Hugging Face's logo

craa
/
100M__6910

Text Generation
Transformers
Safetensors
gpt2
Generated from Trainer
text-generation-inference
Model card Files
xet
 Community
craa commited on
Commit
76884ed
·
verified ·
1 Parent(s): 5137c12

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: 3.3012
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
- | 5.1167 | 0.1078 | 1000 | 5.0316 | 0.2265 |
54
- | 4.5939 | 0.2156 | 2000 | 4.5123 | 0.2701 |
55
- | 4.3174 | 0.3235 | 3000 | 4.2394 | 0.2980 |
56
- | 4.1616 | 0.4313 | 4000 | 4.0950 | 0.3117 |
57
- | 4.0631 | 0.5391 | 5000 | 3.9943 | 0.3210 |
58
- | 4.0063 | 0.6469 | 6000 | 3.9201 | 0.3278 |
59
- | 3.9296 | 0.7547 | 7000 | 3.8656 | 0.3329 |
60
- | 3.8882 | 0.8625 | 8000 | 3.8190 | 0.3372 |
61
- | 3.8432 | 0.9704 | 9000 | 3.7808 | 0.3407 |
62
- | 3.7782 | 1.0782 | 10000 | 3.7569 | 0.3441 |
63
- | 3.7405 | 1.1860 | 11000 | 3.7269 | 0.3471 |
64
- | 3.7342 | 1.2938 | 12000 | 3.7020 | 0.3493 |
65
- | 3.7098 | 1.4016 | 13000 | 3.6816 | 0.3514 |
66
- | 3.7016 | 1.5094 | 14000 | 3.6587 | 0.3536 |
67
- | 3.6781 | 1.6173 | 15000 | 3.6408 | 0.3555 |
68
- | 3.6798 | 1.7251 | 16000 | 3.6209 | 0.3569 |
69
- | 3.6639 | 1.8329 | 17000 | 3.6067 | 0.3586 |
70
- | 3.6461 | 1.9407 | 18000 | 3.5905 | 0.3602 |
71
- | 3.5609 | 2.0485 | 19000 | 3.5818 | 0.3615 |
72
- | 3.5583 | 2.1563 | 20000 | 3.5729 | 0.3628 |
73
- | 3.5592 | 2.2642 | 21000 | 3.5644 | 0.3634 |
74
- | 3.5447 | 2.3720 | 22000 | 3.5518 | 0.3647 |
75
- | 3.5527 | 2.4798 | 23000 | 3.5423 | 0.3656 |
76
- | 3.541 | 2.5876 | 24000 | 3.5317 | 0.3668 |
77
- | 3.5277 | 2.6954 | 25000 | 3.5235 | 0.3674 |
78
- | 3.5395 | 2.8032 | 26000 | 3.5142 | 0.3687 |
79
- | 3.522 | 2.9111 | 27000 | 3.5048 | 0.3694 |
80
- | 3.4455 | 3.0189 | 28000 | 3.5026 | 0.3705 |
81
- | 3.4399 | 3.1267 | 29000 | 3.4956 | 0.3706 |
82
- | 3.4688 | 3.2345 | 30000 | 3.4911 | 0.3718 |
83
- | 3.4748 | 3.3423 | 31000 | 3.4835 | 0.3726 |
84
- | 3.4633 | 3.4501 | 32000 | 3.4767 | 0.3733 |
85
- | 3.4541 | 3.5580 | 33000 | 3.4712 | 0.3735 |
86
- | 3.4513 | 3.6658 | 34000 | 3.4640 | 0.3745 |
87
- | 3.4707 | 3.7736 | 35000 | 3.4572 | 0.3750 |
88
- | 3.4347 | 3.8814 | 36000 | 3.4530 | 0.3755 |
89
- | 3.417 | 3.9892 | 37000 | 3.4462 | 0.3759 |
90
- | 3.3602 | 4.0970 | 38000 | 3.4484 | 0.3764 |
91
- | 3.384 | 4.2049 | 39000 | 3.4457 | 0.3770 |
