metadata
license: mit
base_model: microsoft/phi-2
tags:
- generated_from_trainer
model-index:
- name: V0402MP5
results: []
V0402MP5
This model is a fine-tuned version of microsoft/phi-2 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.1581
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: 0.0003
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 16
- total_train_batch_size: 128
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine_with_restarts
- lr_scheduler_warmup_steps: 20
- num_epochs: 1
- mixed_precision_training: Native AMP
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 2.5643 | 0.09 | 10 | 2.3573 |
| 2.1608 | 0.18 | 20 | 1.8527 |
| 1.6271 | 0.27 | 30 | 1.2732 |
| 1.1558 | 0.36 | 40 | 0.8683 |
| 0.7761 | 0.45 | 50 | 0.5042 |
| 0.4741 | 0.54 | 60 | 0.2558 |
| 0.2937 | 0.63 | 70 | 0.1866 |
| 0.2332 | 0.73 | 80 | 0.1681 |
| 0.2203 | 0.82 | 90 | 0.1613 |
| 0.2151 | 0.91 | 100 | 0.1585 |
| 0.2205 | 1.0 | 110 | 0.1581 |
Framework versions
- Transformers 4.36.0.dev0
- Pytorch 2.1.2+cu121
- Datasets 2.14.6
- Tokenizers 0.14.1