small-mlm-glue-sst2-custom-tokenizer
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: 6.5283
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 |
|---|---|---|---|
| 7.2931 | 0.4 | 500 | 6.8821 |
| 6.7316 | 0.8 | 1000 | 6.5537 |
| 6.3512 | 1.2 | 1500 | 6.5318 |
| 6.3054 | 1.6 | 2000 | 6.6485 |
| 6.2502 | 2.0 | 2500 | 6.5860 |
| 6.0353 | 2.4 | 3000 | 6.4283 |
| 5.9789 | 2.8 | 3500 | 6.5207 |
| 5.9634 | 3.2 | 4000 | 6.4901 |
| 5.9423 | 3.6 | 4500 | 6.5283 |
Framework versions
- Transformers 4.26.0.dev0
- Pytorch 1.13.0+cu116
- Datasets 2.8.1.dev0
- Tokenizers 0.13.2