<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
tiny-mlm-glue-stsb-target-glue-mnli
This model is a fine-tuned version of muhtasham/tiny-mlm-glue-stsb on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.8112
- Accuracy: 0.6365
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 |
|---|---|---|---|---|
| 1.0767 | 0.04 | 500 | 1.0354 | 0.4644 |
| 1.0091 | 0.08 | 1000 | 0.9646 | 0.5496 |
| 0.9629 | 0.12 | 1500 | 0.9236 | 0.5798 |
| 0.9384 | 0.16 | 2000 | 0.9054 | 0.5916 |
| 0.9254 | 0.2 | 2500 | 0.8894 | 0.5995 |
| 0.9167 | 0.24 | 3000 | 0.8788 | 0.6028 |
| 0.9013 | 0.29 | 3500 | 0.8707 | 0.6104 |
| 0.8962 | 0.33 | 4000 | 0.8603 | 0.6132 |
| 0.8802 | 0.37 | 4500 | 0.8561 | 0.6185 |
| 0.8834 | 0.41 | 5000 | 0.8490 | 0.6220 |
| 0.8734 | 0.45 | 5500 | 0.8427 | 0.6227 |
| 0.8721 | 0.49 | 6000 | 0.8399 | 0.6278 |
| 0.8739 | 0.53 | 6500 | 0.8336 | 0.6331 |
| 0.8654 | 0.57 | 7000 | 0.8345 | 0.6294 |
| 0.8579 | 0.61 | 7500 | 0.8192 | 0.6375 |
| 0.8567 | 0.65 | 8000 | 0.8191 | 0.6348 |
| 0.8517 | 0.69 | 8500 | 0.8275 | 0.6315 |
| 0.8528 | 0.73 | 9000 | 0.8060 | 0.6433 |
| 0.8448 | 0.77 | 9500 | 0.8152 | 0.6355 |
| 0.8361 | 0.81 | 10000 | 0.8026 | 0.6415 |
| 0.8398 | 0.86 | 10500 | 0.8112 | 0.6365 |
Framework versions
- Transformers 4.26.0.dev0
- Pytorch 1.13.0+cu116
- Datasets 2.8.1.dev0
- Tokenizers 0.13.2
