llmhw01
This model is a fine-tuned version of distilbert-base-uncased on an unknown dataset.
It achieves the following results on the evaluation set:
- Loss: 0.8872
- Matthews Correlation: 0.5406
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: 2.1796366105472604e-05
- train_batch_size: 16
- eval_batch_size: 16
- seed: 3
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 5
Training results
| Training Loss |
Epoch |
Step |
Validation Loss |
Matthews Correlation |
| 0.518 |
1.0 |
535 |
0.4573 |
0.4819 |
| 0.3464 |
2.0 |
1070 |
0.5350 |
0.4747 |
| 0.2281 |
3.0 |
1605 |
0.6534 |
0.5211 |
| 0.1642 |
4.0 |
2140 |
0.7936 |
0.5472 |
| 0.1181 |
5.0 |
2675 |
0.8872 |
0.5406 |
Framework versions
- Transformers 4.36.2
- Pytorch 2.1.0+cu121
- Datasets 2.16.0
- Tokenizers 0.15.0
