Used dataset: u-10bei/structured_data_with_cot_dataset_512_v2
Treatment of reasoning CoT: Reasoning CoT text was removed entirely, and no <think> tags were added to the training data.
The training data was upsampled using format-specific sampling weight multipliers: TOML ×1.5, XML ×1.3, YAML ×1.2.

Usage

python

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

base = "Qwen/Qwen3-4B-Instruct-2507"
adapter = "CLRafaelR/Qwen3-4B-Instruct-2507-20260302_T105208"

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/structured_data_with_cot_dataset_512_v2

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.