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

16 lines
429 B
JSON
Raw Normal View History

{
"ifeval": {
"pass@1": {
"num_prompts": 541,
"num_instructions": 834,
"average_score": 71.13392465325336,
"prompt_strict_accuracy": 65.06469500924214,
"instruction_strict_accuracy": 74.10071942446042,
"prompt_loose_accuracy": 68.39186691312385,
"instruction_loose_accuracy": 76.97841726618705,
"num_entries": 541,
"avg_tokens": 480,
"gen_seconds": 58
}
}
}