 56000 | 3.3837 | 0.3840 |
109
- | 3.2509 | 6.1456 | 57000 | 3.3834 | 0.3840 |
110
- | 3.2539 | 6.2534 | 58000 | 3.3802 | 0.3846 |
111
- | 3.2851 | 6.3612 | 59000 | 3.3774 | 0.3851 |
112
- | 3.2878 | 6.4690 | 60000 | 3.3725 | 0.3853 |
113
- | 3.2817 | 6.5768 | 61000 | 3.3693 | 0.3855 |
114
- | 3.2722 | 6.6846 | 62000 | 3.3639 | 0.3863 |
115
- | 3.2773 | 6.7925 | 63000 | 3.3609 | 0.3867 |
116
- | 3.2948 | 6.9003 | 64000 | 3.3567 | 0.3872 |
117
- | 3.1985 | 7.0081 | 65000 | 3.3584 | 0.3872 |
118
- | 3.2183 | 7.1159 | 66000 | 3.3596 | 0.3874 |
119
- | 3.2204 | 7.2237 | 67000 | 3.3558 | 0.3878 |
120
- | 3.2237 | 7.3315 | 68000 | 3.3549 | 0.3881 |
121
- | 3.2464 | 7.4394 | 69000 | 3.3498 | 0.3881 |
122
- | 3.2299 | 7.5472 | 70000 | 3.3457 | 0.3887 |
123
- | 3.2273 | 7.6550 | 71000 | 3.3421 | 0.3889 |
124
- | 3.2389 | 7.7628 | 72000 | 3.3363 | 0.3894 |
125
- | 3.2224 | 7.8706 | 73000 | 3.3338 | 0.3899 |
126
- | 3.2305 | 7.9784 | 74000 | 3.3320 | 0.3902 |
127
- | 3.1599 | 8.0863 | 75000 | 3.3381 | 0.3901 |
128
- | 3.1664 | 8.1941 | 76000 | 3.3350 | 0.3903 |
129
- | 3.1774 | 8.3019 | 77000 | 3.3318 | 0.3903 |
130
- | 3.1677 | 8.4097 | 78000 | 3.3281 | 0.3909 |
131
- | 3.1821 | 8.5175 | 79000 | 3.3258 | 0.3913 |
132
- | 3.1805 | 8.6253 | 80000 | 3.3235 | 0.3917 |
133
- | 3.1932 | 8.7332 | 81000 | 3.3191 | 0.3920 |
134
- | 3.1771 | 8.8410 | 82000 | 3.3148 | 0.3924 |
135
- | 3.1927 | 8.9488 | 83000 | 3.3131 | 0.3927 |
136
- | 3.1235 | 9.0566 | 84000 | 3.3163 | 0.3926 |
137
- | 3.1221 | 9.1644 | 85000 | 3.3138 | 0.3929 |
138
- | 3.1312 | 9.2722 | 86000 | 3.3123 | 0.3931 |
139
- | 3.1204 | 9.3801 | 87000 | 3.3111 | 0.3934 |
140
- | 3.1289 | 9.4879 | 88000 | 3.3072 | 0.3939 |
141
- | 3.1452 | 9.5957 | 89000 | 3.3056 | 0.3940 |
142
- | 3.1301 | 9.7035 | 90000 | 3.3030 | 0.3943 |
143
- | 3.1314 | 9.8113 | 91000 | 3.3021 | 0.3943 |
144
- | 3.1061 | 9.9191 | 92000 | 3.3004 | 0.3946 |
 
 
 
145
 
146
 
147
  ### Framework versions
 
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: 3.4624
20
+ - Accuracy: 0.3763
21
 
22
  ## Model description
23
 
 
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: 50
47
  - mixed_precision_training: Native AMP
48
 
