Hugging Face's logo

craa
/
100M__8397

Text Generation
Transformers
Safetensors
gpt2
Generated from Trainer
text-generation-inference
Model card Files
xet
Community
craa commited on
Commit
dc1f265
·
verified ·
1 Parent(s): e1a894b

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: 3.2993
20
- - Accuracy: 0.3945
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
- | 5.0934 | 0.1078 | 1000 | 5.0225 | 0.2275 |
54
- | 4.5937 | 0.2156 | 2000 | 4.5165 | 0.2702 |
55
- | 4.3092 | 0.3235 | 3000 | 4.2461 | 0.2978 |
56
- | 4.1586 | 0.4313 | 4000 | 4.0885 | 0.3127 |
57
- | 4.0492 | 0.5391 | 5000 | 3.9914 | 0.3212 |
58
- | 3.9962 | 0.6469 | 6000 | 3.9193 | 0.3279 |
59
- | 3.9219 | 0.7547 | 7000 | 3.8632 | 0.3335 |
60
- | 3.8566 | 0.8625 | 8000 | 3.8155 | 0.3377 |
61
- | 3.856 | 0.9704 | 9000 | 3.7813 | 0.3408 |
62
- | 3.7649 | 1.0782 | 10000 | 3.7518 | 0.3444 |
63
- | 3.769 | 1.1860 | 11000 | 3.7232 | 0.3474 |
64
- | 3.7288 | 1.2938 | 12000 | 3.6971 | 0.3493 |
65
- | 3.7086 | 1.4016 | 13000 | 3.6748 | 0.3515 |
66
- | 3.703 | 1.5094 | 14000 | 3.6584 | 0.3535 |
67
- | 3.664 | 1.6173 | 15000 | 3.6374 | 0.3555 |
68
- | 3.6639 | 1.7251 | 16000 | 3.6206 | 0.3572 |
69
- | 3.6383 | 1.8329 | 17000 | 3.6051 | 0.3587 |
70
- | 3.6321 | 1.9407 | 18000 | 3.5912 | 0.3600 |
71
- | 3.5672 | 2.0485 | 19000 | 3.5821 | 0.3615 |
72
- | 3.5754 | 2.1563 | 20000 | 3.5733 | 0.3624 |
73
- | 3.5637 | 2.2642 | 21000 | 3.5613 | 0.3637 |
74
- | 3.5706 | 2.3720 | 22000 | 3.5504 | 0.3646 |
75
- | 3.5408 | 2.4798 | 23000 | 3.5402 | 0.3660 |
76
- | 3.532 | 2.5876 | 24000 | 3.5312 | 0.3669 |
77
- | 3.527 | 2.6954 | 25000 | 3.5218 | 0.3673 |
78
- | 3.5333 | 2.8032 | 26000 | 3.5120 | 0.3685 |
79
- | 3.5442 | 2.9111 | 27000 | 3.5054 | 0.3696 |
80
- | 3.4451 | 3.0189 | 28000 | 3.5012 | 0.3700 |
81
- | 3.4295 | 3.1267 | 29000 | 3.4969 | 0.3709 |
82
- | 3.4506 | 3.2345 | 30000 | 3.4902 | 0.3713 |
83
- | 3.4582 | 3.3423 | 31000 | 3.4849 | 0.3719 |
84
- | 3.4597 | 3.4501 | 32000 | 3.4757 | 0.3733 |
85
- | 3.4639 | 3.5580 | 33000 | 3.4722 | 0.3736 |
86
- | 3.4602 | 3.6658 | 34000 | 3.4653 | 0.3735 |
87
- | 3.4613 | 3.7736 | 35000 | 3.4591 | 0.3749 |
88
- | 3.4517 | 3.8814 | 36000 | 3.4511 | 0.3755 |
89
- | 3.4488 | 3.9892 | 37000 | 3.4457 | 0.3761 |
90
- | 3.3644 | 4.0970 | 38000 | 3.4490 | 0.3764 |
91
- | 3.3636 | 4.2049 | 39000 | 3.4452 | 0.3773 |
