---
base_model: HamadaMayu/qwen3-4b-structured-output-lora-v4-CoT
datasets:
- daichira/structured-5k-mix-sft
language:
- en
license: apache-2.0
library_name: peft
pipeline_tag: text-generation
tags:
- qlora
- lora
- structured-output
---

qwen3-4b-structured-output-lora-v4-CoT-structured-5k-mix-sft

This repository provides a **LoRA adapter** fine-tuned from
**HamadaMayu/qwen3-4b-structured-output-lora-v4-CoT** using **QLoRA (4-bit, Unsloth)**.

This repository contains **LoRA adapter weights only**.
The base model must be loaded separately.

## Training Objective

This adapter is trained to improve **structured output accuracy**
(JSON / YAML / XML / TOML / CSV).

Loss is applied only to the final assistant output,
while intermediate reasoning (Chain-of-Thought) is masked.

## Training Configuration

- Base model: HamadaMayu/qwen3-4b-structured-output-lora-v4-CoT
- Method: QLoRA (4-bit)
- Max sequence length: 512
- Epochs: 1
- Learning rate: 5e-07
- LoRA: r=64, alpha=128

## Usage

```python
from transformers import AutoModelForCausalLM, AutoTokenizer
from peft import PeftModel
import torch

base = "HamadaMayu/qwen3-4b-structured-output-lora-v4-CoT"
adapter = "your_id/your-repo"

tokenizer = AutoTokenizer.from_pretrained(base)
model = AutoModelForCausalLM.from_pretrained(
    base,
    torch_dtype=torch.float16,
    device_map="auto",
)
model = PeftModel.from_pretrained(model, adapter)
```

## Sources & Terms (IMPORTANT)

Training data: daichira/structured-5k-mix-sft

Dataset License: MIT License. This dataset is used and distributed under the terms of the MIT License.
Compliance: Users must comply with the MIT license (including copyright notice) and the base model's original terms of use.