tiny-mlm-glue-cola-target-glue-sst2
This model is a fine-tuned version of muhtasham/tiny-mlm-glue-cola on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.4627
- Accuracy: 0.8142
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 |
|---|---|---|---|---|
| 0.5912 | 0.24 | 500 | 0.4989 | 0.7718 |
| 0.445 | 0.48 | 1000 | 0.4712 | 0.7844 |
| 0.3945 | 0.71 | 1500 | 0.4460 | 0.7993 |
| 0.3735 | 0.95 | 2000 | 0.4566 | 0.7913 |
| 0.335 | 1.19 | 2500 | 0.4430 | 0.8028 |
| 0.3154 | 1.43 | 3000 | 0.4454 | 0.8028 |
| 0.2996 | 1.66 | 3500 | 0.4438 | 0.8062 |
| 0.2915 | 1.9 | 4000 | 0.4463 | 0.8165 |
| 0.2737 | 2.14 | 4500 | 0.5033 | 0.7947 |
| 0.2528 | 2.38 | 5000 | 0.4501 | 0.8165 |
| 0.246 | 2.61 | 5500 | 0.4627 | 0.8142 |
Framework versions
- Transformers 4.26.0.dev0
- Pytorch 1.13.0+cu116
- Datasets 2.8.1.dev0
- Tokenizers 0.13.2