Hugging Face's logo

craa
/
100M__495

Text Generation
Transformers
Safetensors
gpt2
Generated from Trainer
text-generation-inference
Model card Files
xet
Community
craa commited on
Commit
4f008ed
·
verified ·
1 Parent(s): 175db17

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: 3.2984
20
- - Accuracy: 0.3950
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
- | 5.1665 | 0.1078 | 1000 | 5.0912 | 0.2206 |
54
- | 4.6555 | 0.2156 | 2000 | 4.6008 | 0.2607 |
55
- | 4.3901 | 0.3235 | 3000 | 4.2963 | 0.2916 |
56
- | 4.2075 | 0.4313 | 4000 | 4.1342 | 0.3082 |
57
- | 4.0819 | 0.5391 | 5000 | 4.0228 | 0.3186 |
58
- | 4.0201 | 0.6469 | 6000 | 3.9409 | 0.3257 |
59
- | 3.9519 | 0.7547 | 7000 | 3.8853 | 0.3312 |
60
- | 3.8917 | 0.8625 | 8000 | 3.8361 | 0.3359 |
61
- | 3.8592 | 0.9704 | 9000 | 3.7949 | 0.3404 |
62
- | 3.7811 | 1.0782 | 10000 | 3.7621 | 0.3437 |
63
- | 3.7908 | 1.1860 | 11000 | 3.7343 | 0.3460 |
64
- | 3.7514 | 1.2938 | 12000 | 3.7092 | 0.3482 |
65
- | 3.7207 | 1.4016 | 13000 | 3.6841 | 0.3513 |
66
- | 3.7092 | 1.5094 | 14000 | 3.6657 | 0.3533 |
67
- | 3.6968 | 1.6173 | 15000 | 3.6474 | 0.3550 |
68
- | 3.6724 | 1.7251 | 16000 | 3.6263 | 0.3571 |
69
- | 3.6737 | 1.8329 | 17000 | 3.6162 | 0.3582 |
70
- | 3.6477 | 1.9407 | 18000 | 3.5972 | 0.3598 |
71
- | 3.5922 | 2.0485 | 19000 | 3.5889 | 0.3611 |
72
- | 3.5706 | 2.1563 | 20000 | 3.5809 | 0.3621 |
73
- | 3.5568 | 2.2642 | 21000 | 3.5704 | 0.3631 |
74
- | 3.5596 | 2.3720 | 22000 | 3.5584 | 0.3645 |
75
- | 3.5529 | 2.4798 | 23000 | 3.5451 | 0.3655 |
76
- | 3.5466 | 2.5876 | 24000 | 3.5379 | 0.3665 |
77
- | 3.555 | 2.6954 | 25000 | 3.5254 | 0.3674 |
78
- | 3.5597 | 2.8032 | 26000 | 3.5180 | 0.3684 |
79
- | 3.5422 | 2.9111 | 27000 | 3.5087 | 0.3693 |
80
- | 3.4422 | 3.0189 | 28000 | 3.5041 | 0.3700 |
81
- | 3.4594 | 3.1267 | 29000 | 3.4994 | 0.3709 |
82
- | 3.4767 | 3.2345 | 30000 | 3.4932 | 0.3715 |
83
- | 3.4629 | 3.3423 | 31000 | 3.4884 | 0.3725 |
84
- | 3.471 | 3.4501 | 32000 | 3.4814 | 0.3728 |
85
- | 3.4594 | 3.5580 | 33000 | 3.4750 | 0.3735 |
86
- | 3.4814 | 3.6658 | 34000 | 3.4692 | 0.3739 |
87
- | 3.4593 | 3.7736 | 35000 | 3.4629 | 0.3747 |
88
- | 3.4435 | 3.8814 | 36000 | 3.4561 | 0.3756 |
89
- | 3.4633 | 3.9892 | 37000 | 3.4494 | 0.3759 |
90
- | 3.381 | 4.0970 | 38000 | 3.4523 | 0.3766 |
91
