ft-bert-with-swag
This model is a fine-tuned version of bert-base-uncased on the swag dataset. It achieves the following results on the evaluation set:
- Loss: 0.9935
- Accuracy: 0.62
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: 5e-05
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| No log | 1.0 | 13 | 1.1779 | 0.5 |
| No log | 2.0 | 26 | 0.9935 | 0.62 |
| No log | 3.0 | 39 | 1.0270 | 0.58 |
| No log | 4.0 | 52 | 1.0306 | 0.58 |
| No log | 5.0 | 65 | 1.0466 | 0.62 |
Framework versions
- Transformers 4.35.2
- Pytorch 2.0.0
- Datasets 2.15.0
- Tokenizers 0.15.0