92
- | 3.4014 | 4.3127 | 40000 | 3.4402 | 0.3773 |
93
- | 3.4152 | 4.4205 | 41000 | 3.4369 | 0.3779 |
94
- | 3.3867 | 4.5283 | 42000 | 3.4306 | 0.3787 |
95
- | 3.4083 | 4.6361 | 43000 | 3.4263 | 0.3791 |
96
- | 3.3929 | 4.7439 | 44000 | 3.4202 | 0.3794 |
97
- | 3.389 | 4.8518 | 45000 | 3.4123 | 0.3803 |
98
- | 3.3876 | 4.9596 | 46000 | 3.4127 | 0.3805 |
99
- | 3.3161 | 5.0674 | 47000 | 3.4123 | 0.3810 |
100
- | 3.3084 | 5.1752 | 48000 | 3.4124 | 0.3809 |
101
- | 3.329 | 5.2830 | 49000 | 3.4083 | 0.3815 |
102
- | 3.3415 | 5.3908 | 50000 | 3.4022 | 0.3817 |
103
- | 3.3172 | 5.4987 | 51000 | 3.3990 | 0.3822 |
104
- | 3.3287 | 5.6065 | 52000 | 3.3959 | 0.3826 |
105
- | 3.3407 | 5.7143 | 53000 | 3.3904 | 0.3830 |
106
- | 3.3252 | 5.8221 | 54000 | 3.3858 | 0.3834 |
107
- | 3.3267 | 5.9299 | 55000 | 3.3821 | 0.3839 |
108
- | 3.2385 | 6.0377 | 56000 | 3.3859 | 0.3840 |
109
- | 3.2695 | 6.1456 | 57000 | 3.3841 | 0.3841 |
110
- | 3.2792 | 6.2534 | 58000 | 3.3813 | 0.3846 |
111
- | 3.2941 | 6.3612 | 59000 | 3.3767 | 0.3850 |
112
- | 3.2853 | 6.4690 | 60000 | 3.3732 | 0.3855 |
113
- | 3.2766 | 6.5768 | 61000 | 3.3684 | 0.3860 |
114
- | 3.2783 | 6.6846 | 62000 | 3.3660 | 0.3860 |
115
- | 3.2875 | 6.7925 | 63000 | 3.3614 | 0.3866 |
116
- | 3.3016 | 6.9003 | 64000 | 3.3557 | 0.3870 |
117
- | 3.2097 | 7.0081 | 65000 | 3.3597 | 0.3869 |
118
- | 3.2227 | 7.1159 | 66000 | 3.3602 | 0.3876 |
119
- | 3.2331 | 7.2237 | 67000 | 3.3562 | 0.3875 |
120
- | 3.2154 | 7.3315 | 68000 | 3.3542 | 0.3880 |
121
- | 3.2493 | 7.4394 | 69000 | 3.3500 | 0.3881 |
122
- | 3.2359 | 7.5472 | 70000 | 3.3460 | 0.3887 |
123
- | 3.236 | 7.6550 | 71000 | 3.3415 | 0.3892 |
124
- | 3.2325 | 7.7628 | 72000 | 3.3391 | 0.3895 |
125
- | 3.2321 | 7.8706 | 73000 | 3.3348 | 0.3898 |
126
- | 3.2455 | 7.9784 | 74000 | 3.3332 | 0.3900 |
127
- | 3.1591 | 8.0863 | 75000 | 3.3374 | 0.3900 |
128
- | 3.1871 | 8.1941 | 76000 | 3.3381 | 0.3901 |
129
- | 3.1747 | 8.3019 | 77000 | 3.3329 | 0.3905 |
130
- | 3.1932 | 8.4097 | 78000 | 3.3302 | 0.3909 |
131
- | 3.1805 | 8.5175 | 79000 | 3.3280 | 0.3911 |
132
- | 3.1777 | 8.6253 | 80000 | 3.3246 | 0.3917 |
133
- | 3.1899 | 8.7332 | 81000 | 3.3193 | 0.3921 |
134
- | 3.196 | 8.8410 | 82000 | 3.3160 | 0.3924 |
135
- | 3.1743 | 8.9488 | 83000 | 3.3144 | 0.3926 |
136
