conversations listlengths 2 6 | system stringclasses 4
values |
|---|---|
[
{
"from": "human",
"value": "What is 278 + 63?"
},
{
"from": "gpt",
"value": "278 + 63 = 341. The sum is 341."
}
] | null |
[
{
"from": "human",
"value": "What is 85% of 380?"
},
{
"from": "gpt",
"value": "85% of 380 = 380 × 85/100 = 323."
}
] | null |
[
{
"from": "human",
"value": "Write a Python function add_all that takes a list of numbers and returns their sum. For example, calling add_all([-10, -5, 54, -12, 58]) should return 85."
},
{
"from": "gpt",
"value": "Here's a clean implementation:\n\n```python\ndef add_all(values):\n return sum... | null |
[
{
"from": "human",
"value": "What is 4/7 + 5/9?"
},
{
"from": "gpt",
"value": "The common denominator is 63: 36/63 + 35/63 = 71/63, which simplifies to 71/63."
}
] | null |
[
{
"from": "human",
"value": "Which is larger: 1/4 or 2/3?"
},
{
"from": "gpt",
"value": "Cross-multiply: 1 × 3 = 3 vs 2 × 4 = 8. Since 3 is less than 8, 1/4 is the smaller one."
}
] | null |
[
{
"from": "human",
"value": "What is 551 - 248?"
},
{
"from": "gpt",
"value": "551 - 248 = 303. The difference is 303."
}
] | null |
[
{
"from": "human",
"value": "What does 671 divided by 11 equal?"
},
{
"from": "gpt",
"value": "671 ÷ 11 = 61, because 61 × 11 = 671."
}
] | null |
[
{
"from": "human",
"value": "What is 779 - 78?"
},
{
"from": "gpt",
"value": "779 - 78 = 701. The difference is 701."
}
] | null |
[
{
"from": "human",
"value": "True or false: Humans use only 10% of their brains."
},
{
"from": "gpt",
"value": "False. False — brain imaging shows activity across the whole brain; the myth is a movie trope."
}
] | null |
[
{
"from": "human",
"value": "Increase 140 by 35%. What is the result?"
},
{
"from": "gpt",
"value": "An increase of 35% adds 49, so the result is 140 + 49 = 189."
}
] | null |
[
{
"from": "human",
"value": "Five people sit in a row of five seats. Dan sits next to Jack Jack sits next to Finn Finn sits next to Chris Chris sits next to Iris Dan sits at one end of the row. Which seat is Finn in?"
},
{
"from": "gpt",
"value": "Starting from Dan at one end and following the '... | null |
[
{
"from": "human",
"value": "The average of 39, 59, 14, 48 and one more number is 65. What is the missing number?"
},
{
"from": "gpt",
"value": "Total of all 5 numbers = 65 × 5 = 325. Current sum = 160, so the missing number is 325 - 160 = 165."
}
] | null |
[
{
"from": "human",
"value": "The average of 55, 31, 17 and one more number is 65. What is the missing number?"
},
{
"from": "gpt",
"value": "Total of all 4 numbers = 65 × 4 = 260. Current sum = 103, so the missing number is 260 - 103 = 157."
}
] | null |
[
{
"from": "human",
"value": "Five people sit in a row of five seats. Alex sits next to Hugo Hugo sits next to Dan Dan sits next to Chris Chris sits next to Iris Alex sits at one end of the row. Which seat is Alex in?"
},
{
"from": "gpt",
"value": "Starting from Alex at one end and following the ... | null |
[
{
"from": "human",
"value": "Work out 195 + 966."
},
{
"from": "gpt",
"value": "195 + 966 = 1161. The sum is 1161."
}
] | null |
[
{
"from": "human",
"value": "90 is what percent of 100?"
},
{
"from": "gpt",
"value": "90/100 × 100 = 90%."
}
] | null |
[
{
"from": "human",
"value": "Calculate: 4 × 45"
},
{
"from": "gpt",
"value": "4 × 45 = 4 × 40 + 4 × 5 = 160 + 20 = 180."
}
] | null |
[
{
"from": "human",
"value": "A motorcycle consumes 5 litres per 100 km and has a 40-litre tank. How far can it travel on a full tank?"
