Datasets:
messages listlengths 3 3 |
|---|
[
{
"role": "system",
"content": "You are a highly capable AI assistant. For every response, you must first show your step-by-step reasoning process within a /think block, followed by the final answer."
},
{
"role": "user",
"content": "Evaluate the integral\n$$\\int\\frac{\\sin(x)}{\\sin(x)+\\cos(... |
[
{
"role": "system",
"content": "You are a highly capable AI assistant. For every response, you must first show your step-by-step reasoning process within a /think block, followed by the final answer."
},
{
"role": "user",
"content": "Given the sequence \\(a_1 = 1\\), \\(a_2 = 3\\), \\(a_3 = 9\\)... |
[
{
"role": "system",
"content": "You are a highly capable AI assistant. For every response, you must first show your step-by-step reasoning process within a /think block, followed by the final answer."
},
{
"role": "user",
"content": "If \\( \\log(xy^3) = 1 \\) and \\( \\log(x^2y) = 1 \\), what i... |
[
{
"role": "system",
"content": "You are a highly capable AI assistant. For every response, you must first show your step-by-step reasoning process within a /think block, followed by the final answer."
},
{
"role": "user",
"content": "What universal scaling laws govern the steady-state probabilit... |
[
{
"role": "system",
"content": "You are a highly capable AI assistant. For every response, you must first show your step-by-step reasoning process within a /think block, followed by the final answer."
},
{
"role": "user",
"content": "Israel has a three-tier court system. At the lowest level are ... |
[
{
"role": "system",
"content": "You are a highly capable AI assistant. For every response, you must first show your step-by-step reasoning process within a /think block, followed by the final answer."
},
{
"role": "user",
"content": "Find all functions \\( f: \\mathbb{R} \\to \\mathbb{R}^+ \\) t... |
[
{
"role": "system",
"content": "You are a highly capable AI assistant. For every response, you must first show your step-by-step reasoning process within a /think block, followed by the final answer."
},
{
"role": "user",
"content": "Write a Python3 script that processes log files from a robotic... |
[
{
"role": "system",
"content": "You are a highly capable AI assistant. For every response, you must first show your step-by-step reasoning process within a /think block, followed by the final answer."
},
{
"role": "user",
"content": "Generate an executable Python function generated from the give... |
[
{
"role": "system",
"content": "You are a highly capable AI assistant. For every response, you must first show your step-by-step reasoning process within a /think block, followed by the final answer."
},
{
"role": "user",
"content": "Why is normal ordering even a valid operation in the first pla... |
[
{
"role": "system",
"content": "You are a highly capable AI assistant. For every response, you must first show your step-by-step reasoning process within a /think block, followed by the final answer."
},
{
"role": "user",
"content": "Return your final response within \\boxed{}. Consider the rail... |
π§ Reasoning Mix 100k β¨
A high-quality, balanced reasoning dataset consisting of 99,999 samples extracted from three reasoning datasets. This dataset is specifically formatted for models to utilize a /think block for step-by-step reasoning.
π Dataset Summary
The Reasoning Mix 100k is a curated collection of reasoning tasks, primarily focused on mathematics, logic, and general problem-solving. It combines the strengths of three high-performing datasets into a unified format optimized for Chain-of-Thought (CoT) training.
| Source | Samples | Format |
|---|---|---|
| GLM-5.1-Reasoning-1M | 33,333 | General/Math Reasoning |
| OpenMathReasoning | 33,333 | Mathematical Chain-of-Thought |
| OpenThoughts-114k | 33,333 | Comprehensive Reasoning |
| Total | 99,999 | ChatML (Reasoning Optimized) |
π οΈ Formatting
The dataset uses the ChatML format with a specialized system prompt and assistant response structure.
π System Prompt
"You are a highly capable AI assistant. For every response, you must first show your step-by-step reasoning process within a /think block, followed by the final answer."
π€ Assistant Responses
Every response is wrapped in a /think block structure:
/think
[Step-by-step reasoning process...]
/think
[Final answer or solution]
π Dataset Structure
Each sample in the reasoning_mix_100k.jsonl file follows this structure:
{
"messages": [
{
"role": "system",
"content": "You are a highly capable AI assistant..."
},
{
"role": "user",
"content": "Evaluate the integral..."
},
{
"role": "assistant",
"content": "/think\n[Reasoning...]\n/think\n\n[Final Answer]"
}
]
}
π Usage
You can load this dataset using the Hugging Face datasets library:
from datasets import load_dataset
dataset = load_dataset("pinkpixel/reasoning-mix-100k")
print(dataset["train"][0])
π License
This dataset is released under the Apache License 2.0.
Made with β€οΈ by Pink Pixel β¨
- Downloads last month
- -