distilbert_add_GLUE_Experiment_stsb_384
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: 2.2840
- Pearson: 0.0810
- Spearmanr: 0.0645
- Combined Score: 0.0728
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 |
|---|---|---|---|---|---|---|
| 4.1199 | 1.0 | 23 | 2.4288 | 0.0129 | 0.0100 | 0.0114 |
| 2.1794 | 2.0 | 46 | 2.3389 | 0.0705 | 0.0708 | 0.0707 |
| 2.1777 | 3.0 | 69 | 2.4374 | 0.0110 | 0.0098 | 0.0104 |
| 2.1687 | 4.0 | 92 | 2.2840 | 0.0810 | 0.0645 | 0.0728 |
| 2.1636 | 5.0 | 115 | 2.3433 | 0.0847 | 0.0760 | 0.0803 |
| 2.0351 | 6.0 | 138 | 2.5863 | 0.0699 | 0.0809 | 0.0754 |
| 1.7847 | 7.0 | 161 | 2.7544 | 0.0920 | 0.1023 | 0.0972 |
| 1.6358 | 8.0 | 184 | 2.8389 | 0.1135 | 0.1197 | 0.1166 |
| 1.5322 | 9.0 | 207 | 2.9114 | 0.1346 | 0.1367 | 0.1356 |
Framework versions
- Transformers 4.26.0
- Pytorch 1.14.0a0+410ce96
- Datasets 2.8.0
- Tokenizers 0.13.2