distilbert_sa_GLUE_Experiment_logit_kd_rte_192
This model is a fine-tuned version of distilbert-base-uncased on the GLUE RTE dataset. It achieves the following results on the evaluation set:
- Loss: 0.4228
- Accuracy: 0.4729
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 |
|---|---|---|---|---|
| 0.4351 | 1.0 | 10 | 0.4238 | 0.4729 |
| 0.4173 | 2.0 | 20 | 0.4246 | 0.4729 |
| 0.4173 | 3.0 | 30 | 0.4238 | 0.4729 |
| 0.4169 | 4.0 | 40 | 0.4237 | 0.4729 |
| 0.4185 | 5.0 | 50 | 0.4242 | 0.4729 |
| 0.4171 | 6.0 | 60 | 0.4237 | 0.4729 |
| 0.417 | 7.0 | 70 | 0.4237 | 0.4729 |
| 0.4175 | 8.0 | 80 | 0.4240 | 0.4729 |
| 0.4166 | 9.0 | 90 | 0.4235 | 0.4729 |
| 0.416 | 10.0 | 100 | 0.4236 | 0.4729 |
| 0.4174 | 11.0 | 110 | 0.4244 | 0.4729 |
| 0.4153 | 12.0 | 120 | 0.4234 | 0.4729 |
| 0.416 | 13.0 | 130 | 0.4238 | 0.4729 |
| 0.4134 | 14.0 | 140 | 0.4228 | 0.4729 |
| 0.4042 | 15.0 | 150 | 0.4279 | 0.4729 |
| 0.3771 | 16.0 | 160 | 0.4453 | 0.4621 |
| 0.3491 | 17.0 | 170 | 0.4650 | 0.4513 |
| 0.3258 | 18.0 | 180 | 0.4842 | 0.4729 |
| 0.3064 | 19.0 | 190 | 0.4996 | 0.4621 |
Framework versions
- Transformers 4.26.0
- Pytorch 1.14.0a0+410ce96
- Datasets 2.9.0
- Tokenizers 0.13.2