92
- | 3.3899 | 4.3127 | 40000 | 3.4419 | 0.3775 |
93
- | 3.3969 | 4.4205 | 41000 | 3.4341 | 0.3782 |
94
- | 3.3848 | 4.5283 | 42000 | 3.4308 | 0.3786 |
95
- | 3.3952 | 4.6361 | 43000 | 3.4244 | 0.3788 |
96
- | 3.3875 | 4.7439 | 44000 | 3.4211 | 0.3793 |
97
- | 3.4071 | 4.8518 | 45000 | 3.4152 | 0.3799 |
98
- | 3.3742 | 4.9596 | 46000 | 3.4095 | 0.3803 |
99
- | 3.3158 | 5.0674 | 47000 | 3.4110 | 0.3807 |
100
- | 3.3119 | 5.1752 | 48000 | 3.4111 | 0.3812 |
101
- | 3.3434 | 5.2830 | 49000 | 3.4074 | 0.3813 |
102
- | 3.3422 | 5.3908 | 50000 | 3.4001 | 0.3820 |
103
- | 3.3445 | 5.4987 | 51000 | 3.3994 | 0.3820 |
104
- | 3.3169 | 5.6065 | 52000 | 3.3929 | 0.3826 |
105
- | 3.333 | 5.7143 | 53000 | 3.3878 | 0.3832 |
106
- | 3.331 | 5.8221 | 54000 | 3.3844 | 0.3835 |
107
- | 3.3342 | 5.9299 | 55000 | 3.3803 | 0.3840 |
108
- | 3.2353 | 6.0377 | 56000 | 3.3844 | 0.3840 |
109
- | 3.2567 | 6.1456 | 57000 | 3.3837 | 0.3843 |
110
- | 3.2751 | 6.2534 | 58000 | 3.3811 | 0.3847 |
111
- | 3.2877 | 6.3612 | 59000 | 3.3768 | 0.3850 |
112
- | 3.2882 | 6.4690 | 60000 | 3.3732 | 0.3854 |
113
- | 3.2801 | 6.5768 | 61000 | 3.3683 | 0.3860 |
114
- | 3.2991 | 6.6846 | 62000 | 3.3624 | 0.3863 |
115
- | 3.2791 | 6.7925 | 63000 | 3.3601 | 0.3866 |
116
- | 3.288 | 6.9003 | 64000 | 3.3546 | 0.3871 |
117
- | 3.1829 | 7.0081 | 65000 | 3.3573 | 0.3871 |
118
- | 3.2234 | 7.1159 | 66000 | 3.3607 | 0.3873 |
119
- | 3.2282 | 7.2237 | 67000 | 3.3572 | 0.3876 |
120
- | 3.2267 | 7.3315 | 68000 | 3.3537 | 0.3878 |
121
- | 3.218 | 7.4394 | 69000 | 3.3486 | 0.3883 |
122
- | 3.2235 | 7.5472 | 70000 | 3.3458 | 0.3885 |
123
- | 3.2512 | 7.6550 | 71000 | 3.3421 | 0.3889 |
124
- | 3.2503 | 7.7628 | 72000 | 3.3384 | 0.3895 |
125
- | 3.2303 | 7.8706 | 73000 | 3.3343 | 0.3901 |
126
- | 3.2506 | 7.9784 | 74000 | 3.3303 | 0.3901 |
127
- | 3.155 | 8.0863 | 75000 | 3.3376 | 0.3902 |
128
- | 3.1552 | 8.1941 | 76000 | 3.3350 | 0.3902 |
129
- | 3.1737 | 8.3019 | 77000 | 3.3308 | 0.3907 |
130
- | 3.167 | 8.4097 | 78000 | 3.3294 | 0.3908 |
131
- | 3.1798 | 8.5175 | 79000 | 3.3247 | 0.3912 |
132
- | 3.1957 | 8.6253 | 80000 | 3.3219 | 0.3918 |
133
- | 3.1876 | 8.7332 | 81000 | 3.3189 | 0.3921 |
134
- | 3.1825 | 8.8410 | 82000 | 3.3150 | 0.3922 |
135
- | 3.1641 | 8.9488 | 83000 | 3.3115 | 0.3927 |
136
- | 3.1278 | 9.0566 | 84000 | 3.3139 | 0.3927 |