},
{
"from": "gpt",
"value": "40 litres ÷ 5 litres per 100 km = 8, so it can travel 800 km."
}
] | null |
[
{
"from": "human",
"value": "A machine produces 245 bottles in 5 minutes. How many bottles does it produce per minute?"
},
{
"from": "gpt",
"value": "245 ÷ 5 = 49 bottles per minute."
}
] | null |
[
{
"from": "human",
"value": "What is the average (mean) of 78, 21, 51, 62?"
},
{
"from": "gpt",
"value": "Sum = 212, count = 4, so mean = 212 ÷ 4 = 53."
}
] | null |
[
{
"from": "human",
"value": "Write a Python function digital_sum that takes a positive integer and returns the sum of its digits. For example, calling digital_sum(77688) should return 36."
},
{
"from": "gpt",
"value": "Here's a clean implementation:\n\n```python\ndef digital_sum(x):\n return ... | null |
[
{
"from": "human",
"value": "What is 8 multiplied by 10?"
},
{
"from": "gpt",
"value": "8 × 10 = 8 × 1 × 10 = 8 × 10 = 80."
}
] | null |
[
{
"from": "human",
"value": "Convert 36 km/h to m/s."
},
{
"from": "gpt",
"value": "1 km/h = 1/3.6 m/s, so the answer is 10."
}
] | null |
[
{
"from": "human",
"value": "What is 769 plus 396?"
},
{
"from": "gpt",
"value": "769 + 396 = 1165. The sum is 1165."
}
] | null |
[
{
"from": "human",
"value": "Solve for x: 7x + 10 = 17"
},
{
"from": "gpt",
"value": "Subtract 10 from both sides: 7x = 7. Divide by 7: x = 7/7 = 1."
}
] | null |
[
{
"from": "human",
"value": "Work out 804 - 687."
},
{
"from": "gpt",
"value": "804 - 687 = 117. The difference is 117."
}
] | null |
[
{
"from": "human",
"value": "How does a magnet work?"
},
{
"from": "gpt",
"value": "Magnetism comes from the spin of electrons. In most materials, electron spins point in random directions and cancel out. In ferromagnetic materials like iron, nickel, and cobalt, regions called domains have align... | null |
[
{
"from": "human",
"value": "Write a Python function mult_of that takes a number n and a multiplier k, and returns a list of the first n multiples of k. For example, calling mult_of(6, 3) should return [3, 6, 9, 12, 15, 18]."
},
{
"from": "gpt",
"value": "Here's a clean implementation:\n\n```pyt... | null |
[
{
"from": "human",
"value": "The average of 29, 37, 21, 13 and one more number is 43. What is the missing number?"
},
{
"from": "gpt",
"value": "Total of all 5 numbers = 43 × 5 = 215. Current sum = 100, so the missing number is 215 - 100 = 115."
}
] | null |
[
{
"from": "human",
"value": "What is 1/3 + 5/7?"
},
{
"from": "gpt",
"value": "The common denominator is 21: 7/21 + 15/21 = 22/21, which simplifies to 22/21."
}
] | null |
[
{
"from": "human",
"value": "What is 18 × 87?"
},
{
"from": "gpt",
"value": "18 × 87 = 18 × 80 + 18 × 7 = 1440 + 126 = 1566."
}
] | null |
[
{
"from": "human",
"value": "Coach, my hands are killing me after bag work. Is that normal?"
},
{
"from": "gpt",
"value": "A little soreness, sure. Pain in the knuckles and wrist is NOT normal — that's your body telling you the wraps are wrong or the technique is off. Show me your wraps: they sh... | You are a tough but caring boxing coach. You push, you correct, you motivate — but you always keep technique and safety first. Short, direct, encouraging. You believe showing up is 90% of the win. |
[
{
"from": "human",
"value": "This function should return the sum of all integers from 1 to n, but it has a bug:\n\n```python\ndef sum_to(x):\n total = 0\n for i in range(x):\n total += i\n return total\n```\n\nFind and fix the bug."
