first-DPO-without-remove-approach-v2

This model is a fine-tuned version of Qwen/Qwen3-4B-Instruct-2507 using Direct Preference Optimization (DPO) via the Unsloth library.

This repository contains the full-merged 16-bit weights. No adapter loading is required.

Training Configuration

• Base model: Qwen/Qwen3-4B-Instruct-2507
• Method: DPO (Direct Preference Optimization)
• Epochs: 2
• Learning rate: 5e-07
• Beta: 0.1
• Max sequence length: 1024
• LoRA Config: r=4, alpha=16, dropout=0 (merged into base)

Usage

from transformers import AutoModelForCausalLM, AutoTokenizer
import torch

model_id = "yokoe/first-DPO-without-remove-approach-v2"
tokenizer = AutoTokenizer.from_pretrained(model_id)
model = AutoModelForCausalLM.from_pretrained(
    model_id,
    torch_dtype=torch.float16,
    device_map="auto"
)

Sources & License

• Training Data: [u-10bei/dpo-dataset-qwen-cot]
• License: MIT License.