<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
multiberts-seed_0_winobias_classifieronly
This model is a fine-tuned version of google/multiberts-seed_0 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.6937
- Accuracy: 0.5088
- Tp: 0.3011
- Tn: 0.2077
- Fp: 0.2923
- Fn: 0.1989
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 5e-05
- train_batch_size: 64
- eval_batch_size: 64
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 50
Training results
Training Loss | Epoch | Step | Validation Loss | Accuracy | Tp | Tn | Fp | Fn |
---|---|---|---|---|---|---|---|---|
0.6992 | 0.8 | 20 | 0.6945 | 0.4943 | 0.2285 | 0.2658 | 0.2342 | 0.2715 |
0.7005 | 1.6 | 40 | 0.6944 | 0.4937 | 0.2702 | 0.2235 | 0.2765 | 0.2298 |
0.7025 | 2.4 | 60 | 0.6958 | 0.5032 | 0.4457 | 0.0574 | 0.4426 | 0.0543 |
0.6937 | 3.2 | 80 | 0.6943 | 0.5044 | 0.3403 | 0.1641 | 0.3359 | 0.1597 |
0.6935 | 4.0 | 100 | 0.6942 | 0.4943 | 0.2847 | 0.2096 | 0.2904 | 0.2153 |
0.6995 | 4.8 | 120 | 0.6942 | 0.4962 | 0.2702 | 0.2260 | 0.2740 | 0.2298 |
0.7035 | 5.6 | 140 | 0.6955 | 0.5 | 0.4324 | 0.0676 | 0.4324 | 0.0676 |
0.7001 | 6.4 | 160 | 0.6941 | 0.5 | 0.2437 | 0.2563 | 0.2437 | 0.2563 |
0.6967 | 7.2 | 180 | 0.6941 | 0.5032 | 0.2418 | 0.2614 | 0.2386 | 0.2582 |
0.6934 | 8.0 | 200 | 0.6943 | 0.4987 | 0.3472 | 0.1515 | 0.3485 | 0.1528 |
0.6963 | 8.8 | 220 | 0.6941 | 0.4956 | 0.3112 | 0.1843 | 0.3157 | 0.1888 |
0.6962 | 9.6 | 240 | 0.6940 | 0.5006 | 0.3112 | 0.1894 | 0.3106 | 0.1888 |
0.6955 | 10.4 | 260 | 0.6944 | 0.4949 | 0.3580 | 0.1370 | 0.3630 | 0.1420 |
0.7032 | 11.2 | 280 | 0.6941 | 0.4912 | 0.2683 | 0.2229 | 0.2771 | 0.2317 |
0.6996 | 12.0 | 300 | 0.6943 | 0.4949 | 0.3580 | 0.1370 | 0.3630 | 0.1420 |
0.6995 | 12.8 | 320 | 0.6940 | 0.5032 | 0.3188 | 0.1843 | 0.3157 | 0.1812 |
0.6985 | 13.6 | 340 | 0.6939 | 0.5057 | 0.2330 | 0.2727 | 0.2273 | 0.2670 |
0.696 | 14.4 | 360 | 0.6947 | 0.5038 | 0.4211 | 0.0827 | 0.4173 | 0.0789 |
0.6988 | 15.2 | 380 | 0.6941 | 0.4937 | 0.2854 | 0.2083 | 0.2917 | 0.2146 |
0.6998 | 16.0 | 400 | 0.6942 | 0.5044 | 0.2191 | 0.2854 | 0.2146 | 0.2809 |
0.6971 | 16.8 | 420 | 0.6940 | 0.5063 | 0.2323 | 0.2740 | 0.2260 | 0.2677 |
0.704 | 17.6 | 440 | 0.6939 | 0.4937 | 0.2664 | 0.2273 | 0.2727 | 0.2336 |
0.695 | 18.4 | 460 | 0.6940 | 0.4956 | 0.2797 | 0.2159 | 0.2841 | 0.2203 |
0.6992 | 19.2 | 480 | 0.6940 | 0.4949 | 0.2872 | 0.2077 | 0.2923 | 0.2128 |
0.6869 | 20.0 | 500 | 0.6940 | 0.4880 | 0.3150 | 0.1730 | 0.3270 | 0.1850 |
0.7022 | 20.8 | 520 | 0.6939 | 0.4968 | 0.3239 | 0.1730 | 0.3270 | 0.1761 |
0.6961 | 21.6 | 540 | 0.6942 | 0.5019 | 0.3857 | 0.1162 | 0.3838 | 0.1143 |
0.7024 | 22.4 | 560 | 0.6940 | 0.4962 | 0.3453 | 0.1509 | 0.3491 | 0.1547 |
