Cell

tower_qwen3.5-397b_basic_s2hp

Qwen3.5-397B on the tower (MT0→MT23w), 1 run. Not cited by the thesis (development or earlier run).

Outcome
Highest floor reached, per runMT4
Died0
Final HP, mean2
Seconds per model call26.6
Parse error rate0.0%
Settings
FloorTower MT0→MT23w
ModelQwen3.5-397B (qwen3.5-397b-a17b, thinking)
Planningreactive
Promptengine-default-tower-basic
Temperature0
Lookahead depth2
Mechanical overrideson (logged per turn)
Engine settings (environment, 1)
THINKING_MAX_TOKENS=8192
Runsopen one to replay it turn by turn
#outcomefinal HPHP / optturnsmodel callsreplay
1MT42–6130replay
Re-run this cell

In a checkout of mt-ai-lab with an API key set (the result goes to a new file rerun_<time>_tower_qwen3.5-397b_basic_s2hp, never over the thesis result):

python scripts/run_catalog.py --codes T5

Result file: results/reliability/tower_qwen3.5-397b_basic_s2hp.json · updated 2026-09-27