small-mlm-glue-stsb
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: 2.7187
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 |
|---|---|---|---|
| 3.2666 | 0.7 | 500 | 2.7896 |
| 3.0117 | 1.39 | 1000 | 2.8245 |
| 2.9461 | 2.09 | 1500 | 2.7108 |
| 2.7341 | 2.78 | 2000 | 2.6721 |
| 2.7235 | 3.48 | 2500 | 2.6946 |
| 2.6687 | 4.17 | 3000 | 2.7103 |
| 2.5373 | 4.87 | 3500 | 2.7187 |
Framework versions
- Transformers 4.26.0.dev0
- Pytorch 1.13.0+cu116
- Datasets 2.8.1.dev0
- Tokenizers 0.13.2