DAPO-Coding-Qwen2.5-1.5B-Instruct

This model is a fine-tuned version of Qwen2.5-1.5B-Instruct using DAPO (Direct Alignment via Preference Optimization) for coding tasks.

Model Details

• Base Model: Qwen/Qwen2.5-1.5B-Instruct
• Fine-tuning Method: DAPO
• Training Steps: 500
• Model Size: 1.5B parameters

Usage

from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("your-username/DAPO-Coding-Qwen2.5-1.5B-Instruct")
model = AutoModelForCausalLM.from_pretrained("your-username/DAPO-Coding-Qwen2.5-1.5B-Instruct")

# Example usage for code generation
prompt = "Write a Python function to calculate fibonacci numbers:"
inputs = tokenizer(prompt, return_tensors="pt")
outputs = model.generate(**inputs, max_length=200)
response = tokenizer.decode(outputs[0], skip_special_tokens=True)
print(response)

Training Details

This model was trained using the DAPO framework for alignment with coding preferences. Training was completed at global step 500.