---
language:
- en
license: apache-2.0
library_name: transformers
tags:
- orpo
- Phi 3
base_model:
- microsoft/Phi-3-mini-128k-instruct
datasets:
- mlabonne/orpo-dpo-mix-40k
---

# Orpo-Phi3-3B-128K

![image/jpeg](https://cdn-uploads.huggingface.co/production/uploads/64fc6d81d75293f417fee1d1/LOJemGwVIPOK4xTczt2MZ.jpeg)

This is an ORPO fine-tune of [microsoft/Phi-3-mini-128k-instruct](https://huggingface.co/microsoft/Phi-3-mini-128k-instruct) on 10k samples of [mlabonne/orpo-dpo-mix-40k](https://huggingface.co/datasets/mlabonne/orpo-dpo-mix-40k).

## 💻 Usage

```python
!pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "Muhammad2003/Orpo-Phi3-3B-128K"
messages = [{"role": "user", "content": "What is a large language model?"}]

tokenizer = AutoTokenizer.from_pretrained(model)
prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
pipeline = transformers.pipeline(
    "text-generation",
    model=model,
    torch_dtype=torch.float16,
    device_map="auto",
)

outputs = pipeline(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
print(outputs[0]["generated_text"])
```

## 📈 Training curves

Wandb Report

![image/png](https://cdn-uploads.huggingface.co/production/uploads/64fc6d81d75293f417fee1d1/uOFRuGlp6z6WLeRDL3sLA.png)

## 🏆 Evaluation
Coming Soon!