Discussion-Phi-4-text
This model is a fine-tuned version of microsoft/phi-4 on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 0.1265
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: 4e-05
- train_batch_size: 1
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 16
- total_train_batch_size: 16
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.95) and epsilon=1e-07 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: linear
- lr_scheduler_warmup_steps: 50
- num_epochs: 3
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 2.6764 | 0.2235 | 10 | 2.4496 |
| 2.1053 | 0.4469 | 20 | 1.9257 |
| 1.222 | 0.6704 | 30 | 1.0594 |
| 0.1878 | 0.8939 | 40 | 0.1615 |
| 0.1642 | 1.1117 | 50 | 0.1395 |
| 0.1127 | 1.3352 | 60 | 0.1343 |
| 0.1483 | 1.5587 | 70 | 0.1332 |
| 0.1342 | 1.7821 | 80 | 0.1338 |
| 0.1529 | 2.0 | 90 | 0.1323 |
| 0.1327 | 2.2235 | 100 | 0.1289 |
| 0.095 | 2.4469 | 110 | 0.1286 |
| 0.1446 | 2.6704 | 120 | 0.1304 |
| 0.1631 | 2.8939 | 130 | 0.1265 |
Framework versions
- PEFT 0.15.2
- Transformers 4.51.3
- Pytorch 2.4.1+cu124
- Datasets 3.5.1
- Tokenizers 0.21.1