Files
rlvrcodemathif-qwen2.5-1.5b/eval-results/hendrycks_math/metrics.json

11 lines
186 B
JSON
Raw Normal View History

{
"hendrycks_math": {
"pass@1": {
"num_entries": 5000,
"avg_tokens": 648,
"gen_seconds": 164,
"symbolic_correct": 55.36,
"no_answer": 5.88
}
}
}