Hugging Face's logo

henryscheible
/
gpt2_crows_pairs_finetuned

Text Classification
Transformers
PyTorch
gpt2
generated_from_trainer
Eval Results
Inference Endpoints
text-generation-inference
Model card Files Community
1
henryscheible commited on
Commit
8ba18fe
•
1 Parent(s): 5a2aa88

update model card README.md

Browse files
Files changed (1) hide show
  1. README.md +55 -55
README.md CHANGED
@@ -21,7 +21,7 @@ model-index:
21
  metrics:
22
  - name: Accuracy
23
  type: accuracy
24
- value: 0.5099337748344371
25
  ---
26
 
27
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
@@ -31,12 +31,12 @@ should probably proofread and complete it, then remove this comment. -->
31
 
32
  This model is a fine-tuned version of [gpt2](https://huggingface.co/gpt2) on the crows_pairs dataset.
33
  It achieves the following results on the evaluation set:
34
- - Loss: 4.2700
35
- - Accuracy: 0.5099
36
- - Tp: 0.2848
37
- - Tn: 0.2252
38
- - Fp: 0.3013
39
- - Fn: 0.1887
40
 
41
  ## Model description
42
 
@@ -55,7 +55,7 @@ More information needed
55
  ### Training hyperparameters
56
 
57
  The following hyperparameters were used during training:
58
- - learning_rate: 0.0001
59
  - train_batch_size: 64
60
  - eval_batch_size: 64
61
  - seed: 42
@@ -67,53 +67,53 @@ The following hyperparameters were used during training:
67
 