49
  ### Training results
50
 
51
+ | Training Loss | Epoch | Step | Accuracy | Validation Loss |
52
+ |:-------------:|:-------:|:-----:|:--------:|:---------------:|
53
+ | 5.0778 | 0.1078 | 1000 | 0.2270 | 5.0244 |
54
+ | 4.5738 | 0.2156 | 2000 | 0.2719 | 4.5008 |
55
+ | 4.309 | 0.3235 | 3000 | 0.2995 | 4.2284 |
56
+ | 4.1524 | 0.4313 | 4000 | 0.3122 | 4.0876 |
57
+ | 4.0714 | 0.5391 | 5000 | 0.3216 | 3.9901 |
58
+ | 3.9631 | 0.6469 | 6000 | 0.3281 | 3.9180 |
59
+ | 3.9285 | 0.7547 | 7000 | 0.3334 | 3.8616 |
60
+ | 3.8676 | 0.8625 | 8000 | 0.3375 | 3.8167 |
61
+ | 3.8648 | 0.9704 | 9000 | 0.3409 | 3.7798 |
62
+ | 3.7661 | 1.0782 | 10000 | 0.3448 | 3.7474 |
63
+ | 3.7415 | 1.1860 | 11000 | 0.3470 | 3.7234 |
64
+ | 3.7241 | 1.2938 | 12000 | 0.3486 | 3.7014 |
65
+ | 3.715 | 1.4016 | 13000 | 0.3512 | 3.6799 |
66
+ | 3.6982 | 1.5094 | 14000 | 0.3534 | 3.6556 |
67
+ | 3.6829 | 1.6173 | 15000 | 0.3553 | 3.6376 |
68
+ | 3.6554 | 1.7251 | 16000 | 0.3569 | 3.6238 |
69
+ | 3.6345 | 1.8329 | 17000 | 0.3583 | 3.6062 |
70
+ | 3.6469 | 1.9407 | 18000 | 0.3599 | 3.5914 |
71
+ | 3.5639 | 2.0485 | 19000 | 0.3609 | 3.5848 |
72
+ | 3.5606 | 2.1563 | 20000 | 0.3622 | 3.5729 |
73
+ | 3.5497 | 2.2642 | 21000 | 0.3636 | 3.5615 |
74
+ | 3.5624 | 2.3720 | 22000 | 0.3644 | 3.5510 |
75
+ | 3.5459 | 2.4798 | 23000 | 0.3655 | 3.5396 |
76
+ | 3.5482 | 2.5876 | 24000 | 0.3666 | 3.5318 |
77
+ | 3.5542 | 2.6954 | 25000 | 0.3674 | 3.5221 |
78
+ | 3.5342 | 2.8032 | 26000 | 0.3682 | 3.5146 |
79
+ | 3.5381 | 2.9111 | 27000 | 0.3693 | 3.5050 |
80
+ | 3.4278 | 3.0189 | 28000 | 0.3700 | 3.5020 |
81
+ | 3.4513 | 3.1267 | 29000 | 0.3708 | 3.4942 |
82
+ | 3.4627 | 3.2345 | 30000 | 0.3713 | 3.4898 |
83
+ | 3.4463 | 3.3423 | 31000 | 0.3720 | 3.4832 |
84
+ | 3.461 | 3.4501 | 32000 | 0.3727 | 3.4776 |
85
+ | 3.4454 | 3.5580 | 33000 | 0.3733 | 3.4710 |
86
+ | 3.4627 | 3.6658 | 34000 | 0.3741 | 3.4632 |
87
+ | 3.444 | 3.7736 | 35000 | 0.3743 | 3.4588 |
88
+ | 3.4513 | 3.8814 | 36000 | 0.3752 | 3.4522 |
89
+ | 3.4411 | 3.9892 | 37000 | 0.3758 | 3.4460 |
90
+ | 3.3781 | 4.0970 | 38000 | 0.3764 | 3.4480 |
91
+ | 3.3767 | 4.2049 | 39000 | 0.3767 | 3.4423 |
92
+ | 3.3947 | 4.3127 | 40000 | 0.3773 | 3.4385 |
93
+ | 3.389 | 4.4205 | 41000 | 0.3776 | 3.4339 |
94
+ | 3.3849 | 4.5283 | 42000 | 0.3786 | 3.4289 |
95
+ | 3.4041 | 4.6361 | 43000 | 0.3791 | 3.4244 |
96
+ | 3.3693 | 4.7439 | 44000 | 0.3796 | 3.4180 |
97
+ | 3.3953 | 4.8518 | 45000 | 0.3799 | 3.4133 |
98
+ | 3.3841 | 4.9596 | 46000 | 0.3805 | 3.4098 |
99
+ | 3.3078 | 5.0674 | 47000 | 0.3803 | 3.4134 |
100
+ | 3.315 | 5.1752 | 48000 | 0.3810 | 3.4103 |
101
+ | 3.3108 | 5.2830 | 49000 | 0.3809 | 3.4087 |
102
+ | 3.3356 | 5.3908 | 50000 | 0.3818 | 3.4030 |
103
+ | 3.3193 | 5.4987 | 51000 | 0.3824 | 3.3966 |
104
+ | 3.3407 | 5.6065 | 52000 | 0.3826 | 3.3946 |
105
+ | 3.348 | 5.7143 | 53000 | 0.3831 | 3.3898 |
106
+ | 3.3288 | 5.8221 | 54000 | 0.3834 | 3.3838 |
107
+ | 3.3283 | 5.9299 | 55000 | 0.3838 | 3.3800 |
108
+ | 3.2613 | 6.0377 | 56000 | 0.3840 | 3.3837 |
109
+ | 3.2509 | 6.1456 | 57000 | 0.3840 | 3.3834 |
110
+ | 3.2539 | 6.2534 | 58000 | 0.3846 | 3.3802 |
111
+ | 3.2851 | 6.3612 | 59000 | 0.3851 | 3.3774 |
112
+ | 3.2878 | 6.4690 | 60000 | 0.3853 | 3.3725 |
113
+ | 3.2817 | 6.5768 | 61000 | 0.3855 | 3.3693 |
114
+ | 3.2722 | 6.6846 | 62000 | 0.3863 | 3.3639 |
115
+ | 3.2773 | 6.7925 | 63000 | 0.3867 | 3.3609 |
116
+ | 3.2948 | 6.9003 | 64000 | 0.3872 | 3.3567 |
117
+ | 3.1985 | 7.0081 | 65000 | 0.3872 | 3.3584 |
118
+ | 3.2183 | 7.1159 | 66000 | 0.3874 | 3.3596 |
119
+ | 3.2204 | 7.2237 | 67000 | 0.3878 | 3.3558 |
120
+ | 3.2237 | 7.3315 | 68000 | 0.3881 | 3.3549 |
121
+ | 3.2464 | 7.4394 | 69000 | 0.3881 | 3.3498 |
122
+ | 3.2299 | 7.5472 | 70000 | 0.3887 | 3.3457 |
123
+ | 3.2273 | 7.6550 | 71000 | 0.3889 | 3.3421 |
124
+ | 3.2389 | 7.7628 | 72000 | 0.3894 | 3.3363 |
125
+ | 3.2224 | 7.8706 | 73000 | 0.3899 | 3.3338 |
126
+ | 3.2305 | 7.9784 | 74000 | 0.3902 | 3.3320 |
127
+ | 3.1599 | 8.0863 | 75000 | 0.3901 | 3.3381 |
128
+ | 3.1664 | 8.1941 | 76000 | 0.3903 | 3.3350 |
129
+ | 3.1774 | 8.3019 | 77000 | 0.3903 | 3.3318 |
130
+ | 3.1677 | 8.4097 | 78000 | 0.3909 | 3.3281 |
131
+ | 3.1821 | 8.5175 | 79000 | 0.3913 | 3.3258 |
132
+ | 3.1805 | 8.6253 | 80000 | 0.3917 | 3.3235 |
133
+ | 3.1932 | 8.7332 | 81000 | 0.3920 | 3.3191 |
134
+ | 3.1771 | 8.8410 | 82000 | 0.3924 | 3.3148 |
135
+ | 3.1927 | 8.9488 | 83000 | 0.3927 | 3.3131 |
136
+ | 3.1235 | 9.0566 | 84000 | 0.3926 | 3.3163 |
137
+ | 3.1221 | 9.1644 | 85000 | 0.3929 | 3.3138 |
138
+ | 3.1312 | 9.2722 | 86000 | 0.3931 | 3.3123 |
139
+ | 3.1204 | 9.3801 | 87000 | 0.3934 | 3.3111 |
140
+ | 3.1289 | 9.4879 | 88000 | 0.3939 | 3.3072 |
141
+ | 3.1452 | 9.5957 | 89000 | 0.3940 | 3.3056 |
142
+ | 3.1301 | 9.7035 | 90000 | 0.3943 | 3.3030 |
143
+ | 3.1314 | 9.8113 | 91000 | 0.3943 | 3.3021 |
144
+ | 3.1061 | 9.9191 | 92000 | 0.3946 | 3.3004 |
145
+ | 3.2987 | 10.0270 | 93000 | 3.4598 | 0.3772 |
146
+ | 3.3802 | 10.1348 | 94000 | 3.4702 | 0.3756 |
147
+ | 3.3824 | 10.2426 | 95000 | 3.4624 | 0.3763 |
148
 
