{"label":"gs_llama8b_bare","preset":"llama-3.1-8b","task_model":"meta-llama-3.1-8b-instruct","task_mode":"instruct","obs":"bare","train":["simple","gen01","gen02"],"test":["gen03","gen04","gen05","gen06","gen07","gen08","gen09","gen10","gen11","gen12","gen13","gen14","gen15"],"repeat":1,"env":{"CONVERSATION_HISTORY_TURNS":"0","MOTA_ORDER_TWO_SIDED":"1","MOTA_FIGHT_KEY_FLOOD":"1","LOOKAHEAD_DEPTH":"3","MOTA_BARE_OBS":"1","MOTA_EVAL_TEMPERATURE":"0.0","THINKING_MAX_TOKENS":"16384","MOTA_DISABLE_OVERRIDES":null,"MOTA_SUPPRESS_HINTS":null},"overrides":"mechanical-only (dead-end / unwinnable / lethal) + auto-collect + auto-fight-0-HP","episodes":[{"prompt":"core","floor":"simple","split":"train","rep":0,"goal_score":0.325,"success":false,"hp_ratio":null,"died":false,"final_hp":397,"hero_atk":13,"hero_def":8,"enemies_killed":13,"enemies_total":14,"llm_calls":16,"parse_errors":0,"forced_turns":20,"lethal_only_stop":null,"run_log":"20260928_012837_080250_simple_llm.json","elapsed_s":53.3},{"prompt":"gepa","floor":"simple","split":"train","rep":0,"goal_score":0.325,"success":false,"hp_ratio":null,"died":false,"final_hp":395,"hero_atk":13,"hero_def":8,"enemies_killed":13,"enemies_total":14,"llm_calls":15,"parse_errors":0,"forced_turns":21,"lethal_only_stop":null,"run_log":"20260928_012930_406864_simple_llm.json","elapsed_s":63.9},{"prompt":"basic","floor":"simple","split":"train","rep":0,"goal_score":0.275,"success":false,"hp_ratio":null,"died":false,"final_hp":179,"hero_atk":13,"hero_def":6,"enemies_killed":11,"enemies_total":14,"llm_calls":14,"parse_errors":3,"forced_turns":14,"lethal_only_stop":null,"run_log":"20260928_013034_298561_simple_llm.json","elapsed_s":76.9},{"prompt":"core","floor":"gen01","split":"train","rep":0,"goal_score":0.744,"success":true,"hp_ratio":0.4881,"died":false,"final_hp":82,"hero_atk":14,"hero_def":7,"enemies_killed":10,"enemies_total":10,"llm_calls":8,"parse_errors":0,"forced_turns":17,"lethal_only_stop":null,"run_log":"20260928_013151_237196_gen01.txt_llm.json","elapsed_s":25.9},{"prompt":"gepa","floor":"gen01","split":"train","rep":0,"goal_score":0.759,"success":true,"hp_ratio":0.5179,"died":false,"final_hp":87,"hero_atk":14,"hero_def":7,"enemies_killed":10,"enemies_total":10,"llm_calls":8,"parse_errors":0,"forced_turns":17,"lethal_only_stop":null,"run_log":"20260928_013217_144184_gen01.txt_llm.json","elapsed_s":31.4},{"prompt":"basic","floor":"gen01","split":"train","rep":0,"goal_score":0.7619,"success":true,"hp_ratio":0.5238,"died":false,"final_hp":88,"hero_atk":14,"hero_def":7,"enemies_killed":10,"enemies_total":10,"llm_calls":8,"parse_errors":3,"forced_turns":17,"lethal_only_stop":null,"run_log":"20260928_013248_597149_gen01.txt_llm.json","elapsed_s":45.8},{"prompt":"core","floor":"gen02","split":"train","rep":0,"goal_score":0.3208,"success":false,"hp_ratio":null,"died":false,"final_hp":290,"hero_atk":14,"hero_def":7,"enemies_killed":11,"enemies_total":12,"llm_calls":11,"parse_errors":0,"forced_turns":14,"lethal_only_stop":null,"run_log":"20260928_013334_341816_gen02.txt_llm.json","elapsed_s":37.2},{"prompt":"gepa","floor":"gen02","split":"train","rep":0,"goal_score":0.3208,"success":false,"hp_ratio":null,"died":false,"final_hp":302,"hero_atk":14,"hero_def":7,"enemies_killed":11,"enemies_total":12,"llm_calls":11,"parse_errors":0,"forced_turns":14,"lethal_only_stop":null,"run_log":"20260928_013411_522196_gen02.txt_llm.json","elapsed_s":37.8},{"prompt":"basic","floor":"gen02","split":"train","rep":0,"goal_score":0.5345,"success":true,"hp_ratio":0.069,"died":false,"final_hp":16,"hero_atk":14,"hero_def":7,"enemies_killed":12,"enemies_total":12,"llm_calls":11,"parse_errors":1,"forced_turns":15,"lethal_only_stop":null,"run_log":"20260928_013449_352518_gen02.txt_llm.json","elapsed_s":49.5},{"prompt":"core","floor":"gen03","split":"test","rep":0,"goal_score":0.2864,"success":false,"hp_ratio":null,"died":false,"final_hp":488,"hero_atk":13,"hero_def":6,"enemies_killed":9,"enemies_total":11,"llm_calls":8,"parse_errors":0,"forced_turns":12,"lethal_only_stop":null,"run_log":"20260928_013538_852093_gen03.txt_llm.json","elapsed_s":26.2},{"prompt":"gepa","floor":"gen03","split":"test","rep":0,"goal_score":0.2864,"success":false,"hp_ratio":null,"died":false,"final_hp":506,"hero_atk":13,"hero_def":6,"enemies_killed":9,"enemies_total":11,"llm_calls":10,"parse_errors":0,"forced_turns":10,"lethal_only_stop":null,"run_log":"20260928_013605_039438_gen03.txt_llm.json","elapsed_s":33.4},{"prompt":"basic","floor":"gen03","split":"test","rep":0,"goal_score":0.2864,"success":false,"hp_ratio":null,"died":false,"final_hp":488,"hero_atk":13,"hero_def":6,"enemies_killed":9,"enemies_total":11,"llm_calls":7,"parse_errors":2,"forced_turns":13,"lethal_only_stop":null,"run_log":"20260928_013638_477723_gen03.txt_llm.json","elapsed_s":59.4},{"prompt":"core","floor":"gen04","split":"test","rep":0,"goal_score":0.7,"success":true,"hp_ratio":0.4,"died":false,"final_hp":90,"hero_atk":14,"hero_def":8,"enemies_killed":11,"enemies_total":11,"llm_calls":8,"parse_errors":0,"forced_turns":21,"lethal_only_stop":null,"run_log":"20260928_013737_866093_gen04.txt_llm.json","elapsed_s":26.2},{"prompt":"gepa","floor":"gen04","split":"test","rep":0,"goal_score":0.7245,"success":true,"hp_ratio":0.4489,"died":false,"final_hp":101,"hero_atk":14,"hero_def":8,"enemies_killed":11,"enemies_total":11,"llm_calls":9,"parse_errors":0,"forced_turns":20,"lethal_only_stop":null,"run_log":"20260928_013804_046969_gen04.txt_llm.json","elapsed_s":29.1},{"prompt":"basic","floor":"gen04","split":"test","rep":0,"goal_score":0.3182,"success":false,"hp_ratio":null,"died":false,"final_hp":321,"hero_atk":13,"hero_def":8,"enemies_killed":10,"enemies_total":11,"llm_calls":8,"parse_errors":0,"forced_turns":17,"lethal_only_stop":null,"run_log":"20260928_013833_188503_gen04.txt_llm.json","elapsed_s":34.5},{"prompt":"core","floor":"gen05","split":"test","rep":0,"goal_score":0.3231,"success":false,"hp_ratio":null,"died":false