mobilebert_add_GLUE_Experiment_logit_kd_mnli_256
This model is a fine-tuned version of google/mobilebert-uncased on the GLUE MNLI dataset. It achieves the following results on the evaluation set:
- Loss: 1.7834
- Accuracy: 0.3295
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 |
|---|---|---|---|---|
| 1.8865 | 1.0 | 3068 | 1.7940 | 0.3274 |
| 1.8864 | 2.0 | 6136 | 1.7940 | 0.3274 |
| 1.8864 | 3.0 | 9204 | 1.7944 | 0.3274 |
| 1.8864 | 4.0 | 12272 | 1.7940 | 0.3274 |
| 1.8864 | 5.0 | 15340 | 1.7938 | 0.3274 |
| 1.8864 | 6.0 | 18408 | 1.7940 | 0.3274 |
| 1.8864 | 7.0 | 21476 | 1.7944 | 0.3274 |
| 1.8864 | 8.0 | 24544 | 1.7939 | 0.3274 |
| 1.8864 | 9.0 | 27612 | 1.7939 | 0.3274 |
| 1.8863 | 10.0 | 30680 | 1.7940 | 0.3274 |
Framework versions
- Transformers 4.26.0
- Pytorch 1.14.0a0+410ce96
- Datasets 2.9.0
- Tokenizers 0.13.2