tiny-mlm-glue-sst2-from-scratch-custom-tokenizer-expand-vocab
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: 6.5699
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 |
|---|---|---|---|
| 9.6351 | 0.4 | 500 | 8.5392 |
| 7.9053 | 0.8 | 1000 | 7.2886 |
| 7.0854 | 1.2 | 1500 | 6.8440 |
| 6.8355 | 1.6 | 2000 | 6.6595 |
| 6.7188 | 2.0 | 2500 | 6.5499 |
| 6.698 | 2.4 | 3000 | 6.6385 |
| 6.6435 | 2.8 | 3500 | 6.6154 |
| 6.6402 | 3.2 | 4000 | 6.5699 |
Framework versions
- Transformers 4.27.0.dev0
- Pytorch 1.13.1+cu116
- Datasets 2.9.1.dev0
- Tokenizers 0.13.2