distilbert-base-uncased-finetuned-winogrande
This model is a fine-tuned version of distilbert-base-uncased on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.9100
- Accuracy: 0.5525
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: 1e-05
- train_batch_size: 8
- eval_batch_size: 4
- seed: 11262
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 4000
- num_epochs: 8
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| No log | 1.0 | 1156 | 0.6931 | 0.5036 |
| No log | 2.0 | 2312 | 0.6926 | 0.5067 |
| No log | 3.0 | 3468 | 0.6929 | 0.5075 |
| No log | 4.0 | 4624 | 0.6908 | 0.5272 |
| 0.6934 | 5.0 | 5780 | 0.6982 | 0.5391 |
| 0.6934 | 6.0 | 6936 | 0.7557 | 0.5312 |
| 0.6934 | 7.0 | 8092 | 0.8402 | 0.5478 |
| 0.6934 | 8.0 | 9248 | 0.9100 | 0.5525 |
Framework versions
- Transformers 4.26.1
- Pytorch 1.13.1+cpu
- Datasets 2.10.1
- Tokenizers 0.13.2