68
  | Training Loss | Epoch | Step | Validation Loss | Accuracy | Tp | Tn | Fp | Fn |
69
  |:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|:------:|:------:|:------:|
70
- | 0.7239 | 1.05 | 20 | 0.7117 | 0.4735 | 0.4735 | 0.0 | 0.5265 | 0.0 |
71
- | 0.6895 | 2.11 | 40 | 0.6892 | 0.5364 | 0.0099 | 0.5265 | 0.0 | 0.4636 |
72
- | 0.5804 | 3.16 | 60 | 0.9994 | 0.5331 | 0.3609 | 0.1722 | 0.3543 | 0.1126 |
73
- | 0.2737 | 4.21 | 80 | 1.6220 | 0.5099 | 0.1954 | 0.3146 | 0.2119 | 0.2781 |
74
- | 0.0966 | 5.26 | 100 | 2.0755 | 0.5364 | 0.1887 | 0.3477 | 0.1788 | 0.2848 |
75
- | 0.254 | 6.32 | 120 | 1.6116 | 0.5099 | 0.2616 | 0.2483 | 0.2781 | 0.2119 |
76
- | 0.0155 | 7.37 | 140 | 2.8087 | 0.5066 | 0.1689 | 0.3377 | 0.1887 | 0.3046 |
77
- | 0.0526 | 8.42 | 160 | 1.9181 | 0.5364 | 0.2417 | 0.2947 | 0.2318 | 0.2318 |
78
- | 0.0211 | 9.47 | 180 | 2.8951 | 0.5298 | 0.2318 | 0.2980 | 0.2285 | 0.2417 |
79
- | 0.0247 | 10.53 | 200 | 2.5335 | 0.5497 | 0.1987 | 0.3510 | 0.1755 | 0.2748 |
80
- | 0.0569 | 11.58 | 220 | 3.0074 | 0.5331 | 0.2682 | 0.2649 | 0.2616 | 0.2053 |
81
- | 0.0063 | 12.63 | 240 | 2.8685 | 0.5166 | 0.1987 | 0.3179 | 0.2086 | 0.2748 |
82
- | 0.015 | 13.68 | 260 | 3.4521 | 0.5298 | 0.3013 | 0.2285 | 0.2980 | 0.1722 |
83
- | 0.0092 | 14.74 | 280 | 3.7708 | 0.5132 | 0.2417 | 0.2715 | 0.2550 | 0.2318 |
84
- | 0.0404 | 15.79 | 300 | 3.5495 | 0.5298 | 0.2517 | 0.2781 | 0.2483 | 0.2219 |
85
- | 0.0005 | 16.84 | 320 | 2.9925 | 0.5728 | 0.2152 | 0.3576 | 0.1689 | 0.2583 |
86
- | 0.0002 | 17.89 | 340 | 3.8998 | 0.5099 | 0.3113 | 0.1987 | 0.3278 | 0.1623 |
87
- | 0.0 | 18.95 | 360 | 4.1014 | 0.5066 | 0.2815 | 0.2252 | 0.3013 | 0.1921 |
88
- | 0.0123 | 20.0 | 380 | 4.1120 | 0.5265 | 0.2914 | 0.2351 | 0.2914 | 0.1821 |
89
- | 0.0002 | 21.05 | 400 | 3.3291 | 0.5265 | 0.3079 | 0.2185 | 0.3079 | 0.1656 |
90
- | 0.0125 | 22.11 | 420 | 2.6652 | 0.5662 | 0.2417 | 0.3245 | 0.2020 | 0.2318 |
91
- | 0.0003 | 23.16 | 440 | 3.1833 | 0.5166 | 0.2980 | 0.2185 | 0.3079 | 0.1755 |
92
- | 0.0001 | 24.21 | 460 | 3.6840 | 0.5132 | 0.2947 | 0.2185 | 0.3079 | 0.1788 |
93
- | 0.0001 | 25.26 | 480 | 3.9037 | 0.5132 | 0.2914 | 0.2219 | 0.3046 | 0.1821 |
94
- | 0.0 | 26.32 | 500 | 3.7120 | 0.5364 | 0.2318 | 0.3046 | 0.2219 | 0.2417 |
95
- | 0.0001 | 27.37 | 520 | 3.8126 | 0.5331 | 0.2219 | 0.3113 | 0.2152 | 0.2517 |
96
- | 0.0001 | 28.42 | 540 | 4.0995 | 0.5265 | 0.2682 | 0.2583 | 0.2682 | 0.2053 |
97
- | 0.0 | 29.47 | 560 | 4.0544 | 0.5265 | 0.2517 | 0.2748 | 0.2517 | 0.2219 |
98
- | 0.0 | 30.53 | 580 | 4.0941 | 0.5232 | 0.2450 | 0.2781 | 0.2483 | 0.2285 |
99
- | 0.0 | 31.58 | 600 | 4.1695 | 0.5232 | 0.2450 | 0.2781 | 0.2483 | 0.2285 |
100
- | 0.0026 | 32.63 | 620 | 4.2884 | 0.5166 | 0.2914 | 0.2252 | 0.3013 | 0.1821 |
101
- | 0.0 | 33.68 | 640 | 4.2271 | 0.5033 | 0.2550 | 0.2483 | 0.2781 | 0.2185 |
102
- | 0.0 | 34.74 | 660 | 4.2139 | 0.5166 | 0.2483 | 0.2682 | 0.2583 | 0.2252 |
103
- | 0.0 | 35.79 | 680 | 4.2499 | 0.5166 | 0.2483 | 0.2682 | 0.2583 | 0.2252 |
104
- | 0.0 | 36.84 | 700 | 4.2932 | 0.5132 | 0.2483 | 0.2649 | 0.2616 | 0.2252 |
105
- | 0.0 | 37.89 | 720 | 4.3296 | 0.5132 | 0.2483 | 0.2649 | 0.2616 | 0.2252 |
106
- | 0.0 | 38.95 | 740 | 4.3720 | 0.5166 | 0.2483 | 0.2682 | 0.2583 | 0.2252 |
107
- | 0.0 | 40.0 | 760 | 4.4985 | 0.5033 | 0.2781 | 0.2252 | 0.3013 | 0.1954 |
108
- | 0.0 | 41.05 | 780 | 4.5338 | 0.5099 | 0.2980 | 0.2119 | 0.3146 | 0.1755 |
109
- | 0.0 | 42.11 | 800 | 4.5692 | 0.5099 | 0.2980 | 0.2119 | 0.3146 | 0.1755 |
110
- | 0.0 | 43.16 | 820 | 4.5827 | 0.5099 | 0.2980 | 0.2119 | 0.3146 | 0.1755 |
111
- | 0.0132 | 44.21 | 840 | 4.3863 | 0.5232 | 0.3113 | 0.2119 | 0.3146 | 0.1623 |
112
- | 0.0 | 45.26 | 860 | 4.2997 | 0.5166 | 0.3079 | 0.2086 | 0.3179 | 0.1656 |
113
- | 0.0 | 46.32 | 880 | 4.3073 | 0.5166 | 0.3013 | 0.2152 | 0.3113 | 0.1722 |
114
- | 0.0094 | 47.37 | 900 | 4.3083 | 0.5066 | 0.2914 | 0.2152 | 0.3113 | 0.1821 |
115
- | 0.0 | 48.42 | 920 | 4.2906 | 0.5033 | 0.2848 | 0.2185 | 0.3079 | 0.1887 |
116
- | 0.0 | 49.47 | 940 | 4.2700 | 0.5099 | 0.2848 | 0.2252 | 0.3013 | 0.1887 |
117
 
