tiny-mlm-glue-wnli-target-glue-qnli
This model is a fine-tuned version of muhtasham/tiny-mlm-glue-wnli on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.4737
- Accuracy: 0.7794
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: 3e-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: constant
- training_steps: 5000
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 0.6298 | 0.15 | 500 | 0.5598 | 0.7249 |
| 0.563 | 0.31 | 1000 | 0.5282 | 0.7435 |
| 0.5386 | 0.46 | 1500 | 0.5010 | 0.7571 |
| 0.527 | 0.61 | 2000 | 0.5312 | 0.7426 |
| 0.5221 | 0.76 | 2500 | 0.4837 | 0.7743 |
| 0.5131 | 0.92 | 3000 | 0.4730 | 0.7785 |
| 0.4991 | 1.07 | 3500 | 0.4643 | 0.7860 |
| 0.4896 | 1.22 | 4000 | 0.4685 | 0.7809 |
| 0.4755 | 1.37 | 4500 | 0.4734 | 0.7783 |
| 0.4829 | 1.53 | 5000 | 0.4737 | 0.7794 |
Framework versions
- Transformers 4.26.0.dev0
- Pytorch 1.13.0+cu116
- Datasets 2.8.1.dev0
- Tokenizers 0.13.2