complex_qwen3.8-27b_reactive_raw
Qwen3.8-27B on the complex floor, 5 episodes. Not cited by the thesis (development or earlier run).
| Wins (boss defeated, alive) | 1/5 [0.04, 0.62] |
| Final HP / exact optimum, mean over wins | 100% |
| Died | 4 |
| Final HP, mean | -65 |
| Seconds per model call | – |
| Parse error rate | – |
| Floor | Complex floor |
| Model | Qwen3.8-27B (qwen3.8-27b, thinking) |
| Planning | reactive |
| Prompt | complex_bfs_strategy.txt |
| Temperature | 0 |
| Lookahead depth | 3 |
| Mechanical overrides | off (all engine replacements and assists disabled) |
| # | outcome | final HP | HP / opt | steps | model calls | replay |
|---|---|---|---|---|---|---|
| 1 | won | 22 | 32% | 52 | – | log not stored |
| 2 | died | -52 | – | 49 | – | log not stored |
| 3 | died | -52 | – | 49 | – | log not stored |
| 4 | died | -190 | – | 47 | – | log not stored |
| 5 | died | -52 | – | 50 | – | log not stored |
This cell has no entry in mt-ai-lab's re-run registry; its settings above and the result file describe how it was run.
Result file: results/reliability/complex_qwen3.8-27b_reactive_raw.json · updated 2026-09-21