v1_phi4_mini_bs1x2x2_lr1e4_dora3264do70_5ep_final
This model is a fine-tuned version of microsoft/Phi-4-mini-instruct on the flock-task17-train and the flock-task17-valid datasets. It achieves the following results on the evaluation set:
- Loss: 1.2035
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: 0.0001
- train_batch_size: 1
- eval_batch_size: 1
- seed: 42
- distributed_type: multi-GPU
- num_devices: 2
- gradient_accumulation_steps: 2
- total_train_batch_size: 4
- total_eval_batch_size: 2
- optimizer: Use paged_adamw_32bit with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: cosine
- lr_scheduler_warmup_steps: 20
- num_epochs: 5
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 1.7819 | 0.3846 | 10 | 1.8132 |
| 1.6989 | 0.7692 | 20 | 1.6344 |
| 1.5387 | 1.1538 | 30 | 1.4881 |
| 1.4922 | 1.5385 | 40 | 1.4193 |
| 1.495 | 1.9231 | 50 | 1.3584 |
| 1.3572 | 2.3077 | 60 | 1.3331 |
| 1.2998 | 2.6923 | 70 | 1.2973 |
| 1.3395 | 3.0769 | 80 | 1.2576 |
| 1.298 | 3.4615 | 90 | 1.2360 |
| 1.2243 | 3.8462 | 100 | 1.2174 |
| 1.2153 | 4.2308 | 110 | 1.2064 |
| 1.245 | 4.6154 | 120 | 1.2034 |
| 1.1783 | 5.0 | 130 | 1.2035 |
Framework versions
- PEFT 0.15.1
- Transformers 4.52.4
- Pytorch 2.9.1+cu128
- Datasets 3.5.0
- Tokenizers 0.21.4
- Downloads last month
- -
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support
Model tree for haqishen/v1_phi4_mini_bs1x2x2_lr1e4_dora3264do70_5ep_final_lora
Base model
microsoft/Phi-4-mini-instruct