small-mlm-glue-mnli-target-glue-qnli
This model is a fine-tuned version of muhtasham/small-mlm-glue-mnli on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.3505
- Accuracy: 0.8576
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 |
|---|---|---|---|---|
| 0.487 | 0.15 | 500 | 0.3918 | 0.8303 |
| 0.4456 | 0.31 | 1000 | 0.3872 | 0.8312 |
| 0.423 | 0.46 | 1500 | 0.3588 | 0.8464 |
| 0.4118 | 0.61 | 2000 | 0.3692 | 0.8360 |
| 0.412 | 0.76 | 2500 | 0.3426 | 0.8514 |
| 0.3945 | 0.92 | 3000 | 0.3313 | 0.8530 |
| 0.3544 | 1.07 | 3500 | 0.3423 | 0.8578 |
| 0.3167 | 1.22 | 4000 | 0.3529 | 0.8570 |
| 0.3152 | 1.37 | 4500 | 0.3512 | 0.8569 |
| 0.3228 | 1.53 | 5000 | 0.3505 | 0.8576 |
Framework versions
- Transformers 4.26.0.dev0
- Pytorch 1.13.0+cu116
- Datasets 2.8.1.dev0
- Tokenizers 0.13.2