This model is a fine-tuned version of
distilbert-base-uncased on the None dataset.
It achieves the following results on the evaluation set:
|
|:-------------:|:-----:|:----:|:---------------:|:------------:|:------:|
| No log | 1.0 | 414 | 6.0798 | 0.2791 | 0.1218 |
| 6.3137 | 2.0 | 828 | 5.3441 | 0.2802 | 0.1242 |
| 5.0492 | 3.0 | 1242 | 4.6148 | 0.2802 | 0.1302 |
| 4.1814 | 4.0 | 1656 | 3.8962 | 0.2895 | 0.1529 |
| 3.3908 | 5.0 | 2070 | 3.2107 | 0.3337 | 0.2453 |
| 3.3908 | 6.0 | 2484 | 2.5814 | 0.3779 | 0.3365 |
| 2.6526 | 7.0 | 2898 | 2.0280 | 0.4256 | 0.4167 |
| 1.9936 | 8.0 | 3312 | 1.5661 | 0.4698 | 0.4800 |
| 1.472 | 9.0 | 3726 | 1.2144 | 0.5209 | 0.5411 |
| 1.1022 | 10.0 | 4140 | 0.9950 | 0.5686 | 0.5908 |
| 0.9323 | 11.0 | 4554 | 0.9181 | 0.5977 | 0.6171 |
| 0.9323 | 12.0 | 4968 | 0.8111 | 0.6221 | 0.6397 |
| 0.8742 | 13.0 | 5382 | 0.7092 | 0.6640 | 0.6779 |
| 0.7487 | 14.0 | 5796 | 0.6502 | 0.6895 | 0.7009 |
| 0.6569 | 15.0 | 6210 | 0.5957 | 0.7581 | 0.7578 |
| 0.5991 | 16.0 | 6624 | 0.5891 | 0.7802 | 0.7822 |
| 0.5488 | 17.0 | 7038 | 0.5739 | 0.7977 | 0.7946 |
| 0.5488 | 18.0 | 7452 | 0.5687 | 0.8012 | 0.7970 |
| 0.5091 | 19.0 | 7866 | 0.5586 | 0.8023 | 0.8016 |
| 0.4818 | 20.0 | 8280 | 0.5570 | 0.8140 | 0.8087 |
| 0.4366 | 21.0 | 8694 | 0.5631 | 0.8116 | 0.8093 |
| 0.4087 | 22.0 | 9108 | 0.5796 | 0.8058 | 0.7981 |
