tiny-mlm-glue-mnli-target-glue-cola
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: 0.7272
- Matthews Correlation: 0.0899
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 | Matthews Correlation |
|---|---|---|---|---|
| 0.6097 | 1.87 | 500 | 0.6214 | 0.0 |
| 0.601 | 3.73 | 1000 | 0.6166 | 0.0 |
| 0.5829 | 5.6 | 1500 | 0.6181 | 0.0630 |
| 0.5537 | 7.46 | 2000 | 0.6384 | 0.0793 |
| 0.5231 | 9.33 | 2500 | 0.6629 | 0.1079 |
| 0.508 | 11.19 | 3000 | 0.6679 | 0.0949 |
| 0.4817 | 13.06 | 3500 | 0.6915 | 0.1062 |
| 0.4661 | 14.93 | 4000 | 0.7272 | 0.0899 |
Framework versions
- Transformers 4.26.0.dev0
- Pytorch 1.13.0+cu116
- Datasets 2.8.1.dev0
- Tokenizers 0.13.2