This model is a fine-tuned version of bert-base-uncased on the squad dataset. It achieves the following results on the evaluation set:
Base model weights were frozen leaving only to finetune the last layer (qa outputs).
Achieved EM: 8.013245033112582, F1: 15.9706088498649
The following hyperparameters were used during training:
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 4.3602 | 1.0 | 5533 | 4.3460 |
| 4.0995 | 2.0 | 11066 | 4.0787 |
| 4.0302 | 3.0 | 16599 | 4.0178 |
pytorch_model.bin 436 MBtraining_args.bin 3 kB