,"final_hp":108,"hero_atk":14,"hero_def":8,"enemies_killed":12,"enemies_total":13,"llm_calls":13,"parse_errors":0,"forced_turns":14,"lethal_only_stop":null,"run_log":"20260928_013907_706249_gen05.txt_llm.json","elapsed_s":42.5},{"prompt":"gepa","floor":"gen05","split":"test","rep":0,"goal_score":0.3231,"success":false,"hp_ratio":null,"died":false,"final_hp":185,"hero_atk":14,"hero_def":8,"enemies_killed":12,"enemies_total":13,"llm_calls":13,"parse_errors":0,"forced_turns":14,"lethal_only_stop":null,"run_log":"20260928_013950_217424_gen05.txt_llm.json","elapsed_s":44},{"prompt":"basic","floor":"gen05","split":"test","rep":0,"goal_score":0.3231,"success":false,"hp_ratio":null,"died":false,"final_hp":174,"hero_atk":14,"hero_def":8,"enemies_killed":12,"enemies_total":13,"llm_calls":13,"parse_errors":2,"forced_turns":14,"lethal_only_stop":null,"run_log":"20260928_014034_231687_gen05.txt_llm.json","elapsed_s":64},{"prompt":"core","floor":"gen06","split":"test","rep":0,"goal_score":0.9422,"success":true,"hp_ratio":0.8843,"died":false,"final_hp":214,"hero_atk":14,"hero_def":7,"enemies_killed":4,"enemies_total":8,"llm_calls":3,"parse_errors":0,"forced_turns":15,"lethal_only_stop":null,"run_log":"20260928_014138_237668_gen06.txt_llm.json","elapsed_s":10.1},{"prompt":"gepa","floor":"gen06","split":"test","rep":0,"goal_score":0.8347,"success":true,"hp_ratio":0.6694,"died":false,"final_hp":162,"hero_atk":14,"hero_def":7,"enemies_killed":5,"enemies_total":8,"llm_calls":4,"parse_errors":0,"forced_turns":15,"lethal_only_stop":null,"run_log":"20260928_014148_366414_gen06.txt_llm.json","elapsed_s":15.8},{"prompt":"basic","floor":"gen06","split":"test","rep":0,"goal_score":0.7769,"success":true,"hp_ratio":0.5537,"died":false,"final_hp":134,"hero_atk":14,"hero_def":7,"enemies_killed":5,"enemies_total":8,"llm_calls":4,"parse_errors":0,"forced_turns":15,"lethal_only_stop":null,"run_log":"20260928_014204_139693_gen06.txt_llm.json","elapsed_s":16.7},{"prompt":"core","floor":"gen07","split":"test","rep":0,"goal_score":0.8491,"success":true,"hp_ratio":0.6982,"died":false,"final_hp":155,"hero_atk":14,"hero_def":8,"enemies_killed":11,"enemies_total":14,"llm_calls":8,"parse_errors":0,"forced_turns":18,"lethal_only_stop":null,"run_log":"20260928_014220_884398_gen07.txt_llm.json","elapsed_s":26.3},{"prompt":"gepa","floor":"gen07","split":"test","rep":0,"goal_score":0.7883,"success":true,"hp_ratio":0.5766,"died":false,"final_hp":128,"hero_atk":14,"hero_def":9,"enemies_killed":14,"enemies_total":14,"llm_calls":11,"parse_errors":0,"forced_turns":19,"lethal_only_stop":null,"run_log":"20260928_014247_228728_gen07.txt_llm.json","elapsed_s":37.1},{"prompt":"basic","floor":"gen07","split":"test","rep":0,"goal_score":0.804,"success":true,"hp_ratio":0.6081,"died":false,"final_hp":135,"hero_atk":14,"hero_def":9,"enemies_killed":14,"enemies_total":14,"llm_calls":12,"parse_errors":0,"forced_turns":18,"lethal_only_stop":null,"run_log":"20260928_014324_305872_gen07.txt_llm.json","elapsed_s":56.1},{"prompt":"core","floor":"gen08","split":"test","rep":0,"goal_score":0.1909,"success":false,"hp_ratio":null,"died":false,"final_hp":22,"hero_atk":14,"hero_def":8,"enemies_killed":6,"enemies_total":11,"llm_calls":7,"parse_errors":0,"forced_turns":14,"lethal_only_stop":{"action":"Fight