- | 3.3899 | 4.2049 | 39000 | 3.4458 | 0.3769 |
92
- | 3.3917 | 4.3127 | 40000 | 3.4431 | 0.3774 |
93
- | 3.4098 | 4.4205 | 41000 | 3.4349 | 0.3781 |
94
- | 3.4073 | 4.5283 | 42000 | 3.4332 | 0.3784 |
95
- | 3.3993 | 4.6361 | 43000 | 3.4270 | 0.3786 |
96
- | 3.4026 | 4.7439 | 44000 | 3.4206 | 0.3797 |
97
- | 3.407 | 4.8518 | 45000 | 3.4162 | 0.3798 |
98
- | 3.3882 | 4.9596 | 46000 | 3.4124 | 0.3807 |
99
- | 3.3153 | 5.0674 | 47000 | 3.4160 | 0.3809 |
100
- | 3.3206 | 5.1752 | 48000 | 3.4111 | 0.3811 |
101
- | 3.3424 | 5.2830 | 49000 | 3.4074 | 0.3814 |
102
- | 3.3458 | 5.3908 | 50000 | 3.4056 | 0.3819 |
103
- | 3.3476 | 5.4987 | 51000 | 3.4017 | 0.3825 |
104
- | 3.3541 | 5.6065 | 52000 | 3.3958 | 0.3827 |
105
- | 3.347 | 5.7143 | 53000 | 3.3918 | 0.3831 |
106
- | 3.3347 | 5.8221 | 54000 | 3.3862 | 0.3836 |
107
- | 3.3565 | 5.9299 | 55000 | 3.3812 | 0.3844 |
108
- | 3.2551 | 6.0377 | 56000 | 3.3851 | 0.3845 |
109
- | 3.2716 | 6.1456 | 57000 | 3.3840 | 0.3843 |
110
- | 3.2895 | 6.2534 | 58000 | 3.3813 | 0.3848 |
111
- | 3.2759 | 6.3612 | 59000 | 3.3767 | 0.3856 |
112
- | 3.2909 | 6.4690 | 60000 | 3.3733 | 0.3853 |
113
- | 3.2835 | 6.5768 | 61000 | 3.3691 | 0.3859 |
114
- | 3.3046 | 6.6846 | 62000 | 3.3650 | 0.3865 |
115
- | 3.2794 | 6.7925 | 63000 | 3.3618 | 0.3867 |
116
- | 3.2831 | 6.9003 | 64000 | 3.3556 | 0.3873 |
117
- | 3.2134 | 7.0081 | 65000 | 3.3601 | 0.3874 |
118
- | 3.2066 | 7.1159 | 66000 | 3.3606 | 0.3873 |
119
- | 3.2313 | 7.2237 | 67000 | 3.3570 | 0.3881 |
120
- | 3.2319 | 7.3315 | 68000 | 3.3540 | 0.3881 |
121
- | 3.2282 | 7.4394 | 69000 | 3.3505 | 0.3882 |
122
- | 3.2386 | 7.5472 | 70000 | 3.3455 | 0.3889 |
123
- | 3.233 | 7.6550 | 71000 | 3.3429 | 0.3891 |
124
- | 3.2386 | 7.7628 | 72000 | 3.3377 | 0.3898 |
125
- | 3.247 | 7.8706 | 73000 | 3.3356 | 0.3899 |
126
- | 3.2571 | 7.9784 | 74000 | 3.3312 | 0.3908 |
127
- | 3.1695 | 8.0863 | 75000 | 3.3377 | 0.3905 |
128
- | 3.1903 | 8.1941 | 76000 | 3.3351 | 0.3906 |
129
- | 3.1892 | 8.3019 | 77000 | 3.3317 | 0.3909 |
130
- | 3.1795 | 8.4097 | 78000 | 3.3293 | 0.3912 |
131
- | 3.179 | 8.5175 | 79000 | 3.3260 | 0.3916 |
132
- | 3.2147 | 8.6253 | 80000 | 3.3226 | 0.3919 |
133
- | 3.1679 | 8.7332 | 81000 | 3.3194 | 0.3922 |
134
- | 3.1853 | 8.8410 | 82000 | 3.3157 | 0.3925 |
135
- | 3.1983 | 8.9488 | 83000 | 3.3126 | 0.3929 |
136
