metadata
base_model: microsoft/Phi-3.5-mini-instruct
library_name: peft
license: mit
tags:
- trl
- sft
- generated_from_trainer
model-index:
- name: outputs
results: []
outputs
This model is a fine-tuned version of microsoft/Phi-3.5-mini-instruct on the None dataset. It achieves the following results on the evaluation set:
- Loss: 1.0926
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: 1
- eval_batch_size: 8
- seed: 42
- gradient_accumulation_steps: 4
- total_train_batch_size: 4
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: cosine_with_restarts
- lr_scheduler_warmup_steps: 20
- training_steps: 160
Training results
| Training Loss | Epoch | Step | Validation Loss |
|---|---|---|---|
| 2.2453 | 1.1429 | 10 | 2.1914 |
| 1.8068 | 2.2857 | 20 | 1.8497 |
| 1.5702 | 3.4286 | 30 | 1.5758 |
| 1.5012 | 4.5714 | 40 | 1.2568 |
| 1.2486 | 5.7143 | 50 | 1.1309 |
| 0.948 | 6.8571 | 60 | 1.0965 |
| 1.0246 | 8.0 | 70 | 1.0826 |
| 0.7834 | 9.1429 | 80 | 1.0786 |
| 0.8802 | 10.2857 | 90 | 1.0755 |
| 0.7285 | 11.4286 | 100 | 1.0781 |
| 0.8049 | 12.5714 | 110 | 1.0855 |
| 0.831 | 13.7143 | 120 | 1.0920 |
| 0.6412 | 14.8571 | 130 | 1.0900 |
| 0.7723 | 16.0 | 140 | 1.0898 |
| 0.8869 | 17.1429 | 150 | 1.0908 |
| 0.6173 | 18.2857 | 160 | 1.0926 |
Framework versions
- PEFT 0.12.0
- Transformers 4.44.2
- Pytorch 2.4.0+cu121
- Datasets 2.21.0
- Tokenizers 0.19.1