distilbert_add_GLUE_Experiment_logit_kd_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: 1.1463
- Pearson: 0.0794
- Spearmanr: 0.0664
- Combined Score: 0.0729
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 |
|---|---|---|---|---|---|---|
| 1.9113 | 1.0 | 23 | 1.2309 | -0.0032 | -0.0024 | -0.0028 |
| 1.0858 | 2.0 | 46 | 1.1748 | nan | nan | nan |
| 1.0898 | 3.0 | 69 | 1.2204 | 0.0435 | 0.0457 | 0.0446 |
| 1.0869 | 4.0 | 92 | 1.1463 | 0.0794 | 0.0664 | 0.0729 |
| 1.0768 | 5.0 | 115 | 1.1601 | 0.0815 | 0.0748 | 0.0782 |
| 1.0097 | 6.0 | 138 | 1.2949 | 0.0709 | 0.0812 | 0.0760 |
| 0.8758 | 7.0 | 161 | 1.4262 | 0.0862 | 0.0948 | 0.0905 |
| 0.8349 | 8.0 | 184 | 1.3240 | 0.1054 | 0.1087 | 0.1070 |
| 0.7651 | 9.0 | 207 | 1.4391 | 0.1215 | 0.1282 | 0.1248 |
Framework versions
- Transformers 4.26.0
- Pytorch 1.14.0a0+410ce96
- Datasets 2.9.0
- Tokenizers 0.13.2