- | 3.1342 | 9.0566 | 84000 | 3.3160 | 0.3928 |
137
- | 3.1281 | 9.1644 | 85000 | 3.3144 | 0.3930 |
138
- | 3.129 | 9.2722 | 86000 | 3.3116 | 0.3934 |
139
- | 3.1315 | 9.3801 | 87000 | 3.3096 | 0.3937 |
140
- | 3.1295 | 9.4879 | 88000 | 3.3074 | 0.3940 |
141
- | 3.1314 | 9.5957 | 89000 | 3.3044 | 0.3942 |
142
- | 3.1393 | 9.7035 | 90000 | 3.3022 | 0.3945 |
143
- | 3.1482 | 9.8113 | 91000 | 3.3000 | 0.3948 |
144
- | 3.1491 | 9.9191 | 92000 | 3.2984 | 0.3950 |
 
 
 
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: 3.4676
20
+ - Accuracy: 0.3760
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
+ | 5.1665 | 0.1078 | 1000 | 0.2206 | 5.0912 |
54
+ | 4.6555 | 0.2156 | 2000 | 0.2607 | 4.6008 |
55
+ | 4.3901 | 0.3235 | 3000 | 0.2916 | 4.2963 |
56
+ | 4.2075 | 0.4313 | 4000 | 0.3082 | 4.1342 |
57
+ | 4.0819 | 0.5391 | 5000 | 0.3186 | 4.0228 |
58
+ | 4.0201 | 0.6469 | 6000 | 0.3257 | 3.9409 |
59
+ | 3.9519 | 0.7547 | 7000 | 0.3312 | 3.8853 |
60
+ | 3.8917 | 0.8625 | 8000 | 0.3359 | 3.8361 |
61
+ | 3.8592 | 0.9704 | 9000 | 0.3404 | 3.7949 |
62
+ | 3.7811 | 1.0782 | 10000 | 0.3437 | 3.7621 |
63
+ | 3.7908 | 1.1860 | 11000 | 0.3460 | 3.7343 |
64
+ | 3.7514 | 1.2938 | 12000 | 0.3482 | 3.7092 |
65
+ | 3.7207 | 1.4016 | 13000 | 0.3513 | 3.6841 |
66
+ | 3.7092 | 1.5094 | 14000 | 0.3533 | 3.6657 |
67
+ | 3.6968 | 1.6173 | 15000 | 0.3550 | 3.6474 |
68
+ | 3.6724 | 1.7251 | 16000 | 0.3571 | 3.6263 |
69
+ | 3.6737 | 1.8329 | 17000 | 0.3582 | 3.6162 |
70
+ | 3.6477 | 1.9407 | 18000 | 0.3598 | 3.5972 |
71
+ | 3.5922 | 2.0485 | 19000 | 0.3611 | 3.5889 |
72
+ | 3.5706 | 2.1563 | 20000 | 0.3621 | 3.5809 |
73
+ | 3.5568 | 2.2642 | 21000 | 0.3631 | 3.5704 |
74
+ | 3.5596 | 2.3720 | 22000 | 0.3645 | 3.5584 |
75
+ | 3.5529 | 2.4798 | 23000 | 0.3655 | 3.5451 |
76
+ | 3.5466 | 2.5876 | 24000 | 0.3665 | 3.5379 |
77
+ | 3.555 | 2.6954 | 25000 | 0.3674 | 3.5254 |
78
+ | 3.5597 | 2.8032 | 26000 | 0.3684 | 3.5180 |
79
+ | 3.5422 | 2.9111 | 27000 | 0.3693 | 3.5087 |
80
+ | 3.4422 | 3.0189 | 28000 | 0.3700 | 3.5041 |
81
+ | 3.4594 | 3.1267 | 29000 | 0.3709 | 3.4994 |
82
+ | 3.4767 | 3.2345 | 30000 | 0.3715 | 3.4932 |
83
+ | 3.4629 | 3.3423 | 31000 | 0.3725 | 3.4884 |
84
+ | 3.471 | 3.4501 | 32000 | 0.3728 | 3.4814 |
85
+ | 3.4594 | 3.5580 | 33000 | 0.3735 | 3.4750 |
86
+ | 3.4814 | 3.6658 | 34000 | 0.3739 | 3.4692 |