137
- | 3.1192 | 9.1644 | 85000 | 3.3138 | 0.3928 |
138
- | 3.1443 | 9.2722 | 86000 | 3.3118 | 0.3932 |
139
- | 3.1211 | 9.3801 | 87000 | 3.3095 | 0.3934 |
140
- | 3.1296 | 9.4879 | 88000 | 3.3073 | 0.3936 |
141
- | 3.1073 | 9.5957 | 89000 | 3.3058 | 0.3939 |
142
- | 3.1362 | 9.7035 | 90000 | 3.3025 | 0.3942 |
143
- | 3.1261 | 9.8113 | 91000 | 3.3008 | 0.3944 |
144
- | 3.1183 | 9.9191 | 92000 | 3.2993 | 0.3945 |
 
 
 
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: 3.4628
20
+ - Accuracy: 0.3762
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
+ | 5.0934 | 0.1078 | 1000 | 0.2275 | 5.0225 |
54
+ | 4.5937 | 0.2156 | 2000 | 0.2702 | 4.5165 |
55
+ | 4.3092 | 0.3235 | 3000 | 0.2978 | 4.2461 |
56
+ | 4.1586 | 0.4313 | 4000 | 0.3127 | 4.0885 |
57
+ | 4.0492 | 0.5391 | 5000 | 0.3212 | 3.9914 |
58
+ | 3.9962 | 0.6469 | 6000 | 0.3279 | 3.9193 |
59
+ | 3.9219 | 0.7547 | 7000 | 0.3335 | 3.8632 |
60
+ | 3.8566 | 0.8625 | 8000 | 0.3377 | 3.8155 |
61
+ | 3.856 | 0.9704 | 9000 | 0.3408 | 3.7813 |
62
+ | 3.7649 | 1.0782 | 10000 | 0.3444 | 3.7518 |
63
+ | 3.769 | 1.1860 | 11000 | 0.3474 | 3.7232 |
64
+ | 3.7288 | 1.2938 | 12000 | 0.3493 | 3.6971 |
65
+ | 3.7086 | 1.4016 | 13000 | 0.3515 | 3.6748 |
66
+ | 3.703 | 1.5094 | 14000 | 0.3535 | 3.6584 |
67
+ | 3.664 | 1.6173 | 15000 | 0.3555 | 3.6374 |
68
+ | 3.6639 | 1.7251 | 16000 | 0.3572 | 3.6206 |
69
+ | 3.6383 | 1.8329 | 17000 | 0.3587 | 3.6051 |
70
+ | 3.6321 | 1.9407 | 18000 | 0.3600 | 3.5912 |
71
+ | 3.5672 | 2.0485 | 19000 | 0.3615 | 3.5821 |
72
+ | 3.5754 | 2.1563 | 20000 | 0.3624 | 3.5733 |
73
+ | 3.5637 | 2.2642 | 21000 | 0.3637 | 3.5613 |
74
+ | 3.5706 | 2.3720 | 22000 | 0.3646 | 3.5504 |
75
+ | 3.5408 | 2.4798 | 23000 | 0.3660 | 3.5402 |
76
+ | 3.532 | 2.5876 | 24000 | 0.3669 | 3.5312 |
77
+ | 3.527 | 2.6954 | 25000 | 0.3673 | 3.5218 |
78
+ | 3.5333 | 2.8032 | 26000 | 0.3685 | 3.5120 |
79
+ | 3.5442 | 2.9111 | 27000 | 0.3696 | 3.5054 |
80
+ | 3.4451 | 3.0189 | 28000 | 0.3700 | 3.5012 |
81
+ | 3.4295 | 3.1267 | 29000 | 0.3709 | 3.4969 |
82
+ | 3.4506 | 3.2345 | 30000 | 0.3713 | 3.4902 |
83
+ | 3.4582 | 3.3423 | 31000 | 0.3719 | 3.4849 |
84
+ | 3.4597 | 3.4501 | 32000 | 0.3733 | 3.4757 |
85
+ | 3.4639 | 3.5580 | 33000 | 0.3736 | 3.4722 |
86
+ | 3.4602 | 3.6658 | 34000 | 0.3735 | 3.4653 |
87
+ | 3.4613 | 3.7736 | 35000 | 0.3749 | 3.4591 |
88
