polyphony · claude-opus-5-5
✓ pass≠ not comparableMetrics
| prose.rc | 1 |
| prose.rn_mean | 0.267 |
| prose.score | 1 (was 0.630) |
| prose.words | 2,796 (was 3,659) |
Labels
| cadence | on-demand |
| experiment | w17-r2-2 |
| harness_sha | 5bb2df25f2a62973f362c31d7cf2ee34bd6a833b |
| judge_provider | groq |
| judge_self_graded | False |
| repeat | 1 |
| trigger | workflow_dispatch |
Compared with the previous run
Baseline 2026-10-25 04:45 UTC · 100% → this run 100% · code diff ↗
- eval:prose: 0.63 → 1.00
Cases (1)
words=2796 path=planned rubrics={'R1': 1.0, 'R2': 1.0, 'R3': 1.0, 'R4': 1.0, 'R5': 1.0} seams=0/0scene 1 R1: All principal canon characters required (Aldous Kerr, Count Vasska) appear correctly and act in line with their descriptions, with no extraneous or mischaracterized figures. | scene 1 R2: All three untagged lines are correctly attributable and each character’s voice remains distinct and faithful to their samples. | scene 1 R3: The scene includes each planned beat in the correct order, with each described change occurring and no extra unplanned events. | scene 1 R4: The passage maintains consistent Aldous‑centered past‑tense POV, matches the eerie tone, and presents varied, publishable prose without structural errors. | scene 1 R5: The prose aligns with canon and planned beats without contradictions, meeting continuity criteria fully. | scene 2 R1: The scene includes exactly the required canon characters Aldous Kerr and Count Vasska, with correct names and behavior, and introduces no unauthorized major characters. | scene 2 R3: The text realizes each planned beat in order, with each described change occurring and no unplanned events added. | scene 2 R4: The passage maintains consistent first‑person‑limited past tense POV, matches the courtly yet unsettling tone, and presents varied, concrete sentences without any structural violations. | scene 2 R5: The prose aligns fully with the established canon and prior text, containing no contradictions. | scene 2 R2: not applicable: under two cast speakers with voice samples have attributed lines | scene 3 R1: All principal characters required (Aldous Kerr and Count Vasska) appear correctly with proper behavior, no missing or misused cast. | scene 3 R3: All five planned beats appear sequentially, each change occurs as described, with no extra or missing events. | scene 3 R4: The scene maintains consistent first‑person‑like limited POV in past tense, matches the intended tone, and is a polished, narrative draft without structural errors. | scene 3 R5: The prose aligns with canon and prior text, introduces no contradictions, and fulfills the planned beats without errors. | scene 3 R2: not applicable: under two cast speakers with voice samples have attributed lines