qwen3-4b-structured-output-lora-v4

This repository provides a LoRA adapter (v4) 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.

Version: v4 — Data Curation

This is v4 of the SFT training, focusing on data curation via vector search. Based on v3's lesson (score dropped from 0.751 to 0.726 with more data), we curated the training data.

Changes from v3

Parameter	v3	v4	Rationale
Dataset	Merged (8,541)	Curated (2,801)	Quality over quantity
MAX_SEQ_LEN	1024	1024	Same as v2/v3
Epochs	1	1	Same as v2/v3
Learning Rate	5e-6	5e-06	Same as v2/v3

Curated Dataset Details

• Method: sentence-transformers + k-NN search for test-similar data
• k: 100 neighbors per test sample
• Total: 2,801 samples (71% of v2)

Source distribution: 2-2_5k_mix(46%), 2-1_3k_mix(35%), 2-3_hard_4k(19%)

Training Objective

This adapter is trained to improve structured output accuracy (JSON / YAML / XML / TOML / CSV) for the StructEval-T benchmark.

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, Unsloth)
• Max sequence length: 1024
• Epochs: 1
• Learning rate: 5e-06
• Batch size: 2 (effective: 16)
• Gradient accumulation: 8
• LoRA: r=64, alpha=128
• CoT masking: enabled (loss on final output only)

Usage

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

base = "Qwen/Qwen3-4B-Instruct-2507"
adapter = "your_id/qwen3-4b-structured-output-lora-v4"

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: merged dataset (1-1_512_v2 + 1-2_512_v4)

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.