<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
mobilebert_sa_GLUE_Experiment_logit_kd_cola_256
This model is a fine-tuned version of google/mobilebert-uncased on the GLUE COLA dataset. It achieves the following results on the evaluation set:
- Loss: 0.6753
- Matthews Correlation: 0.0
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 | Matthews Correlation |
|---|---|---|---|---|
| 0.8155 | 1.0 | 67 | 0.6867 | 0.0 |
| 0.797 | 2.0 | 134 | 0.6862 | 0.0 |
| 0.7961 | 3.0 | 201 | 0.6836 | 0.0 |
| 0.7944 | 4.0 | 268 | 0.6821 | 0.0 |
| 0.7863 | 5.0 | 335 | 0.6753 | 0.0 |
| 0.7138 | 6.0 | 402 | 0.6790 | 0.1085 |
| 0.6262 | 7.0 | 469 | 0.7238 | 0.1231 |
| 0.5782 | 8.0 | 536 | 0.7285 | 0.1281 |
| 0.5482 | 9.0 | 603 | 0.7484 | 0.1281 |
| 0.5318 | 10.0 | 670 | 0.7918 | 0.1182 |
Framework versions
- Transformers 4.26.0
- Pytorch 1.14.0a0+410ce96
- Datasets 2.9.0
- Tokenizers 0.13.2
