small-mlm-glue-sst2-target-glue-qnli
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: 0.3630
- Accuracy: 0.8495
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 |
|---|---|---|---|---|
| 0.4855 | 0.15 | 500 | 0.3953 | 0.8254 |
| 0.4445 | 0.31 | 1000 | 0.3830 | 0.8325 |
| 0.4215 | 0.46 | 1500 | 0.3571 | 0.8422 |
| 0.4085 | 0.61 | 2000 | 0.3721 | 0.8384 |
| 0.4111 | 0.76 | 2500 | 0.3340 | 0.8556 |
| 0.3924 | 0.92 | 3000 | 0.3283 | 0.8556 |
| 0.3534 | 1.07 | 3500 | 0.3431 | 0.8550 |
| 0.3177 | 1.22 | 4000 | 0.3506 | 0.8567 |
| 0.3156 | 1.37 | 4500 | 0.3498 | 0.8547 |
| 0.3208 | 1.53 | 5000 | 0.3630 | 0.8495 |
Framework versions
- Transformers 4.26.0.dev0
- Pytorch 1.13.0+cu116
- Datasets 2.8.1.dev0
- Tokenizers 0.13.2