small-mlm-glue-sst2-target-glue-mrpc
This model is a fine-tuned version of muhtasham/small-mlm-glue-sst2 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.7182
- Accuracy: 0.7917
- F1: 0.8571
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 | F1 |
|---|---|---|---|---|---|
| 0.3886 | 4.35 | 500 | 0.6884 | 0.7892 | 0.8617 |
| 0.0692 | 8.7 | 1000 | 1.3709 | 0.7917 | 0.8627 |
| 0.0318 | 13.04 | 1500 | 1.4689 | 0.7892 | 0.8562 |
| 0.0266 | 17.39 | 2000 | 1.8846 | 0.7745 | 0.8544 |
| 0.0102 | 21.74 | 2500 | 1.7656 | 0.7941 | 0.8571 |
| 0.0139 | 26.09 | 3000 | 1.7271 | 0.7892 | 0.8552 |
| 0.0168 | 30.43 | 3500 | 1.7505 | 0.7966 | 0.8600 |
| 0.0152 | 34.78 | 4000 | 1.6538 | 0.7843 | 0.8483 |
| 0.0135 | 39.13 | 4500 | 1.7268 | 0.7941 | 0.8618 |
| 0.0148 | 43.48 | 5000 | 1.7182 | 0.7917 | 0.8571 |
Framework versions
- Transformers 4.26.0.dev0
- Pytorch 1.13.0+cu116
- Datasets 2.8.1.dev0
- Tokenizers 0.13.2