- | 3.1092 | 9.0566 | 84000 | 3.3175 | 0.3926 |
137
- | 3.1468 | 9.1644 | 85000 | 3.3152 | 0.3929 |
138
- | 3.1336 | 9.2722 | 86000 | 3.3148 | 0.3930 |
139
- | 3.1193 | 9.3801 | 87000 | 3.3122 | 0.3933 |
140
- | 3.1383 | 9.4879 | 88000 | 3.3087 | 0.3938 |
141
- | 3.13 | 9.5957 | 89000 | 3.3068 | 0.3938 |
142
- | 3.135 | 9.7035 | 90000 | 3.3042 | 0.3942 |
143
- | 3.1324 | 9.8113 | 91000 | 3.3028 | 0.3944 |
144
- | 3.1419 | 9.9191 | 92000 | 3.3012 | 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: 3.4646
20
+ - Accuracy: 0.3759
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
+ | 5.1167 | 0.1078 | 1000 | 0.2265 | 5.0316 |
54
+ | 4.5939 | 0.2156 | 2000 | 0.2701 | 4.5123 |
55
+ | 4.3174 | 0.3235 | 3000 | 0.2980 | 4.2394 |
56
+ | 4.1616 | 0.4313 | 4000 | 0.3117 | 4.0950 |
57
+ | 4.0631 | 0.5391 | 5000 | 0.3210 | 3.9943 |
58
+ | 4.0063 | 0.6469 | 6000 | 0.3278 | 3.9201 |
59
+ | 3.9296 | 0.7547 | 7000 | 0.3329 | 3.8656 |
60
+ | 3.8882 | 0.8625 | 8000 | 0.3372 | 3.8190 |
61
+ | 3.8432 | 0.9704 | 9000 | 0.3407 | 3.7808 |
62
+ | 3.7782 | 1.0782 | 10000 | 0.3441 | 3.7569 |
63
+ | 3.7405 | 1.1860 | 11000 | 0.3471 | 3.7269 |
64
+ | 3.7342 | 1.2938 | 12000 | 0.3493 | 3.7020 |
65
+ | 3.7098 | 1.4016 | 13000 | 0.3514 | 3.6816 |
66
+ | 3.7016 | 1.5094 | 14000 | 0.3536 | 3.6587 |
67
+ | 3.6781 | 1.6173 | 15000 | 0.3555 | 3.6408 |
68
+ | 3.6798 | 1.7251 | 16000 | 0.3569 | 3.6209 |
69
+ | 3.6639 | 1.8329 | 17000 | 0.3586 | 3.6067 |
70
+ | 3.6461 | 1.9407 | 18000 | 0.3602 | 3.5905 |
71
+ | 3.5609 | 2.0485 | 19000 | 0.3615 | 3.5818 |
72
+ | 3.5583 | 2.1563 | 20000 | 0.3628 | 3.5729 |
73
+ | 3.5592 | 2.2642 | 21000 | 0.3634 | 3.5644 |
74
+ | 3.5447 | 2.3720 | 22000 | 0.3647 | 3.5518 |
75
+ | 3.5527 | 2.4798 | 23000 | 0.3656 | 3.5423 |
76
+ | 3.541 | 2.5876 | 24000 | 0.3668 | 3.5317 |
77
+ | 3.5277 | 2.6954 | 25000 | 0.3674 | 3.5235 |
78
+ | 3.5395 | 2.8032 | 26000 | 0.3687 | 3.5142 |
79
+ | 3.522 | 2.9111 | 27000 | 0.3694 | 3.5048 |
80
+ | 3.4455 | 3.0189 | 28000 | 0.3705 | 3.5026 |
81
+ | 3.4399 | 3.1267 | 29000 | 0.3706 | 3.4956 |
82
+ | 3.4688 | 3.2345 | 30000 | 0.3718 | 3.4911 |
83
+ | 3.4748 | 3.3423 | 31000 | 0.3726 | 3.4835 |
84
+ | 3.4633 | 3.4501 | 32000 | 0.3733 | 3.4767 |
85
+ | 3.4541 | 3.5580 | 33000 | 0.3735 | 3.4712 |
86
+ | 3.4513 | 3.6658 | 34000 | 0.3745 | 3.4640 |
87
