small-mlm-glue-qnli-target-glue-qnli
This model is a fine-tuned version of muhtasham/small-mlm-glue-qnli on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.3483
- Accuracy: 0.8598
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 |
|---|---|---|---|---|
| 0.4857 | 0.15 | 500 | 0.3911 | 0.8305 |
| 0.444 | 0.31 | 1000 | 0.3833 | 0.8318 |
| 0.421 | 0.46 | 1500 | 0.3587 | 0.8433 |
| 0.4119 | 0.61 | 2000 | 0.3669 | 0.8400 |
| 0.4098 | 0.76 | 2500 | 0.3444 | 0.8451 |
| 0.3909 | 0.92 | 3000 | 0.3305 | 0.8558 |
| 0.3535 | 1.07 | 3500 | 0.3413 | 0.8591 |
| 0.3168 | 1.22 | 4000 | 0.3438 | 0.8622 |
| 0.312 | 1.37 | 4500 | 0.3583 | 0.8539 |
| 0.3183 | 1.53 | 5000 | 0.3483 | 0.8598 |
Framework versions
- Transformers 4.26.0.dev0
- Pytorch 1.13.0+cu116
- Datasets 2.8.1.dev0
- Tokenizers 0.13.2