Downloads · 30 days
8
22% of all-time downloads
Kohsaku/002_qwen3-4b_sft_5kmix-512v4
002_qwen3-4b_sft_5kmix-512v4 is a text generation model from Kohsaku. Use it when you need the model to write or continue text. It is set up for peft. The card lists the license as apache-2.0.
This repository provides a LoRA adapter fine-tuned from Qwen/Qwen3-4B-Instruct-2507 using QLoRA (4-bit, Unsloth).
Downloads · 30 days
8
22% of all-time downloads
All-time downloads
37
Public
Repo size
540 MB
Likes
0
Public
Click a slice to open those files.
.safetensors529 MB · 97%
From the Hugging Face model README
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.
This adapter combines two datasets to leverage their complementary strengths:
daichira/structured-5k-mix-sft (5,000 samples)
u-10bei/structured_data_with_cot_dataset_512_v4 (4,608 samples)
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 for CoT-format data.
from transformers import AutoModelForCausalLM, AutoTokenizer
from peft import PeftModel
import torch
base = "Qwen/Qwen3-4B-Instruct-2507"
adapter = f"Kohsaku/002_qwen3-4b_sft_5kmix-512v4"
tokenizer = AutoTokenizer.from_pretrained(base)
model = AutoModelForCausalLM.from_pretrained(
base,
torch_dtype=torch.float16,
device_map="auto",
)
model = PeftModel.from_pretrained(model, adapter)
Training data:
Dataset License: MIT License. These datasets are 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.