tiny-mlm-glue-mnli-target-glue-qnli
This model is a fine-tuned version of muhtasham/tiny-mlm-glue-mnli on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.4695
- Accuracy: 0.7814
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
- num_epochs: 200
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 0.6034 | 0.15 | 500 | 0.5431 | 0.7335 |
| 0.5403 | 0.31 | 1000 | 0.5253 | 0.7459 |
| 0.5174 | 0.46 | 1500 | 0.4953 | 0.7659 |
| 0.5137 | 0.61 | 2000 | 0.5259 | 0.7483 |
| 0.511 | 0.76 | 2500 | 0.4814 | 0.7750 |
| 0.5032 | 0.92 | 3000 | 0.4670 | 0.7847 |
| 0.4901 | 1.07 | 3500 | 0.4525 | 0.7904 |
| 0.4798 | 1.22 | 4000 | 0.4679 | 0.7836 |
| 0.4667 | 1.37 | 4500 | 0.4752 | 0.7798 |
| 0.4736 | 1.53 | 5000 | 0.4695 | 0.7814 |
Framework versions
- Transformers 4.26.0.dev0
- Pytorch 1.13.0+cu116
- Datasets 2.8.1.dev0
- Tokenizers 0.13.2