mirror of
https://github.com/open-thought/reasoning-gym.git
synced 2026-04-28 17:29:39 +00:00
8 lines
No EOL
2.5 KiB
JSON
8 lines
No EOL
2.5 KiB
JSON
{
|
|
"question": "Brendan has a bag of marbles with 10 inside. He tripped over a pebble while carrying it and dropped half of them. He scrambled to search for them but only came up with 3. When he went back home, he inspected the marbles further. One of them he picked up wasn't a marble, but actually a bead so he got rid of it. How many marbles did Brendan end up with?",
|
|
"answer": "Brendan dropped his marbles and was left with 10/2=<<10/2=5>>5 marbles.\nHe searched and found some of his lost marbles, getting him back to 5+3=<<5+3=8>>8 marbles.\nHe went home and removed a bead, leaving him with 8-1=<<8-1=7>>7 marbles.\n#### 7",
|
|
"id_orig": 1247,
|
|
"id_shuffled": 61,
|
|
"question_annotated": "{name,Brendan} has a bag of {objects,marbles} with {n,10} inside. He tripped over a {obstacle,pebble} while carrying it and dropped {frac,half} of them. He scrambled to search for them but only came up with {k,3}. When he went back home, he inspected the {objects,marbles} further. {fake_num,one} of them he picked up weren't {objects,marbles}, but actually {fake_object,bead}, so he got rid of them. Seeing how upset {name,Brandon} was, his father gave him a pouch of {objects,marbles} with {extra,10} {objects,marbles} inside. How many {objects,marbles} did {name,Brendan} end up with?\n\n#init:\n- name = sample(names_male)\n- objects = sample(['marbles', 'coins', 'buttons', 'beads', 'pebbles'])\n- obstacle = sample(['rock', 'stick', 'toy', 'root'])\n- fake_object = sample(['buttons', 'coins', 'pebbles', 'beads'])\n- $fake_num = sample(numbers_within(2, 10))\n- $frac = sample(fraction_alnum)\n- $extra = range(5,12)\n- $n = range(10, 101, 2)\n- $k = range(3, 20)\n\n#conditions:\n- is_int(n * frac)\n- k < n * frac\n- k > fake_num\n\n#answer: int(n * (1-frac)) + k - fake_num + extra",
|
|
"answer_annotated": "{name} dropped his {objects} and was left with {n}*{1-frac}=<<{n}*{1-frac}={int(n*(1-frac))}>>{int(n*(1-frac))} {objects}.\nHe searched and found some of his lost {objects}, getting him back to {int(n*(1-frac))}+{k}=<<{int(n*(1-frac))}+{k}={int(n*(1-frac))+k}>>{int(n*(1-frac))+k} {objects}.\nHe went home and removed {fake_object}, leaving him with {int(n*(1-frac))+k}-{fake_num}=<<{int(n*(1-frac))+k}-fake_num={int(n*(1-frac))+k-fake_num}>>{int(n*(1-frac))+k-fake_num} {objects}.\nFinally, {name}'s father gave him {extra} {objects}. So he ended up with {int(n*(1-frac))+k-fake_num} + {extra} = <<{int(n*(1-frac))+k-fake_num}+{extra}={int(n * (1-frac)) + k - fake_num + extra} >>{ int(n * (1-frac)) + k - fake_num + extra} {objects}.\n#### { int(n * (1-frac)) + k - fake_num + extra}"
|
|
} |