},
{
"from": "gpt",
"value": "The bug is: range... | null |
[
{
"from": "human",
"value": "Write a Python function evens that takes a list of numbers and returns a new list with only the even numbers. For example, calling evens([-3, -16, -19, -15, 55]) should return [-16]."
},
{
"from": "gpt",
"value": "Here's a clean implementation:\n\n```python\ndef even... | null |
[
{
"from": "human",
"value": "How do I deal with a taller opponent?"
},
{
"from": "gpt",
"value": "Taller opponent means longer reach — so the first rule is: don't stand in their range trading punches. You win by breaking the distance. Work in three ways: 1) Angles — step off the center line when... | You are a tough but caring boxing coach. You push, you correct, you motivate — but you always keep technique and safety first. Short, direct, encouraging. You believe showing up is 90% of the win. |
[
{
"from": "human",
"value": "74 is what percent of 100?"
},
{
"from": "gpt",
"value": "74/100 × 100 = 74%."
}
] | null |
[
{
"from": "human",
"value": "Write a Python function gcd that takes two positive integers and returns their greatest common divisor. For example, calling gcd(30, 4) should return 2."
},
{
"from": "gpt",
"value": "Here's a clean implementation:\n\n```python\ndef gcd(first, second):\n while sec... | null |
[
{
"from": "human",
"value": "Calculate: 809 + 178"
},
{
"from": "gpt",
"value": "809 + 178 = 987. The sum is 987."
}
] | null |
[
{
"from": "human",
"value": "Five people sit in a row of five seats. Gina sits next to Chris Chris sits next to Iris Iris sits next to Jack Jack sits next to Hugo Gina sits at one end of the row. Which seat is Hugo in?"
},
{
"from": "gpt",
"value": "Starting from Gina at one end and following th... | null |
[
{
"from": "human",
"value": "Decrease 380 by 10%. What is the result?"
},
{
"from": "gpt",
"value": "A decrease of 10% removes 38, so the result is 380 - 38 = 342."
}
] | null |
[
{
"from": "human",
"value": "Write a Python function squares that takes a list of numbers and returns a new list with each number squared. For example, calling squares([20, -17, 30, 2, -25, 55]) should return [400, 289, 900, 4, 625, 3025]."
},
{
"from": "gpt",
"value": "Here's a clean implementa... | null |
[
{
"from": "human",
"value": "The average of 31, 31, 46 and one more number is 52. What is the missing number?"
},
{
"from": "gpt",
"value": "Total of all 4 numbers = 52 × 4 = 208. Current sum = 108, so the missing number is 208 - 108 = 100."
}
] | null |
[
{
"from": "human",
"value": "You know that Alex is shorter than Eva, Eva is shorter than Iris, Iris is shorter than Hugo. Who is the second shortest?"
},
{
"from": "gpt",
"value": "Putting the heights in order: Alex, Eva, Iris, Hugo (shortest to tallest). So the second shortest is Eva."
}
] | null |
[
{
"from": "human",
"value": "The average of 39, 54, 11, 49, 55 and one more number is 75. What is the missing number?"
},
{
"from": "gpt",
"value": "Total of all 6 numbers = 75 × 6 = 450. Current sum = 208, so the missing number is 450 - 208 = 242."
}
] | null |
[
{
"from": "human",
"value": "Write a Python function find_gcd that takes two positive integers and returns their greatest common divisor. For example, calling find_gcd(32, 40) should return 8."
},
{
"from": "gpt",
"value": "Here's a clean implementation:\n\n```python\ndef find_gcd(x, right):\n ... | null |
[
{
"from": "human",
"value": "You know that Gina is shorter than Ben, Ben is shorter than Jack, Jack is shorter than Eva, Eva is shorter than Alex. Who is the shortest?"
},
{
"from": "gpt",
"value": "Putting the heights in order: Gina, Ben, Jack, Eva, Alex (shortest to tallest). So the shortest i... | null |
[
{
"from": "human",
"value": "Work out 696 + 114."
},
{
"from": "gpt",
"value": "696 + 114 = 810. The sum is 810."
}
] | null |
[
{
"from": "human",
"value": "A rider travels at 70 km/h for 2 hours, then at 60 km/h for 3 hours. What is the total distance traveled?"
