lavy-Instruct_finding
This model was trained from scratch on an unknown dataset.
It achieves the following results on the evaluation set:
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: 1
- eval_batch_size: 1
- seed: 42
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine
- lr_scheduler_warmup_ratio: 0.03
- num_epochs: 20
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|
| 0.395 | 1.25 | 100 | 0.3798 |
| 0.2238 | 2.5 | 200 | 0.3372 |
| 0.2203 | 3.75 | 300 | 0.3451 |
| 0.1723 | 5.0 | 400 | 0.3716 |
| 0.1269 | 6.25 | 500 | 0.3992 |
| 0.1271 | 7.5 | 600 | 0.3683 |
| 0.121 | 8.75 | 700 | 0.3868 |
| 0.0983 | 10.0 | 800 | 0.4130 |
| 0.0892 | 11.25 | 900 | 0.4376 |
| 0.0764 | 12.5 | 1000 | 0.4336 |
| 0.0703 | 13.75 | 1100 | 0.4661 |
| 0.0733 | 15.0 | 1200 | 0.4852 |
| 0.0513 | 16.25 | 1300 | 0.5114 |
| 0.0518 | 17.5 | 1400 | 0.5383 |
| 0.0443 | 18.75 | 1500 | 0.5647 |
| 0.0407 | 20.0 | 1600 | 0.5674 |
Framework versions
- PEFT 0.5.0
- Transformers 4.37.2
- Pytorch 2.5.1+cu121
- Datasets 3.3.1
- Tokenizers 0.15.2
