<!-- 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. -->
20230824164344
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: 0.9440
- Accuracy: 0.7329
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 | 0.7410 | 0.5162 |
| No log | 2.0 | 312 | 1.0443 | 0.4729 |
| No log | 3.0 | 468 | 0.6773 | 0.5054 |
| 0.9803 | 4.0 | 624 | 0.8278 | 0.5343 |
| 0.9803 | 5.0 | 780 | 0.6367 | 0.6137 |
| 0.9803 | 6.0 | 936 | 0.6217 | 0.6426 |
| 0.8339 | 7.0 | 1092 | 1.2109 | 0.5776 |
| 0.8339 | 8.0 | 1248 | 0.5718 | 0.6859 |
| 0.8339 | 9.0 | 1404 | 0.7100 | 0.6606 |
| 0.7334 | 10.0 | 1560 | 1.3794 | 0.5993 |
| 0.7334 | 11.0 | 1716 | 0.7077 | 0.5668 |
| 0.7334 | 12.0 | 1872 | 0.5683 | 0.7040 |
| 0.6828 | 13.0 | 2028 | 0.5391 | 0.7329 |
| 0.6828 | 14.0 | 2184 | 0.7041 | 0.7292 |
| 0.6828 | 15.0 | 2340 | 0.7170 | 0.6679 |
| 0.6828 | 16.0 | 2496 | 1.1745 | 0.6029 |
| 0.622 | 17.0 | 2652 | 0.6299 | 0.7112 |
| 0.622 | 18.0 | 2808 | 0.5566 | 0.7437 |
| 0.622 | 19.0 | 2964 | 0.5614 | 0.7509 |
| 0.5718 | 20.0 | 3120 | 1.6971 | 0.6390 |
| 0.5718 | 21.0 | 3276 | 0.6663 | 0.7076 |
| 0.5718 | 22.0 | 3432 | 0.6859 | 0.6498 |
| 0.5554 | 23.0 | 3588 | 0.7722 | 0.7112 |
| 0.5554 | 24.0 | 3744 | 0.6040 | 0.7256 |
| 0.5554 | 25.0 | 3900 | 0.8333 | 0.7329 |
| 0.4565 | 26.0 | 4056 | 0.5782 | 0.7220 |
| 0.4565 | 27.0 | 4212 | 0.6536 | 0.6968 |
| 0.4565 | 28.0 | 4368 | 0.8468 | 0.7292 |
| 0.4326 | 29.0 | 4524 | 0.7304 | 0.7148 |
| 0.4326 | 30.0 | 4680 | 0.8690 | 0.6968 |
| 0.4326 | 31.0 | 4836 | 0.8080 | 0.7148 |
| 0.4326 | 32.0 | 4992 | 0.6306 | 0.7292 |
| 0.3528 | 33.0 | 5148 | 0.8862 | 0.7220 |
| 0.3528 | 34.0 | 5304 | 0.8333 | 0.7365 |
| 0.3528 | 35.0 | 5460 | 0.6612 | 0.7329 |
| 0.3155 | 36.0 | 5616 | 0.7407 | 0.7401 |
| 0.3155 | 37.0 | 5772 | 0.8019 | 0.7365 |
| 0.3155 | 38.0 | 5928 | 0.9540 | 0.7401 |
| 0.2632 | 39.0 | 6084 | 0.9973 | 0.7365 |
| 0.2632 | 40.0 | 6240 | 0.7745 | 0.7401 |
| 0.2632 | 41.0 | 6396 | 0.7636 | 0.7473 |
| 0.2516 | 42.0 | 6552 | 0.8117 | 0.7401 |
| 0.2516 | 43.0 | 6708 | 0.8688 | 0.7329 |
| 0.2516 | 44.0 | 6864 | 0.8390 | 0.7509 |
| 0.219 | 45.0 | 7020 | 0.9181 | 0.7401 |
| 0.219 | 46.0 | 7176 | 0.8596 | 0.7509 |
| 0.219 | 47.0 | 7332 | 0.9130 | 0.7437 |
| 0.219 | 48.0 | 7488 | 0.9129 | 0.7437 |
| 0.2039 | 49.0 | 7644 | 0.7271 | 0.7545 |
| 0.2039 | 50.0 | 7800 | 0.8405 | 0.7437 |
| 0.2039 | 51.0 | 7956 | 0.8249 | 0.7653 |
| 0.1809 | 52.0 | 8112 | 0.8916 | 0.7581 |
| 0.1809 | 53.0 | 8268 | 0.9851 | 0.7437 |
| 0.1809 | 54.0 | 8424 | 0.8449 | 0.7653 |
| 0.1588 | 55.0 | 8580 | 0.8400 | 0.7437 |
| 0.1588 | 56.0 | 8736 | 0.9869 | 0.7473 |
| 0.1588 | 57.0 | 8892 | 0.7289 | 0.7509 |
| 0.1563 | 58.0 | 9048 | 0.9168 | 0.7437 |
| 0.1563 | 59.0 | 9204 | 1.0048 | 0.7401 |
| 0.1563 | 60.0 | 9360 | 0.9174 | 0.7581 |
| 0.1434 | 61.0 | 9516 | 1.0328 | 0.7437 |
| 0.1434 | 62.0 | 9672 | 0.9543 | 0.7509 |
| 0.1434 | 63.0 | 9828 | 0.9841 | 0.7509 |
| 0.1434 | 64.0 | 9984 | 0.9057 | 0.7509 |
| 0.1345 | 65.0 | 10140 | 0.9597 | 0.7509 |
| 0.1345 | 66.0 | 10296 | 0.9686 | 0.7509 |
| 0.1345 | 67.0 | 10452 | 0.9621 | 0.7581 |
| 0.1363 | 68.0 | 10608 | 1.0869 | 0.7292 |
| 0.1363 | 69.0 | 10764 | 1.0265 | 0.7365 |
| 0.1363 | 70.0 | 10920 | 0.9629 | 0.7509 |
| 0.1166 | 71.0 | 11076 | 0.8672 | 0.7509 |
| 0.1166 | 72.0 | 11232 | 0.9515 | 0.7401 |
| 0.1166 | 73.0 | 11388 | 0.9453 | 0.7401 |
| 0.1196 | 74.0 | 11544 | 0.9168 | 0.7473 |
| 0.1196 | 75.0 | 11700 | 0.9455 | 0.7437 |
| 0.1196 | 76.0 | 11856 | 0.9246 | 0.7437 |
| 0.1184 | 77.0 | 12012 | 1.0048 | 0.7329 |
| 0.1184 | 78.0 | 12168 | 0.9510 | 0.7329 |
| 0.1184 | 79.0 | 12324 | 0.9356 | 0.7365 |
| 0.1184 | 80.0 | 12480 | 0.9440 | 0.7329 |
Framework versions
- Transformers 4.26.1
- Pytorch 2.0.1+cu118
- Datasets 2.12.0
- Tokenizers 0.13.3