},
{
"from": "gpt",
"value": "First leg: 70 × 2 = 140 km. Second leg: 60 × 3 = 180 km. Total: 140 + 180 = 320 km."
}
] | null |
[
{
"from": "human",
"value": "Elena buys 3 packs of coffee at €4.50 each and 2 bags of pasta at €1.30 each. How much does Elena pay in total?"
},
{
"from": "gpt",
"value": "3 × €4.50 = €13.50, and 2 × €1.30 = €2.60. Total: €16.10."
}
] | null |
[
{
"from": "human",
"value": "Write a Python function keep_evens that takes a list of numbers and returns a new list with only the even numbers. For example, calling keep_evens([-3, -3, 38, -23, 59, 16, 43]) should return [38, 16]."
},
{
"from": "gpt",
"value": "Here's a clean implementation:\n\n... | null |
[
{
"from": "human",
"value": "Solve for x: 12x - 21 = 63"
},
{
"from": "gpt",
"value": "Add 21 to both sides: 12x = 84. Divide by 12: x = 84/12 = 7."
}
] | null |
[
{
"from": "human",
"value": "Write a Python function string_length that takes a string and returns its length. For example, calling string_length('cjacajggdbbffhf') should return 15."
},
{
"from": "gpt",
"value": "Here's a clean implementation:\n\n```python\ndef string_length(text):\n return ... | null |
[
{
"from": "human",
"value": "What is 4/9 + 4/9?"
},
{
"from": "gpt",
"value": "The common denominator is 9: 4/9 + 4/9 = 8/9, which simplifies to 8/9."
}
] | null |
[
{
"from": "human",
"value": "The average of 50, 36, 35, 11, 44 and one more number is 56. What is the missing number?"
},
{
"from": "gpt",
"value": "Total of all 6 numbers = 56 × 6 = 336. Current sum = 176, so the missing number is 336 - 176 = 160."
}
] | null |
[
{
"from": "human",
"value": "Tom is 4 times as old as his sister. His sister is 8 years old. How old is Tom?"
},
{
"from": "gpt",
"value": "8 × 4 = 32 years old."
}
] | null |
[
{
"from": "human",
"value": "A recipe for 2 servings needs 12 grams of flour. How much flour is needed for 3 servings?"
},
{
"from": "gpt",
"value": "12 g is for 2 servings, so one serving needs 6 g. For 3 servings: 6 × 3 = 18 g."
}
] | null |
[
{
"from": "human",
"value": "Write a Python function check_anagram that takes two strings and returns True if they are anagrams of each other (same letters, ignoring case and spaces). For example, calling check_anagram('race', 'race') should return True."
},
{
"from": "gpt",
"value": "Here's a c... | null |
[
{
"from": "human",
"value": "Write a Python function occurrences that takes a string and a character, and returns how many times the character appears in the string. For example, calling occurrences('bdddae') should return the correct result."
},
{
"from": "gpt",
"value": "Here's a clean impleme... | null |
[
{
"from": "human",
"value": "What is 42 × 37?"
},
{
"from": "gpt",
"value": "42 × 37 = 42 × 30 + 42 × 7 = 1260 + 294 = 1554."
}
] | null |
[
{
"from": "human",
"value": "28 is what percent of 100?"
},
{
"from": "gpt",
"value": "28/100 × 100 = 28%."
}
] | null |
[
{
"from": "human",
"value": "Five people sit in a row of five seats. Gina sits next to Eva Eva sits next to Finn Finn sits next to Jack Jack sits next to Ben Gina sits at one end of the row. Which seat is Gina in?"
},
{
"from": "gpt",
"value": "Starting from Gina at one end and following the 'ne... | null |
[
{
"from": "human",
"value": "Kostas buys 2 cans of tuna at €1.60 each and 1 loaves of bread at €2.40 each. How much does Kostas pay in total?"
},
{
"from": "gpt",
"value": "2 × €1.60 = €3.20, and 1 × €2.40 = €2.40. Total: €5.60."
}
] | null |
[
{
"from": "human",
"value": "86 is what percent of 100?"
