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 | 2.5912 | 0.7128 | 0.6101 |
| 3.7111 | 2.0 | 828 | 2.2184 | 0.6663 | 0.5942 |
| 2.7081 | 3.0 | 1242 | 1.9706 | 0.6244 | 0.5875 |
| 2.2729 | 4.0 | 1656 | 1.8491 | 0.5895 | 0.5732 |
| 2.0505 | 5.0 | 2070 | 1.7316 | 0.5919 | 0.5764 |
| 2.0505 | 6.0 | 2484 | 1.5095 | 0.7105 | 0.6440 |
| 1.7673 | 7.0 | 2898 | 1.2680 | 0.7093 | 0.6567 |
| 1.3485 | 8.0 | 3312 | 1.0762 | 0.7047 | 0.6666 |
| 1.0343 | 9.0 | 3726 | 0.9565 | 0.7174 | 0.6912 |
| 0.8004 | 10.0 | 4140 | 0.9240 | 0.7140 | 0.7073 |
| 0.668 | 11.0 | 4554 | 0.9017 | 0.7709 | 0.7563 |
| 0.668 | 12.0 | 4968 | 0.8650 | 0.7907 | 0.7884 |
| 0.5588 | 13.0 | 5382 | 0.9005 | 0.7837 | 0.7876 |
| 0.4851 | 14.0 | 5796 | 0.8801 | 0.8081 | 0.8056 |
