metadata
base_model: microsoft/Phi-3-mini-4k-instruct
library_name: peft
license: mit
tags:
- trl
- sft
- generated_from_trainer
model-index:
- name: outputdir
results: []
outputdir
This model is a fine-tuned version of microsoft/Phi-3-mini-4k-instruct on an unknown dataset. It achieves the following results on the evaluation set:
- Loss: 1.2794
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: 1e-05
- train_batch_size: 8
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 32
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- lr_scheduler_warmup_ratio: 0.1
- num_epochs: 2
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 3.0506 | 0.0771 | 100 | 2.8512 |
| 2.4484 | 0.1542 | 200 | 1.7779 |
| 1.5032 | 0.2313 | 300 | 1.3608 |
| 1.3472 | 0.3085 | 400 | 1.3325 |
| 1.3304 | 0.3856 | 500 | 1.3195 |
| 1.3156 | 0.4627 | 600 | 1.3103 |
| 1.313 | 0.5398 | 700 | 1.3038 |
| 1.2976 | 0.6169 | 800 | 1.2991 |
| 1.3001 | 0.6940 | 900 | 1.2956 |
| 1.2976 | 0.7712 | 1000 | 1.2927 |
| 1.2902 | 0.8483 | 1100 | 1.2907 |
| 1.2831 | 0.9254 | 1200 | 1.2888 |
| 1.2839 | 1.0025 | 1300 | 1.2874 |
| 1.2792 | 1.0796 | 1400 | 1.2860 |
| 1.295 | 1.1567 | 1500 | 1.2845 |
| 1.287 | 1.2339 | 1600 | 1.2838 |
| 1.2831 | 1.3110 | 1700 | 1.2831 |
| 1.2764 | 1.3881 | 1800 | 1.2821 |
| 1.2836 | 1.4652 | 1900 | 1.2815 |
| 1.2844 | 1.5423 | 2000 | 1.2810 |
| 1.2791 | 1.6194 | 2100 | 1.2804 |
| 1.2869 | 1.6965 | 2200 | 1.2799 |
| 1.2814 | 1.7737 | 2300 | 1.2798 |
| 1.2775 | 1.8508 | 2400 | 1.2796 |
| 1.2837 | 1.9279 | 2500 | 1.2794 |
Framework versions
- PEFT 0.12.0
- Transformers 4.42.4
- Pytorch 2.3.1+cu121
- Datasets 2.21.0
- Tokenizers 0.19.1