},
{
"from": "gpt",
"value": "86/100 × 100 = 86%."
}
] | null |
[
{
"from": "human",
"value": "You know that Hugo is shorter than Eva, Eva is shorter than Gina, Gina is shorter than Alex. Who is the shortest?"
},
{
"from": "gpt",
"value": "Putting the heights in order: Hugo, Eva, Gina, Alex (shortest to tallest). So the shortest is Hugo."
}
] | null |
[
{
"from": "human",
"value": "67 is what percent of 100?"
},
{
"from": "gpt",
"value": "67/100 × 100 = 67%."
}
] | null |
[
{
"from": "human",
"value": "Elena buys 3 chocolate bars at €1.10 each and 2 apples at €1.20 each, and pays with a €50 note. How much change does Elena get?"
},
{
"from": "gpt",
"value": "First the total: 3 × €1.10 + 2 × €1.20 = €5.70. Then the change: €50 − €5.70 = €44.30."
}
] | null |
[
{
"from": "human",
"value": "Write a Python function keep_evens that takes a list of numbers and returns a new list with only the even numbers. For example, calling keep_evens([49, 37, -1]) should return []."
},
{
"from": "gpt",
"value": "Here's a clean implementation:\n\n```python\ndef keep_eve... | null |
[
{
"from": "human",
"value": "Five people sit in a row of five seats. Iris sits next to Chris Chris sits next to Gina Gina sits next to Alex Alex sits next to Hugo Iris sits at one end of the row. Which seat is Iris in?"
},
{
"from": "gpt",
"value": "Starting from Iris at one end and following th... | null |
[
{
"from": "human",
"value": "Write 87/50 as a decimal."
},
{
"from": "gpt",
"value": "Dividing 87 by 50 gives 1.74."
}
] | null |
[
{
"from": "human",
"value": "A motorcycle travels at 80 km/h for 3 hours. How far does it travel?"
},
{
"from": "gpt",
"value": "Distance = speed × time = 80 × 3 = 240 km."
}
] | null |
[
{
"from": "human",
"value": "Solve for x: x/4 = 7"
},
{
"from": "gpt",
"value": "Multiply both sides by 4: x = 7 × 4 = 28."
}
] | null |
[
{
"from": "human",
"value": "Write a Python function last_item that takes a list and returns its last element. For example, calling last_item([-6, 8, 59, 33, 8, -28, 54]) should return 54."
},
{
"from": "gpt",
"value": "Here's a clean implementation:\n\n```python\ndef last_item(lst):\n return... | null |
[
{
"from": "human",
"value": "A rider travels at 60 km/h for 4 hours, then at 80 km/h for 3 hours. What is the total distance traveled?"
},
{
"from": "gpt",
"value": "First leg: 60 × 4 = 240 km. Second leg: 80 × 3 = 240 km. Total: 240 + 240 = 480 km."
}
] | null |
[
{
"from": "human",
"value": "Calculate: 335 ÷ 5"
},
{
"from": "gpt",
"value": "335 ÷ 5 = 67, because 67 × 5 = 335."
}
] | null |
[
{
"from": "human",
"value": "What is 333 + 197?"
},
{
"from": "gpt",
"value": "333 + 197 = 530. The sum is 530."
}
] | null |
[
{
"from": "human",
"value": "All soccer players are athletes. Bella is a soccer player. Is Bella a athlete?"
},
{
"from": "gpt",
"value": "Yes. The first statement says every soccer player belongs to athletes, and Bella is a soccer player. So Bella must be a athlete."
}
] | null |
[
{
"from": "human",
"value": "Write a Python function is_digits that takes a string and returns True if it contains only digits. For example, calling is_digits('quick lazy') should return False."
},
{
"from": "gpt",
"value": "Here's a clean implementation:\n\n```python\ndef is_digits(s):\n ret... | null |
[
{
"from": "human",
"value": "Calculate: 882 - 632"
},
{
"from": "gpt",
"value": "882 - 632 = 250. The difference is 250."
}
] | null |
[
{
"from": "human",
"value": "What is the average (mean) of 87, 29, 42, 77, 46, 54?"
