mobilebert_add_GLUE_Experiment_logit_kd_sst2_256
This model is a fine-tuned version of google/mobilebert-uncased on the GLUE SST2 dataset. It achieves the following results on the evaluation set:
- Loss: 1.2641
- Accuracy: 0.7076
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: 128
- eval_batch_size: 128
- 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
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 1.5438 | 1.0 | 527 | 1.4012 | 0.5814 |
| 1.364 | 2.0 | 1054 | 1.5474 | 0.5413 |
| 1.2907 | 3.0 | 1581 | 1.5138 | 0.5642 |
| 1.257 | 4.0 | 2108 | 1.4409 | 0.5665 |
| 1.2417 | 5.0 | 2635 | 1.4473 | 0.5929 |
| 1.2056 | 6.0 | 3162 | 1.2641 | 0.7076 |
| 0.6274 | 7.0 | 3689 | nan | 0.4908 |
| 0.0 | 8.0 | 4216 | nan | 0.4908 |
| 0.0 | 9.0 | 4743 | nan | 0.4908 |
| 0.0 | 10.0 | 5270 | nan | 0.4908 |
| 0.0 | 11.0 | 5797 | nan | 0.4908 |
Framework versions
- Transformers 4.26.0
- Pytorch 1.14.0a0+410ce96
- Datasets 2.9.0
- Tokenizers 0.13.2