distilbert_sa_GLUE_Experiment_stsb_256
This model is a fine-tuned version of distilbert-base-uncased on the GLUE STSB dataset. It achieves the following results on the evaluation set:
- Loss: 2.2582
- Pearson: -0.0216
- Spearmanr: -0.0200
- Combined Score: -0.0208
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: 5e-05
- train_batch_size: 256
- eval_batch_size: 256
- seed: 10
- distributed_type: multi-GPU
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 50
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss | Pearson | Spearmanr | Combined Score |
|---|---|---|---|---|---|---|
| 6.555 | 1.0 | 23 | 3.5514 | -0.0389 | -0.0280 | -0.0335 |
| 3.774 | 2.0 | 46 | 2.6498 | -0.0559 | -0.0576 | -0.0567 |
| 2.7302 | 3.0 | 69 | 2.2582 | -0.0216 | -0.0200 | -0.0208 |
| 2.2286 | 4.0 | 92 | 2.3753 | 0.0290 | 0.0275 | 0.0283 |
| 2.1694 | 5.0 | 115 | 2.3590 | 0.0512 | 0.0607 | 0.0559 |
| 2.059 | 6.0 | 138 | 2.4605 | 0.0601 | 0.0797 | 0.0699 |
| 1.8739 | 7.0 | 161 | 2.6062 | 0.1242 | 0.1311 | 0.1276 |
| 1.6112 | 8.0 | 184 | 2.4597 | 0.2007 | 0.2205 | 0.2106 |
Framework versions
- Transformers 4.26.0
- Pytorch 1.14.0a0+410ce96
- Datasets 2.8.0
- Tokenizers 0.13.2