RaindropA practical guide to choosing decision modelsGitHub ↗

Rule application · question 54

Back to Rule application · All capabilities

Question ID: cruxeval/c78bd20037af9eacc078
Source: CRUXEval
Input format: adapted choices

Explore this question’s altered versions

Context

def f(text, n):
    length = len(text)
    return text[length*(n%4):length ]

Question

What does f('abc', 1) return?

Answer choices

  • A: 'b'
  • B: 'a'
  • C: '0'
  • D: ''

Reference answer: '' (D)

Exact evaluated input as JSON
{
  "state": "def f(text, n):\n    length = len(text)\n    return text[length*(n%4):length ]",
  "question": "What does f('abc', 1) return?",
  "choices": {
    "A": "'b'",
    "B": "'a'",
    "C": "'0'",
    "D": "''"
  }
}

Recorded model answers

Model Answer Outcome
Astra '' Correct
Opus 5.5 '' Correct
GLiDE v1.2 '' Correct
Decider v1.1 '' Correct
Jev '' Correct
Microsoft Decision-1 'a' Incorrect
Clef '' Correct
Luna Decisions '' Correct
Clef Flash '' Correct
Strands Decider '' Correct
GLiNER2.5 Decide 'b' Incorrect
Laya 'b' Incorrect
CalDec v1 'b' Incorrect

Source and sharing

Attribution: Meta; CRUXEval authors.
Terms: MIT. Source sharing decision.
The capability assignment and choice-format normalization are part of this experiment. Source data retain their own terms.

Exact source locator and input fingerprint
{
  "source_locator": {
    "path": "data/cruxeval.jsonl",
    "record": "598",
    "kind": "source_file_record"
  },
  "source_revision": "190faf16d175b5847b0af05d937872b1fb395942",
  "input_hash": "8fd45ad52951829f6383ad3bcd6f78cadf47f16d503554521e660871c079c4cb"
}

Type to search questions, variants and documentation.