distilbert_add_GLUE_Experiment_logit_kd_sst2_192
This model is a fine-tuned version of distilbert-base-uncased on the GLUE SST2 dataset. It achieves the following results on the evaluation set:
- Loss: 0.9978
- Accuracy: 0.7592
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 |
|---|---|---|---|---|
| 1.6036 | 1.0 | 264 | 1.4704 | 0.5092 |
| 1.5492 | 2.0 | 528 | 1.2486 | 0.6915 |
| 0.8905 | 3.0 | 792 | 0.9978 | 0.7592 |
| 0.6841 | 4.0 | 1056 | 1.0505 | 0.7833 |
| 0.5427 | 5.0 | 1320 | 1.1682 | 0.7661 |
| 0.4648 | 6.0 | 1584 | 1.0544 | 0.7867 |
| 0.4365 | 7.0 | 1848 | 1.1258 | 0.7741 |
| 0.3939 | 8.0 | 2112 | 1.2815 | 0.7580 |
Framework versions
- Transformers 4.26.0
- Pytorch 1.14.0a0+410ce96
- Datasets 2.9.0
- Tokenizers 0.13.2