Discussion-Phi-4-multimodal-instruct-audio-level
This model is a fine-tuned version of microsoft/Phi-4-multimodal-instruct on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 14.8465
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 |
|---|---|---|---|
| 4.4085 | 0.2235 | 10 | 15.3387 |
| 0.2499 | 0.4469 | 20 | 14.8507 |
| 0.2277 | 0.6704 | 30 | 14.7963 |
| 0.2977 | 0.8939 | 40 | 14.8726 |
| 0.2456 | 1.1117 | 50 | 14.8255 |
| 0.1969 | 1.3352 | 60 | 14.9068 |
| 0.2661 | 1.5587 | 70 | 14.7583 |
| 0.1688 | 1.7821 | 80 | 14.9709 |
| 0.0946 | 2.0 | 90 | 14.7753 |
| 0.1948 | 2.2235 | 100 | 14.7443 |
| 0.0879 | 2.4469 | 110 | 14.7647 |
| 0.057 | 2.6704 | 120 | 14.8623 |
| 0.0259 | 2.8939 | 130 | 14.8465 |
Framework versions
- Transformers 4.51.3
- Pytorch 2.4.1+cu124
- Datasets 3.6.0
- Tokenizers 0.21.1