+ | 3.4517 | 3.8814 | 36000 | 0.3755 | 3.4511 |
89
+ | 3.4488 | 3.9892 | 37000 | 0.3761 | 3.4457 |
90
+ | 3.3644 | 4.0970 | 38000 | 0.3764 | 3.4490 |
91
+ | 3.3636 | 4.2049 | 39000 | 0.3773 | 3.4452 |
92
+ | 3.3899 | 4.3127 | 40000 | 0.3775 | 3.4419 |
93
+ | 3.3969 | 4.4205 | 41000 | 0.3782 | 3.4341 |
94
+ | 3.3848 | 4.5283 | 42000 | 0.3786 | 3.4308 |
95
+ | 3.3952 | 4.6361 | 43000 | 0.3788 | 3.4244 |
96
+ | 3.3875 | 4.7439 | 44000 | 0.3793 | 3.4211 |
97
+ | 3.4071 | 4.8518 | 45000 | 0.3799 | 3.4152 |
98
+ | 3.3742 | 4.9596 | 46000 | 0.3803 | 3.4095 |
99
+ | 3.3158 | 5.0674 | 47000 | 0.3807 | 3.4110 |
100
+ | 3.3119 | 5.1752 | 48000 | 0.3812 | 3.4111 |
101
+ | 3.3434 | 5.2830 | 49000 | 0.3813 | 3.4074 |
102
+ | 3.3422 | 5.3908 | 50000 | 0.3820 | 3.4001 |
103
+ | 3.3445 | 5.4987 | 51000 | 0.3820 | 3.3994 |
104
+ | 3.3169 | 5.6065 | 52000 | 0.3826 | 3.3929 |
105
+ | 3.333 | 5.7143 | 53000 | 0.3832 | 3.3878 |
106
+ | 3.331 | 5.8221 | 54000 | 0.3835 | 3.3844 |
107
+ | 3.3342 | 5.9299 | 55000 | 0.3840 | 3.3803 |
108
+ | 3.2353 | 6.0377 | 56000 | 0.3840 | 3.3844 |
109
+ | 3.2567 | 6.1456 | 57000 | 0.3843 | 3.3837 |
110
+ | 3.2751 | 6.2534 | 58000 | 0.3847 | 3.3811 |
111
+ | 3.2877 | 6.3612 | 59000 | 0.3850 | 3.3768 |
112
+ | 3.2882 | 6.4690 | 60000 | 0.3854 | 3.3732 |
113
+ | 3.2801 | 6.5768 | 61000 | 0.3860 | 3.3683 |
114
+ | 3.2991 | 6.6846 | 62000 | 0.3863 | 3.3624 |
115
+ | 3.2791 | 6.7925 | 63000 | 0.3866 | 3.3601 |
116
+ | 3.288 | 6.9003 | 64000 | 0.3871 | 3.3546 |
117
+ | 3.1829 | 7.0081 | 65000 | 0.3871 | 3.3573 |
118
+ | 3.2234 | 7.1159 | 66000 | 0.3873 | 3.3607 |
119
+ | 3.2282 | 7.2237 | 67000 | 0.3876 | 3.3572 |
120
+ | 3.2267 | 7.3315 | 68000 | 0.3878 | 3.3537 |
121
+ | 3.218 | 7.4394 | 69000 | 0.3883 | 3.3486 |
122
+ | 3.2235 | 7.5472 | 70000 | 0.3885 | 3.3458 |
123
+ | 3.2512 | 7.6550 | 71000 | 0.3889 | 3.3421 |
124
+ | 3.2503 | 7.7628 | 72000 | 0.3895 | 3.3384 |
125
+ | 3.2303 | 7.8706 | 73000 | 0.3901 | 3.3343 |
126
+ | 3.2506 | 7.9784 | 74000 | 0.3901 | 3.3303 |
127
+ | 3.155 | 8.0863 | 75000 | 0.3902 | 3.3376 |
128
+ | 3.1552 | 8.1941 | 76000 | 0.3902 | 3.3350 |
129
+ | 3.1737 | 8.3019 | 77000 | 0.3907 | 3.3308 |
130
+ | 3.167 | 8.4097 | 78000 | 0.3908 | 3.3294 |
131
+ | 3.1798 | 8.5175 | 79000 | 0.3912 | 3.3247 |
132
+ | 3.1957 | 8.6253 | 80000 | 0.3918 | 3.3219 |
133