149
 
150
  ### Framework versions
all_results.json ADDED
@@ -0,0 +1,16 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 10.0,
3
+ "eval_accuracy": 0.39428448868113974,
4
+ "eval_loss": 3.3030037879943848,
5
+ "eval_runtime": 181.4646,
6
+ "eval_samples": 18011,
7
+ "eval_samples_per_second": 99.254,
8
+ "eval_steps_per_second": 6.205,
9
+ "perplexity": 27.194201977682727,
10
+ "total_flos": 7.75449427968e+17,
11
+ "train_loss": 3.4548003072815763,
12
+ "train_runtime": 79154.1696,
13
+ "train_samples": 296775,
14
+ "train_samples_per_second": 37.493,
15
+ "train_steps_per_second": 1.172
16
+ }
eval_results.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 10.0,
3
+ "eval_accuracy": 0.39428448868113974,
4
+ "eval_loss": 3.3030037879943848,
5
+ "eval_runtime": 181.4646,
6
+ "eval_samples": 18011,
7
+ "eval_samples_per_second": 99.254,
8
+ "eval_steps_per_second": 6.205,
9
+ "perplexity": 27.194201977682727
10
+ }
train_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 10.0,
3
+ "total_flos": 7.75449427968e+17,
4
+ "train_loss": 3.4548003072815763,
5
+ "train_runtime": 79154.1696,
6
+ "train_samples": 296775,
7
+ "train_samples_per_second": 37.493,
8
+ "train_steps_per_second": 1.172
9
+ }
trainer_state.json ADDED
The diff for this file is too large to render. See raw diff
 
training_args.bin CHANGED
@@ -1,3 +1,3 @@
1
  version https://git-lfs.github.com/spec/v1
2
- oid sha256:869e284510a25da056f1bb9a81ca3fecc0da8e7b5155c416dd9c5c0c11ed1983
3
  size 5304
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:6dbba059ae4e965260f4be9ebe73f66dc0ec9b205428e29ea65d2a6a6ded9fe2
3
  size 5304