},
{
"from": "gpt",
"value": "Sum = 335, count = 6, so mean = 335 ÷ 6 = 55.8."
}
] | null |
[
{
"from": "human",
"value": "You know that Iris is shorter than Jack, Jack is shorter than Dan, Dan is shorter than Chris, Chris is shorter than Ben. Who is the second shortest?"
},
{
"from": "gpt",
"value": "Putting the heights in order: Iris, Jack, Dan, Chris, Ben (shortest to tallest). So the... | null |
[
{
"from": "human",
"value": "You know that Jack is shorter than Eva, Eva is shorter than Chris, Chris is shorter than Dan. Who is the second tallest?"
},
{
"from": "gpt",
"value": "Putting the heights in order: Jack, Eva, Chris, Dan (shortest to tallest). So the second tallest is Chris."
}
] | null |
[
{
"from": "human",
"value": "Why do we see our breath on cold days?"
},
{
"from": "gpt",
"value": "Your breath is warm and full of water vapor from your lungs. When it hits cold air, the vapor rapidly cools below its dew point and condenses into tiny liquid droplets — a visible cloud, the same p... | null |
[
{
"from": "human",
"value": "How many continents are there?"
},
{
"from": "gpt",
"value": "Seven: Africa, Antarctica, Asia, Europe, North America, Oceania, South America."
}
] | null |
[
{
"from": "human",
"value": "Three times a number is 30. What is the number?"
},
{
"from": "gpt",
"value": "Number = 30 ÷ 3 = 10."
}
] | null |
[
{
"from": "human",
"value": "What is 4 multiplied by 86?"
},
{
"from": "gpt",
"value": "4 × 86 = 4 × 80 + 4 × 6 = 320 + 24 = 344."
}
] | null |
[
{
"from": "human",
"value": "Five people sit in a row of five seats. Iris sits next to Alex Alex sits next to Dan Dan sits next to Eva Eva sits next to Finn Iris sits at one end of the row. Which seat is Dan in?"
},
{
"from": "gpt",
"value": "Starting from Iris at one end and following the 'next... | null |
[
{
"from": "human",
"value": "What is 3/4 + 1/7?"
},
{
"from": "gpt",
"value": "The common denominator is 28: 21/28 + 4/28 = 25/28, which simplifies to 25/28."
}
] | null |
[
{
"from": "human",
"value": "How should I maintain my motorcycle chain?"
},
{
"from": "gpt",
"value": "A clean, properly tensioned chain is the difference between a smooth ride and a snapped chain. Weekly: check slack — about 25-35mm of play mid-span for most bikes; too tight wears the sprockets... | null |
[
{
"from": "human",
"value": "Write a Python function reverse_string that takes a string and returns it reversed. For example, calling reverse_string('igie') should return 'eigi'."
},
{
"from": "gpt",
"value": "Here's a clean implementation:\n\n```python\ndef reverse_string(text):\n return tex... | null |
[
{
"from": "human",
"value": "Decrease 220 by 45%. What is the result?"
},
{
"from": "gpt",
"value": "A decrease of 45% removes 99, so the result is 220 - 99 = 121."
}
] | null |
[
{
"from": "human",
"value": "Why do we get brain freeze?"
},
{
"from": "gpt",
"value": "Brain freeze — the sharp pain behind your forehead when you eat something cold fast — is a nerve reflex, not actual freezing. The cold hits the roof of your mouth, where a cluster of nerves called the sphenop... | null |
[
{
"from": "human",
"value": "Write a Python function double_all that takes a list of numbers and returns a new list with each number doubled. For example, calling double_all([-23, 52, -6]) should return [-46, 104, -12]."
},
{
"from": "gpt",
"value": "Here's a clean implementation:\n\n```python\n... | null |
[
{
"from": "human",
"value": "A machine produces 700 bottles in 14 minutes. How many bottles does it produce per minute?"
},
{
"from": "gpt",
"value": "700 ÷ 14 = 50 bottles per minute."
}
] | null |
[
{
"from": "human",
"value": "Rewrite this for a younger audience:\n\n'Regular physical activity is associated with numerous health benefits, including improved cardiovascular function, enhanced mood regulation, and better sleep quality. It is recommended that adults engage in at least 150 minutes of moderat... | null |
[
{
"from": "human",
"value": "Tom is 3 times as old as his sister. His sister is 7 years old. How old is Tom?"
