small-mlm-snli-target-glue-stsb
This model is a fine-tuned version of muhtasham/small-mlm-snli on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.5736
- Pearson: 0.8731
- Spearmanr: 0.8693
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 | Pearson | Spearmanr |
|---|---|---|---|---|---|
| 0.851 | 2.78 | 500 | 0.6505 | 0.8616 | 0.8636 |
| 0.3135 | 5.56 | 1000 | 0.5968 | 0.8663 | 0.8646 |
| 0.1838 | 8.33 | 1500 | 0.6331 | 0.8648 | 0.8614 |
| 0.1314 | 11.11 | 2000 | 0.6045 | 0.8680 | 0.8640 |
| 0.1046 | 13.89 | 2500 | 0.5871 | 0.8693 | 0.8660 |
| 0.0859 | 16.67 | 3000 | 0.6002 | 0.8690 | 0.8645 |
| 0.073 | 19.44 | 3500 | 0.5644 | 0.8720 | 0.8676 |
| 0.064 | 22.22 | 4000 | 0.5672 | 0.8734 | 0.8690 |
| 0.0576 | 25.0 | 4500 | 0.5835 | 0.8715 | 0.8680 |
| 0.052 | 27.78 | 5000 | 0.5736 | 0.8731 | 0.8693 |
Framework versions
- Transformers 4.26.0.dev0
- Pytorch 1.13.0+cu116
- Datasets 2.8.1.dev0
- Tokenizers 0.13.2