🔎 CRUXEval Sample Explorer 🔎

CRUXEval is a benchmark complementary to HumanEval and MBPP measuring code reasoning, understanding, and execution capabilities!

Homepage Paper Code HF Dataset Leaderboard

model: temp:

def f(s1,s2):
    res = []
    i = s1.rfind(s2)
    while i != -1:
        res.append(i+len(s2)-1)
        i = s1.rfind(s2, 0, i)
    return res
assert f('abcdefghabc', 'abc') == [10, 2]

CRUXEval-I assert f(??) == [10, 2]

CRUXEval-O assert f('abcdefghabc', 'abc') == ??

The site is inspired by and based on Minerva sample explorer and Llemma. Thanks!