Test fixture
Arithmetic, symbolic steps, and structured problem solving.
The model receives the prompt (and optional system message). The run uses scorer exact_match with the JSON configuration below. Pass/fail and partial credit are determined entirely by that scorer against the model output; no human grading.
Sequence: 5, 9, 17, 33, ?. Return only the next number.
{
"expected": "65"
}temperature
0
max_tokens
48
timeout (s)
120
type
scored
file
reasoning_medium_10.json