255yossya/matsuollm2025_main_lora_repo_v8

qwen3-4b-json-yaml-xml-output-lora

This repository provides a LoRA adapter fine-tuned from Qwen/Qwen3-4B-Instruct-2507 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: Qwen/Qwen3-4B-Instruct-2507
• —Method: QLoRA (4-bit)
• —Max sequence length: 1024
• —Epochs: 1
• —Learning rate: 5e-05
• —LoRA: r=64, alpha=128

Usage

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

base = "Qwen/Qwen3-4B-Instruct-2507"
adapter = "255yossya/matsuollm2025_main_lora_repo_v8"

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:

• —u-10bei/structureddatawithcotdataset512v2
• —u-10bei/structureddatawithcotdataset512v4
• —u-10bei/structureddatawithcotdataset512v5
• —u-10bei/structureddatawithcotdataset_512
• —u-10bei/structureddatawithcotdataset_v2
• —u-10bei/structureddatawithcotdataset

License notes:

• —u-10bei/* datasets are MIT-licensed.
• —daichira/* datasets are CC-BY-4.0 licensed (attribution required).
• —Please comply with all applicable dataset licenses and the base model's terms of use.