metadata dict | messages listlengths 2 2 |
|---|---|
{
"answers": [
{
"group": "GEOLOGICAL MATERIALS",
"level": 0,
"members": [
"QUARTZ",
"MICA",
"SAND",
"GRANITE"
]
},
{
"group": "SPORTS",
"level": 1,
"members": [
"BASKETBALL",
"SWIMMING",
"TENNIS",
"CRICKET"
]
},
{
"group": "MEAL COURSES",
"level": 2,
"members": [
"DESSERT",
"APPETIZER",
"ENTREE",
"SNACK"
]
},
{
"group": "CULTURAL DISHES",
"level": 3,
"members": [
"CEVICHE",
"SUSHI",
"TACO",
"BURGER"
]
}
],
"custom_id": "train_1",
"reasoning_length": 3500
} | [
{
"content": "Solve this Connections puzzle by finding 4 groups of 4 related words:\nWords: QUARTZ, MICA, SAND, GRANITE, BASKETBALL, SWIMMING, TENNIS, CRICKET, DESSERT, APPETIZER, ENTREE, SNACK, CEVICHE, SUSHI, TACO, BURGER",
"role": "user"
},
{
"content": "Alright, let's break down this Connections... |
{
"answers": [
{
"group": "NATURAL DISASTERS",
"level": 0,
"members": [
"VOLCANO",
"TORNADO",
"HURRICANE",
"EARTHQUAKE"
]
},
{
"group": "MUSICAL INSTRUMENTS",
"level": 1,
"members": [
"PIANO",
"VIOLIN",
"DRUM",
"FLUTE"
]
},
{
"group": "CULINARY PROFESSIONS",
"level": 2,
"members": [
"CHEF",
"WAITER",
"BAKER",
"SOMMELIER"
]
},
{
"group": "TRAVEL ACTIVITIES",
"level": 3,
"members": [
"TRAVEL",
"ADVENTURE",
"RELAXATION",
"EXPLORATION"
]
}
],
"custom_id": "train_2",
"reasoning_length": 3600
} | [
{
"content": "Solve this Connections puzzle by finding 4 groups of 4 related words:\nWords: VOLCANO, TORNADO, HURRICANE, EARTHQUAKE, PIANO, VIOLIN, DRUM, FLUTE, CHEF, WAITER, BAKER, SOMMELIER, TRAVEL, ADVENTURE, RELAXATION, EXPLORATION",
"role": "user"
},
{
"content": "Let's analyze this Connections... |
{
"answers": [
{
"group": "ASTRONOMICAL TERMS",
"level": 0,
"members": [
"GALAXY",
"PLANET",
"STAR",
"COMET"
]
},
{
"group": "FORMS OF ART",
"level": 1,
"members": [
"DANCE",
"SING",
"PAINT",
"SCULPT"
]
},
{
"group": "TRANSPORTATION MODES",
"level": 2,
"members": [
"BICYCLE",
"SKATEBOARD",
"SCOOTER",
"ROLLERBLADES"
]
},
{
"group": "FRUITS",
"level": 3,
"members": [
"STRAWBERRY",
"BANANA",
"APPLE",
"GRAPE"
]
}
],
"custom_id": "train_3",
"reasoning_length": 3400
} | [
{
"content": "Solve this Connections puzzle by finding 4 groups of 4 related words:\nWords: GALAXY, PLANET, STAR, COMET, DANCE, SING, PAINT, SCULPT, BICYCLE, SKATEBOARD, SCOOTER, ROLLERBLADES, STRAWBERRY, BANANA, APPLE, GRAPE",
"role": "user"
},
{
"content": "Let’s solve this Connections puzzle. The... |
{
"answers": [
{
"group": "TIMES OF DAY",
"level": 0,
"members": [
"HORIZON",
"SUNSET",
"NIGHT",
"DAWN"
]
},
{
"group": "FRUITS",
"level": 1,
"members": [
"APPLE",
"ORANGE",
"GRAPEFRUIT",
"LEMON"
]
},
{
"group": "SPORTS",
"level": 2,
"members": [
"MARTIAL ARTS",
"BASKETBALL",
"SOCCER",
"TENNIS"
]
},
{
"group": "WATERCRAFT",
"level": 3,
"members": [
"SAILBOAT",
"YACHT",
"KAYAK",
"CANOE"
]
}
],
"custom_id": "train_4",
"reasoning_length": 3300
} | [
{
"content": "Solve this Connections puzzle by finding 4 groups of 4 related words:\nWords: HORIZON, SUNSET, NIGHT, DAWN, APPLE, ORANGE, GRAPEFRUIT, LEMON, MARTIAL ARTS, BASKETBALL, SOCCER, TENNIS, SAILBOAT, YACHT, KAYAK, CANOE",
"role": "user"
},
{
"content": "Let's solve this Connections puzzle. T... |
{
"answers": [
{
"group": "DAIRY PRODUCTS",
"level": 0,
"members": [
"HONEY",
"MILK",
"CHEESE",
"YOGURT"
]
},
{
"group": "VEGETABLES",
"level": 1,
"members": [
"SPINACH",
"BROCCOLI",
"CARROT",
"LETTUCE"
]
},
{
"group": "PHYSICAL ACTIVITIES",
"level": 2,
"members": [
"HIKING",
"RUNNING",
"SWIMMING",
"CYCLING"
]
},
{
"group": "LITERARY FORMS",
"level": 3,
"members": [
"NOVEL",
"POEM",
"ESSAY",
"SHORT STORY"
]
}
],
"custom_id": "train_5",
"reasoning_length": 3700
} | [
{
"content": "Solve this Connections puzzle by finding 4 groups of 4 related words:\nWords: HONEY, MILK, CHEESE, YOGURT, SPINACH, BROCCOLI, CARROT, LETTUCE, HIKING, RUNNING, SWIMMING, CYCLING, NOVEL, POEM, ESSAY, SHORT STORY",
"role": "user"
},
{
"content": "Let's analyze this Connections puzzle. Th... |
nyt-connections-cot-enhanced
Dataset Description
The Enhanced Connections Puzzle Dataset is designed to facilitate the development and evaluation of models in the realm of word association and reasoning tasks. This dataset features a collection of puzzles where users are tasked with identifying four groups of related words from a given list. Each entry consists of detailed reasoning that outlines the thought process behind the grouping, making it a valuable resource for researchers and developers. Key improvements include increased diversity in themes and vocabulary combinations, optimized reasoning processes for clarity and efficiency, and the introduction of puzzles with varying difficulty levels to enhance challenge and engagement. This dataset provides a clear JSON structure that includes comprehensive messages and metadata, making it easy to analyze and utilize in machine learning applications. This dataset only contains sample data. For more data please visit https://www.mobiusi.com/?refer=huggingface or contact us at contact@mobiusi.com.
Usage
from datasets import load_dataset
dataset = load_dataset("Mobiusi/nyt-connections-cot-enhanced")
Created: 2025-10-20 12:37:48
- Downloads last month
- 22