+ | 3.1876 | 8.7332 | 81000 | 0.3921 | 3.3189 |
134
+ | 3.1825 | 8.8410 | 82000 | 0.3922 | 3.3150 |
135
+ | 3.1641 | 8.9488 | 83000 | 0.3927 | 3.3115 |
136
+ | 3.1278 | 9.0566 | 84000 | 0.3927 | 3.3139 |
137
+ | 3.1192 | 9.1644 | 85000 | 0.3928 | 3.3138 |
138
+ | 3.1443 | 9.2722 | 86000 | 0.3932 | 3.3118 |
139
+ | 3.1211 | 9.3801 | 87000 | 0.3934 | 3.3095 |
140
+ | 3.1296 | 9.4879 | 88000 | 0.3936 | 3.3073 |
141
+ | 3.1073 | 9.5957 | 89000 | 0.3939 | 3.3058 |
142
+ | 3.1362 | 9.7035 | 90000 | 0.3942 | 3.3025 |
143
+ | 3.1261 | 9.8113 | 91000 | 0.3944 | 3.3008 |
144
+ | 3.1183 | 9.9191 | 92000 | 0.3945 | 3.2993 |
145
+ | 3.3291 | 10.0270 | 93000 | 3.4618 | 0.3782 |
146
+ | 3.3625 | 10.1348 | 94000 | 3.4680 | 0.3760 |
147
+ | 3.3897 | 10.2426 | 95000 | 3.4628 | 0.3762 |
148
 
149
 
150
  ### Framework versions
all_results.json ADDED
@@ -0,0 +1,16 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 10.0,
3
+ "eval_accuracy": 0.3942061499490255,
4
+ "eval_loss": 3.302462100982666,
5
+ "eval_runtime": 185.2652,
6
+ "eval_samples": 18011,
7
+ "eval_samples_per_second": 97.217,
8
+ "eval_steps_per_second": 6.078,
9
+ "perplexity": 27.179475220683955,
10
+ "total_flos": 7.75449427968e+17,
11
+ "train_loss": 3.4548489175883907,
12
+ "train_runtime": 79989.2225,
13
+ "train_samples": 296775,
14
+ "train_samples_per_second": 37.102,
15
+ "train_steps_per_second": 1.16
16
+ }
eval_results.json ADDED
@@ -0,0 +1,10 @@
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 10.0,
3
+ "eval_accuracy": 0.3942061499490255,
4
+ "eval_loss": 3.302462100982666,
5
+ "eval_runtime": 185.2652,
6
+ "eval_samples": 18011,
7
+ "eval_samples_per_second": 97.217,
8
+ "eval_steps_per_second": 6.078,
9
+ "perplexity": 27.179475220683955
10
+ }
train_results.json ADDED
@@ -0,0 +1,9 @@
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "epoch": 10.0,
3
+ "total_flos": 7.75449427968e+17,
4
+ "train_loss": 3.4548489175883907,
5
+ "train_runtime": 79989.2225,
6
+ "train_samples": 296775,
7
+ "train_samples_per_second": 37.102,
8
+ "train_steps_per_second": 1.16
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 sha256:4128eda3e9c8c97041ff17245f83cc1f26210a065e6d51188d1e8a5776110948
3
  size 5304
 
1
  version https://git-lfs.github.com/spec/v1
2
+ oid sha256:8073d66d223082d7af2c859c8fc1d730cf6444a6538568483d6df796390a4a5a
3
  size 5304