small-mlm-glue-wnli-custom-tokenizer-expand-vocab
This model is a fine-tuned version of google/bert_uncased_L-4_H-512_A-8 on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.1184
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 |
|---|---|---|---|
| 3.3425 | 6.25 | 500 | 2.3392 |
| 1.9269 | 12.5 | 1000 | 1.5744 |
| 1.2226 | 18.75 | 1500 | 1.1470 |
| 0.8299 | 25.0 | 2000 | 0.9460 |
| 0.5836 | 31.25 | 2500 | 0.7930 |
| 0.469 | 37.5 | 3000 | 0.3551 |
| 0.3617 | 43.75 | 3500 | 0.4110 |
| 0.2696 | 50.0 | 4000 | 0.2649 |
| 0.2099 | 56.25 | 4500 | 0.3647 |
| 0.2048 | 62.5 | 5000 | 0.1184 |
Framework versions
- Transformers 4.27.0.dev0
- Pytorch 1.13.1+cu116
- Datasets 2.9.1.dev0
- Tokenizers 0.13.2