tiny-mlm-snli-target-glue-cola
This model is a fine-tuned version of muhtasham/tiny-mlm-snli on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.7851
- Matthews Correlation: 0.1125
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 | Matthews Correlation |
|---|---|---|---|---|
| 0.61 | 1.87 | 500 | 0.6206 | 0.0 |
| 0.6008 | 3.73 | 1000 | 0.6153 | 0.0257 |
| 0.5837 | 5.6 | 1500 | 0.6237 | 0.0218 |
| 0.5551 | 7.46 | 2000 | 0.6447 | 0.0688 |
| 0.5306 | 9.33 | 2500 | 0.6594 | 0.0973 |
| 0.5103 | 11.19 | 3000 | 0.6779 | 0.0957 |
| 0.4842 | 13.06 | 3500 | 0.6971 | 0.1010 |
| 0.4648 | 14.93 | 4000 | 0.7289 | 0.1170 |
| 0.4467 | 16.79 | 4500 | 0.7530 | 0.0991 |
| 0.4266 | 18.66 | 5000 | 0.7851 | 0.1125 |
Framework versions
- Transformers 4.26.0.dev0
- Pytorch 1.13.0+cu116
- Datasets 2.8.1.dev0
- Tokenizers 0.13.2