Chan-Y/Opus-4.6-Reasoning-3000x-filtered-tr

---
dataset_info:
features:
- name: problem
dtype: string
- name: thinking
dtype: string
- name: solution
dtype: string
- name: difficulty
dtype: string
- name: category
dtype: string
splits:
- name: train
num_bytes: 5935468
num_examples: 2326
download_size: 3322553
dataset_size: 5935468
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
language:
- tr
---
# 🇹🇷 Turkish Translation of Opus-4.6 Reasoning Dataset

## Dataset Description

* **Name:** `Opus-4.6-Reasoning-3000x-filtered-tr`
* **Original Dataset:** `nohurry/Opus-4.6-Reasoning-3000x-filtered`
* **Language:** Turkish (translated from English)
* **Size:** ~3000 samples
* **Task:** Reasoning / Instruction Following / Chain-of-Thought

This dataset is a Turkish translation of the original *Opus-4.6 Reasoning 3000x filtered* dataset. The goal is to provide high-quality reasoning data for Turkish LLM training, evaluation, and alignment tasks.

---

## Motivation

High-quality reasoning datasets in Turkish are limited. This dataset aims to:

* Improve Turkish LLM reasoning capabilities
* Enable better instruction-following in Turkish
* Support fine-tuning and evaluation for Turkish AI systems

---

## Data Source

* Original dataset: [`nohurry/Opus-4.6-Reasoning-3000x-filtered`](https://huggingface.co/datasets/nohurry/Opus-4.6-Reasoning-3000x-filtered)
* The dataset consists of curated reasoning samples with structured prompts and responses.

---

## Translation Details

* **Model used:** [`tencent/HY-MT1.5-1.8B`](https://huggingface.co/tencent/HY-MT1.5-1.8B)
* **Method:** Fully automatic translation

数据集信息：
特征字段：
- 字段名：problem，数据类型：字符串
- 字段名：thinking，数据类型：字符串
- 字段名：solution，数据类型：字符串
- 字段名：difficulty，数据类型：字符串
- 字段名：category，数据类型：字符串
数据划分：
- 划分名称：train，数据字节数：5935468，样本数量：2326
下载体积：3322553 字节，数据集总大小：5935468 字节
配置项：
- 配置名称：default，数据文件：
- 对应划分：train，文件路径：data/train-*
支持语言：
- tr

# 🇹🇷 Opus-4.6 推理数据集土耳其语翻译版

## 数据集说明
* **数据集名称**：`Opus-4.6-Reasoning-3000x-filtered-tr`
* **原始数据集**：`nohurry/Opus-4.6-Reasoning-3000x-filtered`
* **语言属性**：土耳其语（由英语翻译生成）
* **样本规模**：约3000条样本
* **任务范畴**：推理/指令遵循/思维链（Chain-of-Thought）

本数据集为原始*Opus-4.6 推理3000x过滤版*数据集的土耳其语翻译版本，旨在为土耳其语大语言模型（Large Language Model，LLM）的训练、评估与对齐任务提供高质量的推理数据支撑。

## 项目动机
当前面向土耳其语的高质量推理数据集较为匮乏，本数据集旨在实现以下目标：
* 提升土耳其语大语言模型的推理能力
* 优化土耳其语场景下的指令遵循效果
* 为土耳其语AI系统的微调与评估提供支撑

## 数据来源
* 原始数据集：[`nohurry/Opus-4.6-Reasoning-3000x-filtered`](https://huggingface.co/datasets/nohurry/Opus-4.6-Reasoning-3000x-filtered)
* 本数据集包含经过精心筛选的推理样本，均带有结构化的提示词与回复内容。

## 翻译细节
* **所用翻译模型**：[`tencent/HY-MT1.5-1.8B`](https://huggingface.co/tencent/HY-MT1.5-1.8B)
* **翻译方法**：全自动机器翻译