<!-- 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. -->
bert-base-uncased-finetuned-cola
This model is a fine-tuned version of bert-base-uncased on the glue dataset. It achieves the following results on the evaluation set:
- Loss: 1.5922
- Matthews Correlation: 0.5678
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: 9.207256119784435e-06
- train_batch_size: 4
- eval_batch_size: 16
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 10
Training results
| Training Loss | Epoch | Step | Validation Loss | Matthews Correlation |
|---|---|---|---|---|
| 0.5811 | 1.0 | 2138 | 0.6179 | 0.4846 |
| 0.4698 | 2.0 | 4276 | 0.8083 | 0.5495 |
| 0.3161 | 3.0 | 6414 | 1.1152 | 0.5389 |
| 0.2499 | 4.0 | 8552 | 1.0719 | 0.5624 |
| 0.1755 | 5.0 | 10690 | 1.1734 | 0.5709 |
| 0.1511 | 6.0 | 12828 | 1.2383 | 0.5699 |
| 0.0738 | 7.0 | 14966 | 1.3802 | 0.5598 |
| 0.0677 | 8.0 | 17104 | 1.4711 | 0.5599 |
| 0.0509 | 9.0 | 19242 | 1.5751 | 0.5678 |
| 0.0397 | 10.0 | 21380 | 1.5922 | 0.5678 |
Framework versions
- Transformers 4.28.1
- Pytorch 2.0.0
- Datasets 2.12.0
- Tokenizers 0.13.3