87
+ | 3.4593 | 3.7736 | 35000 | 0.3747 | 3.4629 |
88
+ | 3.4435 | 3.8814 | 36000 | 0.3756 | 3.4561 |
89
+ | 3.4633 | 3.9892 | 37000 | 0.3759 | 3.4494 |
90
+ | 3.381 | 4.0970 | 38000 | 0.3766 | 3.4523 |
91
+ | 3.3899 | 4.2049 | 39000 | 0.3769 | 3.4458 |
92
+ | 3.3917 | 4.3127 | 40000 | 0.3774 | 3.4431 |
93
+ | 3.4098 | 4.4205 | 41000 | 0.3781 | 3.4349 |
94
+ | 3.4073 | 4.5283 | 42000 | 0.3784 | 3.4332 |
95
+ | 3.3993 | 4.6361 | 43000 | 0.3786 | 3.4270 |
96
+ | 3.4026 | 4.7439 | 44000 | 0.3797 | 3.4206 |
97
+ | 3.407 | 4.8518 | 45000 | 0.3798 | 3.4162 |
98
+ | 3.3882 | 4.9596 | 46000 | 0.3807 | 3.4124 |
99
+ | 3.3153 | 5.0674 | 47000 | 0.3809 | 3.4160 |
100
+ | 3.3206 | 5.1752 | 48000 | 0.3811 | 3.4111 |
101
+ | 3.3424 | 5.2830 | 49000 | 0.3814 | 3.4074 |
102
+ | 3.3458 | 5.3908 | 50000 | 0.3819 | 3.4056 |
103
+ | 3.3476 | 5.4987 | 51000 | 0.3825 | 3.4017 |
104
+ | 3.3541 | 5.6065 | 52000 | 0.3827 | 3.3958 |
105
+ | 3.347 | 5.7143 | 53000 | 0.3831 | 3.3918 |
106
+ | 3.3347 | 5.8221 | 54000 | 0.3836 | 3.3862 |
107
+ | 3.3565 | 5.9299 | 55000 | 0.3844 | 3.3812 |
108
+ | 3.2551 | 6.0377 | 56000 | 0.3845 | 3.3851 |
109
+ | 3.2716 | 6.1456 | 57000 | 0.3843 | 3.3840 |
110
+ | 3.2895 | 6.2534 | 58000 | 0.3848 | 3.3813 |
111
+ | 3.2759 | 6.3612 | 59000 | 0.3856 | 3.3767 |
112
+ | 3.2909 | 6.4690 | 60000 | 0.3853 | 3.3733 |
113
+ | 3.2835 | 6.5768 | 61000 | 0.3859 | 3.3691 |
114
+ | 3.3046 | 6.6846 | 62000 | 0.3865 | 3.3650 |
115
+ | 3.2794 | 6.7925 | 63000 | 0.3867 | 3.3618 |
116
+ | 3.2831 | 6.9003 | 64000 | 0.3873 | 3.3556 |
117
+ | 3.2134 | 7.0081 | 65000 | 0.3874 | 3.3601 |
118
+ | 3.2066 | 7.1159 | 66000 | 0.3873 | 3.3606 |
119
+ | 3.2313 | 7.2237 | 67000 | 0.3881 | 3.3570 |
120
+ | 3.2319 | 7.3315 | 68000 | 0.3881 | 3.3540 |
121
+ | 3.2282 | 7.4394 | 69000 | 0.3882 | 3.3505 |
122
+ | 3.2386 | 7.5472 | 70000 | 0.3889 | 3.3455 |
123
+ | 3.233 | 7.6550 | 71000 | 0.3891 | 3.3429 |
124
+ | 3.2386 | 7.7628 | 72000 | 0.3898 | 3.3377 |
125
+ | 3.247 | 7.8706 | 73000 | 0.3899 | 3.3356 |
126
+ | 3.2571 | 7.9784 | 74000 | 0.3908 | 3.3312 |
127
+ | 3.1695 | 8.0863 | 75000 | 0.3905 | 3.3377 |
128
+ | 3.1903 | 8.1941 | 76000 | 0.3906 | 3.3351 |
129
+ | 3.1892 | 8.3019 | 77000 | 0.3909 | 3.3317 |
130
+ | 3.1795 | 8.4097 | 78000 | 0.3912 | 3.3293 |
131
+ | 3.179 | 8.5175 | 79000 | 0.3916 | 3.3260 |
