small-vanilla-target-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: 0.5625
- Pearson: 0.8713
- Spearmanr: 0.8677
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 | Pearson | Spearmanr |
|---|---|---|---|---|---|
| 0.823 | 2.78 | 500 | 0.5972 | 0.8689 | 0.8689 |
| 0.2951 | 5.56 | 1000 | 0.5683 | 0.8725 | 0.8710 |
| 0.181 | 8.33 | 1500 | 0.5985 | 0.8695 | 0.8657 |
| 0.1349 | 11.11 | 2000 | 0.5915 | 0.8708 | 0.8679 |
| 0.1067 | 13.89 | 2500 | 0.5625 | 0.8713 | 0.8677 |
Framework versions
- Transformers 4.26.0.dev0
- Pytorch 1.13.0+cu116
- Datasets 2.8.1.dev0
- Tokenizers 0.13.2