+ | 3.4707 | 3.7736 | 35000 | 0.3750 | 3.4572 |
88
+ | 3.4347 | 3.8814 | 36000 | 0.3755 | 3.4530 |
89
+ | 3.417 | 3.9892 | 37000 | 0.3759 | 3.4462 |
90
+ | 3.3602 | 4.0970 | 38000 | 0.3764 | 3.4484 |
91
+ | 3.384 | 4.2049 | 39000 | 0.3770 | 3.4457 |
92
+ | 3.4014 | 4.3127 | 40000 | 0.3773 | 3.4402 |
93
+ | 3.4152 | 4.4205 | 41000 | 0.3779 | 3.4369 |
94
+ | 3.3867 | 4.5283 | 42000 | 0.3787 | 3.4306 |
95
+ | 3.4083 | 4.6361 | 43000 | 0.3791 | 3.4263 |
96
+ | 3.3929 | 4.7439 | 44000 | 0.3794 | 3.4202 |
97
+ | 3.389 | 4.8518 | 45000 | 0.3803 | 3.4123 |
98
+ | 3.3876 | 4.9596 | 46000 | 0.3805 | 3.4127 |
99
+ | 3.3161 | 5.0674 | 47000 | 0.3810 | 3.4123 |
100
+ | 3.3084 | 5.1752 | 48000 | 0.3809 | 3.4124 |
101
+ | 3.329 | 5.2830 | 49000 | 0.3815 | 3.4083 |
102
+ | 3.3415 | 5.3908 | 50000 | 0.3817 | 3.4022 |
103
+ | 3.3172 | 5.4987 | 51000 | 0.3822 | 3.3990 |
104
+ | 3.3287 | 5.6065 | 52000 | 0.3826 | 3.3959 |
105
+ | 3.3407 | 5.7143 | 53000 | 0.3830 | 3.3904 |
106
+ | 3.3252 | 5.8221 | 54000 | 0.3834 | 3.3858 |
107
+ | 3.3267 | 5.9299 | 55000 | 0.3839 | 3.3821 |
108
+ | 3.2385 | 6.0377 | 56000 | 0.3840 | 3.3859 |
109
+ | 3.2695 | 6.1456 | 57000 | 0.3841 | 3.3841 |
110
+ | 3.2792 | 6.2534 | 58000 | 0.3846 | 3.3813 |
111
+ | 3.2941 | 6.3612 | 59000 | 0.3850 | 3.3767 |
112
+ | 3.2853 | 6.4690 | 60000 | 0.3855 | 3.3732 |
113
+ | 3.2766 | 6.5768 | 61000 | 0.3860 | 3.3684 |
114
+ | 3.2783 | 6.6846 | 62000 | 0.3860 | 3.3660 |
115
+ | 3.2875 | 6.7925 | 63000 | 0.3866 | 3.3614 |
116
+ | 3.3016 | 6.9003 | 64000 | 0.3870 | 3.3557 |
117
+ | 3.2097 | 7.0081 | 65000 | 0.3869 | 3.3597 |
118
+ | 3.2227 | 7.1159 | 66000 | 0.3876 | 3.3602 |
119
+ | 3.2331 | 7.2237 | 67000 | 0.3875 | 3.3562 |
120
+ | 3.2154 | 7.3315 | 68000 | 0.3880 | 3.3542 |
121
+ | 3.2493 | 7.4394 | 69000 | 0.3881 | 3.3500 |
122
+ | 3.2359 | 7.5472 | 70000 | 0.3887 | 3.3460 |
123
+ | 3.236 | 7.6550 | 71000 | 0.3892 | 3.3415 |
124
+ | 3.2325 | 7.7628 | 72000 | 0.3895 | 3.3391 |
125
+ | 3.2321 | 7.8706 | 73000 | 0.3898 | 3.3348 |
126
+ | 3.2455 | 7.9784 | 74000 | 0.3900 | 3.3332 |
127
+ | 3.1591 | 8.0863 | 75000 | 0.3900 | 3.3374 |
128
+ | 3.1871 | 8.1941 | 76000 | 0.3901 | 3.3381 |
129
+ | 3.1747 | 8.3019 | 77000 | 0.3905 | 3.3329 |
130
+ | 3.1932 | 8.4097 | 78000 | 0.3909 | 3.3302 |
131
+ | 3.1805 | 8.5175 | 79000 | 0.3911 | 3.3280 |
132
