distilbert_add_GLUE_Experiment_logit_kd_mnli_384
This model is a fine-tuned version of distilbert-base-uncased on the GLUE MNLI dataset. It achieves the following results on the evaluation set:
- Loss: 0.5304
- Accuracy: 0.5770
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 | Accuracy |
|---|---|---|---|---|
| 0.6035 | 1.0 | 1534 | 0.5764 | 0.4805 |
| 0.5667 | 2.0 | 3068 | 0.5578 | 0.5171 |
| 0.5542 | 3.0 | 4602 | 0.5520 | 0.5243 |
| 0.5447 | 4.0 | 6136 | 0.5460 | 0.5422 |
| 0.5338 | 5.0 | 7670 | 0.5387 | 0.5671 |
| 0.5172 | 6.0 | 9204 | 0.5304 | 0.5781 |
| 0.4993 | 7.0 | 10738 | 0.5333 | 0.5847 |
| 0.482 | 8.0 | 12272 | 0.5317 | 0.5901 |
| 0.4654 | 9.0 | 13806 | 0.5323 | 0.5949 |
| 0.4504 | 10.0 | 15340 | 0.5368 | 0.5957 |
| 0.4369 | 11.0 | 16874 | 0.5405 | 0.5980 |
Framework versions
- Transformers 4.26.0
- Pytorch 1.14.0a0+410ce96
- Datasets 2.9.0
- Tokenizers 0.13.2