118
 
119
  ### Framework versions
 
21
  metrics:
22
  - name: Accuracy
23
  type: accuracy
24
+ value: 0.5364238410596026
25
  ---
26
 
27
  <!-- This model card has been generated automatically according to the information the Trainer had access to. You
 
31
 
32
  This model is a fine-tuned version of [gpt2](https://huggingface.co/gpt2) on the crows_pairs dataset.
33
  It achieves the following results on the evaluation set:
34
+ - Loss: 4.7088
35
+ - Accuracy: 0.5364
36
+ - Tp: 0.2384
37
+ - Tn: 0.2980
38
+ - Fp: 0.2185
39
+ - Fn: 0.2450
40
 
41
  ## Model description
42
 
 
55
  ### Training hyperparameters
56
 
57
  The following hyperparameters were used during training:
58
+ - learning_rate: 5e-05
59
  - train_batch_size: 64
60
  - eval_batch_size: 64
61
  - seed: 42
 
67
 
68
  | Training Loss | Epoch | Step | Validation Loss | Accuracy | Tp | Tn | Fp | Fn |
69
  |:-------------:|:-----:|:----:|:---------------:|:--------:|:------:|:------:|:------:|:------:|
70
+ | 0.861 | 1.05 | 20 | 0.7012 | 0.4834 | 0.4834 | 0.0 | 0.5166 | 0.0 |
71
+ | 0.8018 | 2.11 | 40 | 0.8255 | 0.4834 | 0.4834 | 0.0 | 0.5166 | 0.0 |
72
+ | 0.7346 | 3.16 | 60 | 0.7108 | 0.4834 | 0.4834 | 0.0 | 0.5166 | 0.0 |
73
+ | 0.556 | 4.21 | 80 | 0.9991 | 0.5232 | 0.1424 | 0.3808 | 0.1358 | 0.3411 |
74
+ | 0.3663 | 5.26 | 100 | 1.0435 | 0.5596 | 0.2715 | 0.2881 | 0.2285 | 0.2119 |
75
+ | 0.2264 | 6.32 | 120 | 1.2216 | 0.5662 | 0.2417 | 0.3245 | 0.1921 | 0.2417 |
76
+ | 0.1094 | 7.37 | 140 | 1.6917 | 0.5430 | 0.3046 | 0.2384 | 0.2781 | 0.1788 |
77
+ | 0.0548 | 8.42 | 160 | 3.3528 | 0.5331 | 0.2682 | 0.2649 | 0.2517 | 0.2152 |
78
+ | 0.0991 | 9.47 | 180 | 1.8176 | 0.5430 | 0.2417 | 0.3013 | 0.2152 | 0.2417 |
79
+ | 0.0292 | 10.53 | 200 | 2.6706 | 0.5232 | 0.1821 | 0.3411 | 0.1755 | 0.3013 |
80
+ | 0.0187 | 11.58 | 220 | 3.8548 | 0.5132 | 0.1854 | 0.3278 | 0.1887 | 0.2980 |
81
+ | 0.0149 | 12.63 | 240 | 3.9376 | 0.5298 | 0.2351 | 0.2947 | 0.2219 | 0.2483 |
82
+ | 0.0025 | 13.68 | 260 | 3.7839 | 0.5430 | 0.1755 | 0.3675 | 0.1490 | 0.3079 |
83
+ | 0.0338 | 14.74 | 280 | 3.0663 | 0.5596 | 0.1954 | 0.3642 | 0.1523 | 0.2881 |
84
+ | 0.004 | 15.79 | 300 | 3.3492 | 0.5397 | 0.2550 | 0.2848 | 0.2318 | 0.2285 |
85
+ | 0.0003 | 16.84 | 320 | 4.1201 | 0.5629 | 0.1887 | 0.3742 | 0.1424 | 0.2947 |
86
+ | 0.0185 | 17.89 | 340 | 3.9285 | 0.5563 | 0.1954 | 0.3609 | 0.1556 | 0.2881 |
87
+ | 0.0001 | 18.95 | 360 | 3.7377 | 0.5430 | 0.2550 | 0.2881 | 0.2285 | 0.2285 |