+ | 3.1777 | 8.6253 | 80000 | 0.3917 | 3.3246 |
133
+ | 3.1899 | 8.7332 | 81000 | 0.3921 | 3.3193 |
134
+ | 3.196 | 8.8410 | 82000 | 0.3924 | 3.3160 |
135
+ | 3.1743 | 8.9488 | 83000 | 0.3926 | 3.3144 |
136
+ | 3.1092 | 9.0566 | 84000 | 0.3926 | 3.3175 |
137
+ | 3.1468 | 9.1644 | 85000 | 0.3929 | 3.3152 |
138
+ | 3.1336 | 9.2722 | 86000 | 0.3930 | 3.3148 |
139
+ | 3.1193 | 9.3801 | 87000 | 0.3933 | 3.3122 |
140
+ | 3.1383 | 9.4879 | 88000 | 0.3938 | 3.3087 |
141
+ | 3.13 | 9.5957 | 89000 | 0.3938 | 3.3068 |
142
+ | 3.135 | 9.7035 | 90000 | 0.3942 | 3.3042 |
143
+ | 3.1324 | 9.8113 | 91000 | 0.3944 | 3.3028 |
144
+ | 3.1419 | 9.9191 | 92000 | 0.3946 | 3.3012 |
145
+ | 3.2899 | 10.0270 | 93000 | 3.4552 | 0.3784 |
146
+ | 3.3629 | 10.1348 | 94000 | 3.4698 | 0.3757 |
147
+ | 3.3727 | 10.2426 | 95000 | 3.4646 | 0.3759 |
148
 
149
 
150
  ### Framework versions
all_results.json ADDED
@@ -0,0 +1,16 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 10.0,
3
+ "eval_accuracy": 0.3941857232061164,
4
+ "eval_loss": 3.3042373657226562,
5
+ "eval_runtime": 183.0156,
6
+ "eval_samples": 18011,
7
+ "eval_samples_per_second": 98.412,
8
+ "eval_steps_per_second": 6.152,
9
+ "perplexity": 27.227768838990173,
10
+ "total_flos": 7.75449427968e+17,
11
+ "train_loss": 3.4577919183201544,
12
+ "train_runtime": 79361.548,
13
+ "train_samples": 296775,
14
+ "train_samples_per_second": 37.395,
15
+ "train_steps_per_second": 1.169
16
+ }
eval_results.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 10.0,
3
+ "eval_accuracy": 0.3941857232061164,
4
+ "eval_loss": 3.3042373657226562,
5
+ "eval_runtime": 183.0156,
6
+ "eval_samples": 18011,
7
+ "eval_samples_per_second": 98.412,
8
+ "eval_steps_per_second": 6.152,
9
+ "perplexity": 27.227768838990173
10
+ }
train_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 10.0,
3
+ "total_flos": 7.75449427968e+17,
4
+ "train_loss": 3.4577919183201544,
5
+ "train_runtime": 79361.548,
6
+ "train_samples": 296775,
7
+ "train_samples_per_second": 37.395,
8
+ "train_steps_per_second": 1.169
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 sha256:0d35e51458d6994cce9ea8f8150dbd01a34f4cef3ae97d9da84c514e46faf8a3
3
  size 5304
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:a5370cb9d79e4d219f2c8f35a18d33b03650be712ffdb19623c3e2a34c403257
3
  size 5304