132
+ | 3.2147 | 8.6253 | 80000 | 0.3919 | 3.3226 |
133
+ | 3.1679 | 8.7332 | 81000 | 0.3922 | 3.3194 |
134
+ | 3.1853 | 8.8410 | 82000 | 0.3925 | 3.3157 |
135
+ | 3.1983 | 8.9488 | 83000 | 0.3929 | 3.3126 |
136
+ | 3.1342 | 9.0566 | 84000 | 0.3928 | 3.3160 |
137
+ | 3.1281 | 9.1644 | 85000 | 0.3930 | 3.3144 |
138
+ | 3.129 | 9.2722 | 86000 | 0.3934 | 3.3116 |
139
+ | 3.1315 | 9.3801 | 87000 | 0.3937 | 3.3096 |
140
+ | 3.1295 | 9.4879 | 88000 | 0.3940 | 3.3074 |
141
+ | 3.1314 | 9.5957 | 89000 | 0.3942 | 3.3044 |
142
+ | 3.1393 | 9.7035 | 90000 | 0.3945 | 3.3022 |
143
+ | 3.1482 | 9.8113 | 91000 | 0.3948 | 3.3000 |
144
+ | 3.1491 | 9.9191 | 92000 | 0.3950 | 3.2984 |
145
+ | 3.3065 | 10.0270 | 93000 | 3.4581 | 0.3773 |
146
+ | 3.3805 | 10.1348 | 94000 | 3.4696 | 0.3763 |
147
+ | 3.3884 | 10.2426 | 95000 | 3.4676 | 0.3760 |
148
 
149
 
150
  ### Framework versions
all_results.json ADDED
@@ -0,0 +1,16 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 10.0,
3
+ "eval_accuracy": 0.3945371066453084,
4
+ "eval_loss": 3.302194118499756,
5
+ "eval_runtime": 144.899,
6
+ "eval_samples": 18011,
7
+ "eval_samples_per_second": 124.3,
8
+ "eval_steps_per_second": 7.771,
9
+ "perplexity": 27.17219257328467,
10
+ "total_flos": 7.75449427968e+17,
11
+ "train_loss": 3.4695698613991954,
12
+ "train_runtime": 61247.9496,
13
+ "train_samples": 296775,
14
+ "train_samples_per_second": 48.455,
15
+ "train_steps_per_second": 1.514
16
+ }
eval_results.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 10.0,
3
+ "eval_accuracy": 0.3945371066453084,
4
+ "eval_loss": 3.302194118499756,
5
+ "eval_runtime": 144.899,
6
+ "eval_samples": 18011,
7
+ "eval_samples_per_second": 124.3,
8
+ "eval_steps_per_second": 7.771,
9
+ "perplexity": 27.17219257328467
10
+ }
train_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 10.0,
3
+ "total_flos": 7.75449427968e+17,
4
+ "train_loss": 3.4695698613991954,
5
+ "train_runtime": 61247.9496,
6
+ "train_samples": 296775,
7
+ "train_samples_per_second": 48.455,
8
+ "train_steps_per_second": 1.514
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 sha256:55f163471d892c378f64db98f5c5595dd2d777d1501861716698287d58ed0c89
3
  size 5304
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:7d24f4ca1b52c9db2ed02b327a68a7c52c020ab559dd4b89a1344f00db4d14cb
3
  size 5304