distilbert_sa_GLUE_Experiment_logit_kd_cola_192
This model is a fine-tuned version of distilbert-base-uncased on the GLUE COLA dataset. It achieves the following results on the evaluation set:
- Loss: 0.6737
- Matthews Correlation: 0.0137
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 | Matthews Correlation |
|---|---|---|---|---|
| 0.8307 | 1.0 | 34 | 0.6854 | 0.0 |
| 0.7984 | 2.0 | 68 | 0.6840 | 0.0 |
| 0.7949 | 3.0 | 102 | 0.6839 | 0.0 |
| 0.7958 | 4.0 | 136 | 0.6839 | 0.0 |
| 0.7978 | 5.0 | 170 | 0.6836 | 0.0 |
| 0.7943 | 6.0 | 204 | 0.6814 | 0.0 |
| 0.7713 | 7.0 | 238 | 0.6737 | 0.0137 |
| 0.717 | 8.0 | 272 | 0.6892 | 0.1227 |
| 0.6569 | 9.0 | 306 | 0.6887 | 0.0754 |
| 0.6257 | 10.0 | 340 | 0.7207 | 0.1426 |
| 0.5916 | 11.0 | 374 | 0.7593 | 0.1155 |
| 0.5709 | 12.0 | 408 | 0.7747 | 0.1218 |
Framework versions
- Transformers 4.26.0
- Pytorch 1.14.0a0+410ce96
- Datasets 2.9.0
- Tokenizers 0.13.2