tiny-mlm-glue-stsb
This model is a fine-tuned version of google/bert_uncased_L-2_H-128_A-2 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 3.7830
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 |
|---|---|---|---|
| 4.7548 | 0.7 | 500 | 3.9253 |
| 4.4535 | 1.39 | 1000 | 3.9069 |
| 4.364 | 2.09 | 1500 | 3.8392 |
| 4.1534 | 2.78 | 2000 | 3.7830 |
| 4.2317 | 3.48 | 2500 | 3.7450 |
| 4.1233 | 4.17 | 3000 | 3.7755 |
| 4.0383 | 4.87 | 3500 | 3.7060 |
| 4.0459 | 5.56 | 4000 | 3.8708 |
| 3.9321 | 6.26 | 4500 | 3.8573 |
| 4.0206 | 6.95 | 5000 | 3.7830 |
Framework versions
- Transformers 4.26.0.dev0
- Pytorch 1.13.0+cu116
- Datasets 2.8.1.dev0
- Tokenizers 0.13.2