0.6995 | 23.2 | 580 | 0.6940 | 0.5019 | 0.3542 | 0.1477 | 0.3523 | 0.1458 |
0.7 | 24.0 | 600 | 0.6939 | 0.5019 | 0.3131 | 0.1888 | 0.3112 | 0.1869 |
0.7035 | 24.8 | 620 | 0.6939 | 0.5 | 0.3396 | 0.1604 | 0.3396 | 0.1604 |
0.6978 | 25.6 | 640 | 0.6938 | 0.5133 | 0.2064 | 0.3068 | 0.1932 | 0.2936 |
0.7032 | 26.4 | 660 | 0.6938 | 0.4981 | 0.2399 | 0.2582 | 0.2418 | 0.2601 |
0.6898 | 27.2 | 680 | 0.6938 | 0.4956 | 0.2449 | 0.2506 | 0.2494 | 0.2551 |
0.6923 | 28.0 | 700 | 0.6939 | 0.4924 | 0.2929 | 0.1995 | 0.3005 | 0.2071 |
0.7025 | 28.8 | 720 | 0.6938 | 0.4968 | 0.2809 | 0.2159 | 0.2841 | 0.2191 |
0.6999 | 29.6 | 740 | 0.6943 | 0.5025 | 0.3838 | 0.1187 | 0.3813 | 0.1162 |
0.6993 | 30.4 | 760 | 0.6940 | 0.5006 | 0.3573 | 0.1433 | 0.3567 | 0.1427 |
0.7013 | 31.2 | 780 | 0.6939 | 0.4968 | 0.3333 | 0.1635 | 0.3365 | 0.1667 |
0.6985 | 32.0 | 800 | 0.6939 | 0.4981 | 0.3460 | 0.1521 | 0.3479 | 0.1540 |
0.6957 | 32.8 | 820 | 0.6938 | 0.5076 | 0.3277 | 0.1799 | 0.3201 | 0.1723 |
0.6978 | 33.6 | 840 | 0.6937 | 0.5107 | 0.2475 | 0.2633 | 0.2367 | 0.2525 |
0.6953 | 34.4 | 860 | 0.6938 | 0.5133 | 0.3201 | 0.1932 | 0.3068 | 0.1799 |
0.6923 | 35.2 | 880 | 0.6939 | 0.5 | 0.3485 | 0.1515 | 0.3485 | 0.1515 |
0.6947 | 36.0 | 900 | 0.6939 | 0.4987 | 0.3491 | 0.1496 | 0.3504 | 0.1509 |
0.7018 | 36.8 | 920 | 0.6937 | 0.5095 | 0.3018 | 0.2077 | 0.2923 | 0.1982 |
0.6957 | 37.6 | 940 | 0.6939 | 0.4975 | 0.3403 | 0.1572 | 0.3428 | 0.1597 |
0.698 | 38.4 | 960 | 0.6939 | 0.4981 | 0.3415 | 0.1566 | 0.3434 | 0.1585 |
0.6902 | 39.2 | 980 | 0.6938 | 0.4968 | 0.3289 | 0.1679 | 0.3321 | 0.1711 |
0.6973 | 40.0 | 1000 | 0.6938 | 0.4956 | 0.3346 | 0.1610 | 0.3390 | 0.1654 |
0.6962 | 40.8 | 1020 | 0.6938 | 0.5025 | 0.3194 | 0.1831 | 0.3169 | 0.1806 |
0.699 | 41.6 | 1040 | 0.6937 | 0.5019 | 0.3093 | 0.1926 | 0.3074 | 0.1907 |
0.6965 | 42.4 | 1060 | 0.6937 | 0.5013 | 0.2816 | 0.2197 | 0.2803 | 0.2184 |
0.694 | 43.2 | 1080 | 0.6937 | 0.5013 | 0.2677 | 0.2336 | 0.2664 | 0.2323 |
0.699 | 44.0 | 1100 | 0.6937 | 0.4962 | 0.2664 | 0.2298 | 0.2702 | 0.2336 |
0.6932 | 44.8 | 1120 | 0.6937 | 0.4975 | 0.2645 | 0.2330 | 0.2670 | 0.2355 |
0.6994 | 45.6 | 1140 | 0.6937 | 0.5006 | 0.2696 | 0.2311 | 0.2689 | 0.2304 |
0.7008 | 46.4 | 1160 | 0.6937 | 0.5095 | 0.2992 | 0.2102 | 0.2898 | 0.2008 |
0.7006 | 47.2 | 1180 | 0.6937 | 0.5082 | 0.3049 | 0.2033 | 0.2967 | 0.1951 |
0.6954 | 48.0 | 1200 | 0.6937 | 0.5101 | 0.3030 | 0.2071 | 0.2929 | 0.1970 |
0.7044 | 48.8 | 1220 | 0.6937 | 0.5088 | 0.3030 | 0.2058 | 0.2942 | 0.1970 |
0.6993 | 49.6 | 1240 | 0.6937 | 0.5088 | 0.3011 | 0.2077 | 0.2923 | 0.1989 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.13.1
- Datasets 2.10.1
- Tokenizers 0.13.2