small-mlm-snli-target-glue-mnli
This model is a fine-tuned version of muhtasham/small-mlm-snli on the None dataset. It achieves the following results on the evaluation set:
- Loss: 0.6523
- Accuracy: 0.7211
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: 3e-05
- train_batch_size: 32
- eval_batch_size: 32
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: constant
- training_steps: 5000
Training results
| Training Loss | Epoch | Step | Validation Loss | Accuracy |
|---|---|---|---|---|
| 0.9112 | 0.04 | 500 | 0.8191 | 0.6414 |
| 0.8114 | 0.08 | 1000 | 0.7787 | 0.6652 |
| 0.7747 | 0.12 | 1500 | 0.7499 | 0.6732 |
| 0.7494 | 0.16 | 2000 | 0.7257 | 0.6909 |
| 0.7451 | 0.2 | 2500 | 0.7153 | 0.6921 |
| 0.7307 | 0.24 | 3000 | 0.7069 | 0.6995 |
| 0.7125 | 0.29 | 3500 | 0.6837 | 0.7104 |
| 0.7031 | 0.33 | 4000 | 0.6690 | 0.7202 |
| 0.6821 | 0.37 | 4500 | 0.6802 | 0.7201 |
| 0.6824 | 0.41 | 5000 | 0.6523 | 0.7211 |
Framework versions
- Transformers 4.26.0.dev0
- Pytorch 1.13.0+cu116
- Datasets 2.8.1.dev0
- Tokenizers 0.13.2