88
+ | 0.0082 | 20.0 | 380 | 3.6933 | 0.5662 | 0.2815 | 0.2848 | 0.2318 | 0.2020 |
89
+ | 0.0049 | 21.05 | 400 | 3.8695 | 0.5629 | 0.2649 | 0.2980 | 0.2185 | 0.2185 |
90
+ | 0.0129 | 22.11 | 420 | 4.2052 | 0.5199 | 0.2185 | 0.3013 | 0.2152 | 0.2649 |
91
+ | 0.0158 | 23.16 | 440 | 3.9680 | 0.5397 | 0.2781 | 0.2616 | 0.2550 | 0.2053 |
92
+ | 0.0009 | 24.21 | 460 | 3.8547 | 0.5596 | 0.1987 | 0.3609 | 0.1556 | 0.2848 |
93
+ | 0.0013 | 25.26 | 480 | 4.2549 | 0.5430 | 0.2517 | 0.2914 | 0.2252 | 0.2318 |
94
+ | 0.0 | 26.32 | 500 | 4.4493 | 0.5497 | 0.2583 | 0.2914 | 0.2252 | 0.2252 |
95
+ | 0.0018 | 27.37 | 520 | 4.6134 | 0.5662 | 0.2384 | 0.3278 | 0.1887 | 0.2450 |
96
+ | 0.0021 | 28.42 | 540 | 4.4612 | 0.5166 | 0.2450 | 0.2715 | 0.2450 | 0.2384 |
97
+ | 0.0 | 29.47 | 560 | 4.5389 | 0.5331 | 0.2086 | 0.3245 | 0.1921 | 0.2748 |
98
+ | 0.0 | 30.53 | 580 | 4.3807 | 0.5629 | 0.2947 | 0.2682 | 0.2483 | 0.1887 |
99
+ | 0.0042 | 31.58 | 600 | 4.6153 | 0.5298 | 0.2318 | 0.2980 | 0.2185 | 0.2517 |
100
+ | 0.0 | 32.63 | 620 | 4.7666 | 0.5298 | 0.2020 | 0.3278 | 0.1887 | 0.2815 |
101
+ | 0.0099 | 33.68 | 640 | 4.8298 | 0.5298 | 0.2053 | 0.3245 | 0.1921 | 0.2781 |
102
+ | 0.0 | 34.74 | 660 | 4.6201 | 0.5364 | 0.2219 | 0.3146 | 0.2020 | 0.2616 |
103
+ | 0.0 | 35.79 | 680 | 4.6495 | 0.5364 | 0.2185 | 0.3179 | 0.1987 | 0.2649 |
104
+ | 0.0005 | 36.84 | 700 | 4.5861 | 0.5464 | 0.2450 | 0.3013 | 0.2152 | 0.2384 |
105
+ | 0.0 | 37.89 | 720 | 4.8236 | 0.5331 | 0.2020 | 0.3311 | 0.1854 | 0.2815 |
106
+ | 0.0 | 38.95 | 740 | 4.8646 | 0.5364 | 0.2020 | 0.3344 | 0.1821 | 0.2815 |
107
+ | 0.0 | 40.0 | 760 | 4.8922 | 0.5331 | 0.2020 | 0.3311 | 0.1854 | 0.2815 |
108
+ | 0.0 | 41.05 | 780 | 4.7690 | 0.5430 | 0.2318 | 0.3113 | 0.2053 | 0.2517 |
109
+ | 0.0 | 42.11 | 800 | 4.7544 | 0.5397 | 0.2351 | 0.3046 | 0.2119 | 0.2483 |
110
+ | 0.0 | 43.16 | 820 | 4.8132 | 0.5464 | 0.2351 | 0.3113 | 0.2053 | 0.2483 |
111
+ | 0.0 | 44.21 | 840 | 4.8464 | 0.5430 | 0.2318 | 0.3113 | 0.2053 | 0.2517 |
112
+ | 0.0129 | 45.26 | 860 | 4.6306 | 0.5497 | 0.2682 | 0.2815 | 0.2351 | 0.2152 |
113
+ | 0.0 | 46.32 | 880 | 4.6495 | 0.5364 | 0.2450 | 0.2914 | 0.2252 | 0.2384 |
114
+ | 0.0 | 47.37 | 900 | 4.6940 | 0.5364 | 0.2384 | 0.2980 | 0.2185 | 0.2450 |
115
+ | 0.0 | 48.42 | 920 | 4.7050 | 0.5364 | 0.2384 | 0.2980 | 0.2185 | 0.2450 |
116
+ | 0.0 | 49.47 | 940 | 4.7088 | 0.5364 | 0.2384 | 0.2980 | 0.2185 | 0.2450 |
117
 
118
 
119
  ### Framework versions