<!-- 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. -->
20230825003351
This model is a fine-tuned version of bert-large-cased on the super_glue dataset. It achieves the following results on the evaluation set:
- Loss: 0.8357
- Accuracy: 0.7545
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: 0.005
- train_batch_size: 16
- eval_batch_size: 8
- seed: 11
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 80.0
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| No log | 1.0 | 156 | 0.6981 | 0.5307 |
| No log | 2.0 | 312 | 1.0094 | 0.4729 |
| No log | 3.0 | 468 | 0.9523 | 0.4729 |
| 0.9081 | 4.0 | 624 | 0.6610 | 0.5487 |
| 0.9081 | 5.0 | 780 | 0.8823 | 0.5415 |
| 0.9081 | 6.0 | 936 | 0.6125 | 0.6318 |
| 0.8417 | 7.0 | 1092 | 1.1255 | 0.6282 |
| 0.8417 | 8.0 | 1248 | 1.1191 | 0.5487 |
| 0.8417 | 9.0 | 1404 | 0.7418 | 0.6318 |
| 0.7229 | 10.0 | 1560 | 0.6357 | 0.6859 |
| 0.7229 | 11.0 | 1716 | 0.6525 | 0.6354 |
| 0.7229 | 12.0 | 1872 | 0.9733 | 0.6426 |
| 0.6627 | 13.0 | 2028 | 0.7120 | 0.6715 |
| 0.6627 | 14.0 | 2184 | 1.0585 | 0.6606 |
| 0.6627 | 15.0 | 2340 | 0.6682 | 0.6751 |
| 0.6627 | 16.0 | 2496 | 1.0978 | 0.6462 |
| 0.6611 | 17.0 | 2652 | 0.5480 | 0.7256 |
| 0.6611 | 18.0 | 2808 | 0.5491 | 0.7220 |
| 0.6611 | 19.0 | 2964 | 0.5740 | 0.7112 |
| 0.5716 | 20.0 | 3120 | 0.7281 | 0.6751 |
| 0.5716 | 21.0 | 3276 | 0.6144 | 0.6931 |
| 0.5716 | 22.0 | 3432 | 0.5663 | 0.7076 |
| 0.53 | 23.0 | 3588 | 0.6161 | 0.7329 |
| 0.53 | 24.0 | 3744 | 0.7898 | 0.6968 |
| 0.53 | 25.0 | 3900 | 0.9875 | 0.6715 |
| 0.5203 | 26.0 | 4056 | 0.5164 | 0.7329 |
| 0.5203 | 27.0 | 4212 | 0.5534 | 0.7473 |
| 0.5203 | 28.0 | 4368 | 0.6047 | 0.7473 |
| 0.4624 | 29.0 | 4524 | 0.6346 | 0.7292 |
| 0.4624 | 30.0 | 4680 | 0.8954 | 0.7040 |
| 0.4624 | 31.0 | 4836 | 0.6913 | 0.7004 |
| 0.4624 | 32.0 | 4992 | 0.6815 | 0.7401 |
| 0.4178 | 33.0 | 5148 | 0.6964 | 0.7220 |
| 0.4178 | 34.0 | 5304 | 0.6707 | 0.7184 |
| 0.4178 | 35.0 | 5460 | 0.6211 | 0.7581 |
| 0.4111 | 36.0 | 5616 | 0.7246 | 0.7329 |
| 0.4111 | 37.0 | 5772 | 0.8112 | 0.7401 |
| 0.4111 | 38.0 | 5928 | 0.8703 | 0.7220 |
| 0.3451 | 39.0 | 6084 | 0.6927 | 0.7220 |
| 0.3451 | 40.0 | 6240 | 0.8796 | 0.7292 |
| 0.3451 | 41.0 | 6396 | 0.7368 | 0.7329 |
| 0.3129 | 42.0 | 6552 | 0.7534 | 0.7365 |
| 0.3129 | 43.0 | 6708 | 0.8497 | 0.7040 |
| 0.3129 | 44.0 | 6864 | 0.6551 | 0.7581 |
| 0.3028 | 45.0 | 7020 | 0.9768 | 0.7076 |
| 0.3028 | 46.0 | 7176 | 0.7248 | 0.7509 |
| 0.3028 | 47.0 | 7332 | 0.7248 | 0.7473 |
| 0.3028 | 48.0 | 7488 | 0.8685 | 0.7581 |
| 0.2749 | 49.0 | 7644 | 0.7781 | 0.7437 |
| 0.2749 | 50.0 | 7800 | 0.8690 | 0.7365 |
| 0.2749 | 51.0 | 7956 | 1.0267 | 0.7473 |
| 0.2695 | 52.0 | 8112 | 0.7892 | 0.7437 |
| 0.2695 | 53.0 | 8268 | 0.8535 | 0.7256 |
| 0.2695 | 54.0 | 8424 | 0.7670 | 0.7473 |
| 0.2482 | 55.0 | 8580 | 0.7970 | 0.7509 |
| 0.2482 | 56.0 | 8736 | 0.9177 | 0.7473 |
| 0.2482 | 57.0 | 8892 | 0.9749 | 0.7437 |
| 0.2201 | 58.0 | 9048 | 0.7507 | 0.7256 |
| 0.2201 | 59.0 | 9204 | 0.9523 | 0.7401 |
| 0.2201 | 60.0 | 9360 | 0.7579 | 0.7545 |
| 0.2158 | 61.0 | 9516 | 1.0789 | 0.7437 |
| 0.2158 | 62.0 | 9672 | 0.9747 | 0.7437 |
| 0.2158 | 63.0 | 9828 | 1.1461 | 0.7220 |
| 0.2158 | 64.0 | 9984 | 0.8201 | 0.7509 |
| 0.2003 | 65.0 | 10140 | 0.9460 | 0.7365 |
| 0.2003 | 66.0 | 10296 | 0.9926 | 0.7617 |
| 0.2003 | 67.0 | 10452 | 1.0429 | 0.7329 |
| 0.1944 | 68.0 | 10608 | 0.9319 | 0.7401 |
| 0.1944 | 69.0 | 10764 | 0.8939 | 0.7545 |
| 0.1944 | 70.0 | 10920 | 0.9424 | 0.7473 |
| 0.179 | 71.0 | 11076 | 0.8541 | 0.7545 |
| 0.179 | 72.0 | 11232 | 0.9214 | 0.7581 |
| 0.179 | 73.0 | 11388 | 0.9028 | 0.7473 |
| 0.1751 | 74.0 | 11544 | 0.9283 | 0.7509 |
| 0.1751 | 75.0 | 11700 | 0.8972 | 0.7473 |
| 0.1751 | 76.0 | 11856 | 0.8062 | 0.7581 |
| 0.1661 | 77.0 | 12012 | 0.8653 | 0.7473 |
| 0.1661 | 78.0 | 12168 | 0.8238 | 0.7653 |
| 0.1661 | 79.0 | 12324 | 0.8519 | 0.7473 |
| 0.1661 | 80.0 | 12480 | 0.8357 | 0.7545 |
Framework versions
- Transformers 4.26.1
- Pytorch 2.0.1+cu118
- Datasets 2.12.0
- Tokenizers 0.13.3
