🔎 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

    
def f(array):
    result = []
    index = 0
    while index < len(array):
        result.append(array.pop())
        index += 2
    return result
assert f([8, 8, -4, -9, 2, 8, -1, 8]) == [8, -1, 8]
CRUXEval-I 
assert f(??) == [8, -1, 8]
 

            		
          

          

            
              CRUXEval-O
              
assert f([8, 8, -4, -9, 2, 8, -1, 8]) == ??

              

              

            		
          

        

      


      
      

      

        
        
        
        

          546 / 800
        

      


      
      

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