Files
yao-meta-skill/evals/sample_trigger_report.json
T

38 lines
1.0 KiB
JSON

{
"threshold": 0.42,
"threshold_explanation": "Prompts at or above the threshold are treated as trigger matches. Final scores combine token overlap, positive-token bonuses, negative-token penalties, and explicit negative-pattern penalties. Scores near the threshold should be reviewed as boundary cases.",
"false_positives": 0,
"false_negatives": 0,
"precision": 1.0,
"recall": 1.0,
"bucket_stats": {
"should_trigger": {
"total": 8,
"passed": 8,
"pass_rate": 1.0
},
"should_not_trigger": {
"total": 7,
"passed": 7,
"pass_rate": 1.0
},
"near_neighbor": {
"total": 7,
"passed": 7,
"pass_rate": 1.0
}
},
"comparison": {
"baseline_false_positives": 0,
"baseline_false_negatives": 8,
"improved_false_positives": 0,
"improved_false_negatives": 0,
"false_positive_delta": 0,
"false_negative_delta": -8,
"baseline_precision": null,
"improved_precision": 1.0,
"baseline_recall": 0.0,
"improved_recall": 1.0
}
}