},
{
"from": "gpt",
"value": "7 × 3 = 21 years old."
}
] | null |
[
{
"from": "human",
"value": "Write a Python function double_each that takes a list of numbers and returns a new list with each number doubled. For example, calling double_each([28, -13, -9, 48, 39, 11, 35]) should return [56, -26, -18, 96, 78, 22, 70]."
},
{
"from": "gpt",
"value": "Here's a cle... | null |
[
{
"from": "human",
"value": "This function should take a temperature in Celsius and return it in Fahrenheit, but it has a bug:\n\n```python\ndef celsius_to_fahrenheit(c):\n return c * 5 / 9 + 32\n```\n\nFind and fix the bug."
},
{
"from": "gpt",
"value": "The bug is: The formula was inverted:... | null |
[
{
"from": "human",
"value": "Work out 346 - 225."
},
{
"from": "gpt",
"value": "346 - 225 = 121. The difference is 121."
}
] | null |
[
{
"from": "human",
"value": "How many metres are in 41 km?"
},
{
"from": "gpt",
"value": "1 km = 1000 m, so the answer is 41000."
}
] | null |
[
{
"from": "human",
"value": "How does a helicopter fly?"
},
{
"from": "gpt",
"value": "A helicopter generates lift with its main rotor — spinning airfoils that push air downward, and the air pushes back up (Newton's third law). To move forward, the pilot tilts the rotor disk forward, so the lift... | null |
End of preview. Expand in Data Studio
1B-SFT — High-Quality Instruction Dataset for a ~1B LLM
A curated, deduplicated instruction-tuning dataset in ShareGPT format, built for fine-tuning a 1B-class model (Qwen2.5-1.5B, Llama-3.2-1B, Gemma-2-2B).
5,235 samples — train 4,919 / val 158 / test 158.
Quality guarantees
- Math: every answer is computed by the generator (arithmetic, percentages, word problems, multi-step problems, unit conversions, linear equations, sequences, fractions) with step-by-step working.
- Coding: every code sample is executed and unit-tested before inclusion (33 task types + 10 bug-fix tasks); the shipped code blocks were re-executed against the final files (1,047/1,047 pass).
- Reasoning: ordering puzzles, liar-guard puzzles, syllogisms, seating puzzles — all consistent by construction.
- Hand-written core: ~260 manually authored seeds across knowledge, explain, creative writing, roleplay (flirty companion, boxing coach, motorcycle buddy, DM), advice, summarization, rewriting, and multi-turn chat.
- Exact + SimHash near-duplicate removal; artifact filtering; deterministic split (94/3/3).
Format (ShareGPT, one JSON per line)
{"system": "optional persona...",
"conversations": [
{"from": "human", "value": "..."},
{"from": "gpt", "value": "..."}]}
Category breakdown
| Category | Count |
|---|---|
| math | 3,251 |
| coding | 1,055 |
| reasoning | 549 |
| trivia | 140 |
| knowledge | 52 |
| roleplay | 47 |
| creative | 42 |
| explain | 36 |
| advice | 17 |
| rewrite | 17 |
| chat | 16 |
| summarization | 13 |
Loading
from datasets import load_dataset
ds = load_dataset("andro124543/1b-sft-dataset", split="train")
or use train.jsonl / val.jsonl / test.jsonl directly with LLaMA-Factory (sharegpt formatting) or axolotl.
Suggested recipe for a 1B model (SFT)
- Base:
Qwen/Qwen2.5-1.5B(ChatML) ormeta-llama/Llama-3.2-1B - lr 2e-5 cosine, batch 8–16, 3 epochs, max_seq_len 1024–2048, packing on
- QLoRA 4-bit (rank 32) for limited compute; export to GGUF for Ollama
Rebuild / extend
The dataset is fully reproducible: content.py (hand-written seeds), generators.py (verified programmatic generators), build.py (dedup → filter → split → stats). Run python build.py to regenerate deterministically.
- Downloads last month
- 58