Bat at (9,6) dmg=72","hp_after":-50,"damage":72},"run_log":"20260928_014420_429897_gen08.txt_llm.json","elapsed_s":26.2},{"prompt":"gepa","floor":"gen08","split":"test","rep":0,"goal_score":0.1591,"success":false,"hp_ratio":null,"died":false,"final_hp":74,"hero_atk":13,"hero_def":7,"enemies_killed":5,"enemies_total":11,"llm_calls":6,"parse_errors":1,"forced_turns":12,"lethal_only_stop":{"action":"Fight Bat at (4,2) dmg=90","hp_after":-16,"damage":90},"run_log":"20260928_014446_602194_gen08.txt_llm.json","elapsed_s":25.8},{"prompt":"basic","floor":"gen08","split":"test","rep":0,"goal_score":0.1591,"success":false,"hp_ratio":null,"died":false,"final_hp":74,"hero_atk":13,"hero_def":7,"enemies_killed":5,"enemies_total":11,"llm_calls":6,"parse_errors":1,"forced_turns":12,"lethal_only_stop":{"action":"Fight Bat at (4,2) dmg=90","hp_after":-16,"damage":90},"run_log":"20260928_014512_394765_gen08.txt_llm.json","elapsed_s":34.5},{"prompt":"core","floor":"gen09","split":"test","rep":0,"goal_score":0.7721,"success":true,"hp_ratio":0.5441,"died":false,"final_hp":111,"hero_atk":14,"hero_def":7,"enemies_killed":9,"enemies_total":11,"llm_calls":8,"parse_errors":0,"forced_turns":16,"lethal_only_stop":null,"run_log":"20260928_014546_877012_gen09.txt_llm.json","elapsed_s":26.6},{"prompt":"gepa","floor":"gen09","split":"test","rep":0,"goal_score":0.5931,"success":true,"hp_ratio":0.1863,"died":false,"final_hp":38,"hero_atk":14,"hero_def":8,"enemies_killed":11,"enemies_total":11,"llm_calls":11,"parse_errors":0,"forced_turns":18,"lethal_only_stop":null,"run_log":"20260928_014613_503520_gen09.txt_llm.json","elapsed_s":36.9},{"prompt":"basic","floor":"gen09","split":"test","rep":0,"goal_score":0.6887,"success":true,"hp_ratio":0.3775,"died":false,"final_hp":77,"hero_atk":14,"hero_def":8,"enemies_killed":10,"enemies_total":11,"llm_calls":9,"parse_errors":2,"forced_turns":19,"lethal_only_stop":null,"run_log":"20260928_014650_428621_gen09.txt_llm.json","elapsed_s":59.8},{"prompt":"core","floor":"gen10","split":"test","rep":0,"goal_score":0.2917,"success":false,"hp_ratio":null,"died":false,"final_hp":247,"hero_atk":14,"hero_def":6,"enemies_killed":10,"enemies_total":12,"llm_calls":9,"parse_errors":0,"forced_turns":15,"lethal_only_stop":null,"run_log":"20260928_014750_222677_gen10.txt_llm.json","elapsed_s":29.1},{"prompt":"gepa","floor":"gen10","split":"test","rep":0,"goal_score":0.2917,"success":false,"hp_ratio":null,"died":false,"final_hp":271,"hero_atk":14,"hero_def":6,"enemies_killed":10,"enemies_total":12,"llm_calls":9,"parse_errors":0,"forced_turns":15,"lethal_only_stop":null,"run_log":"20260928_014819_357460_gen10.txt_llm.json","elapsed_s":31.5},{"prompt":"basic","floor":"gen10","split":"test","rep":0,"goal_score":0.2917,"success":false,"hp_ratio":null,"died":false,"final_hp":311,"hero_atk":14,"hero_def":6,"enemies_killed":10,"enemies_total":12,"llm_calls":9,"parse_errors":4,"forced_turns":15,"lethal_only_stop":null,"run_log":"20260928_014850_887654_gen10.txt_llm.json","elapsed_s":73.1},{"prompt":"core","floor":"gen11","split":"test","rep":0,"goal_score":0.1556,"success":false,"hp_ratio":null,"died":false,"final_hp":16,"hero_atk":10,"hero_def":6,"enemies_killed":4,"enemies_total":9,"llm_calls":5,"parse_errors":0,"forced_turns":6,"lethal_only_stop":null,"run_log":"20260928_015004_001154_gen11.txt_llm.json","elapsed_s":16.5},{"prompt":"gepa","floor":"gen11","split":"test","rep":0,"goal_score":0.1944,"success":false,"hp_ratio":null,"died":false,"final_hp":426,"hero_atk":13,"hero_def":7,"enemies_killed":5,"enemies_total":9,"llm_calls":7,"parse_errors":0,"forced_turns":14,"lethal_only_stop":null,"run_log":"20260928_015020_480457_gen11.txt_llm.json","elapsed_s":26.4},{"prompt":"basic","floor":"gen11","split":"test","rep":0,"goal_score":0.1167,"success":false,"hp_ratio":null,"died":false,"final_hp":109,"hero_atk":9,"hero_def":6,"enemies_killed":3,"enemies_total":9,"llm_calls":5,"parse_errors":0,"forced_turns":5,"lethal_only_stop":{"action":"Fight Bat at (7,1) dmg=154","hp_after":-45,"damage":154},"run_log":"20260928_015046_880938_gen11.txt_llm.json","elapsed_s":24.9},{"prompt":"core","floor":"gen12","split":"test","rep":0,"goal_score":0.1273,"success":false,"hp_ratio":null,"died":false,"final_hp":741,"hero_atk":13,"hero_def":4,"enemies_killed":4,"enemies_total":11,"llm_calls":5,"parse_errors":0,"forced_turns":10,"lethal_only_stop":null,"run_log":"20260928_015111_789207_gen12.txt_llm.json","elapsed_s":16.4},{"prompt":"gepa","floor":"gen12","split":"test","rep":0,"goal_score":0.1273,"success":false,"hp_ratio":null,"died":false,"final_hp":765,"hero_atk":13,"hero_def":4,"enemies_killed":4,"enemies_total":11,"llm_calls":5,"parse_errors":0,"forced_turns":10,"lethal_only_stop":null,"run_log":"20260928_015128_228479_gen12.txt_llm.json","elapsed_s":17.9},{"prompt":"basic","floor":"gen12","split":"test","rep":0,"goal_score":0.1273,"success":false,"hp_ratio":null,"died":false,"final_hp":741,"hero_atk":13,"hero_def":4,"enemies_killed":4,"enemies_total":11,"llm_calls":5,"parse_errors":1,"forced_turns":10,"lethal_only_stop":null,"run_log":"20260928_015146_132315_gen12.txt_llm.json","elapsed_s":26.4},{"prompt":"core","floor":"gen13","split":"test","rep":0,"goal_score":0.175,"success":false,"hp_ratio":null,"died":false,"final_hp":45,"hero_atk":10,"hero_def":6,"enemies_killed":6,"enemies_total":12,"llm_calls":4,"parse_errors":0,"forced_turns":7,"lethal_only_stop":{"action":"Fight Slime Lord at (7,2) dmg=60","hp_after":-15,"damage":60},"run_log":"20260928_015212_567700_gen13.txt_llm.json","elapsed_s":17.2},{"prompt":"gepa","floor":"gen13","split":"test","rep":0,"goal_score":0.2333,"success":false,"hp_ratio":null,"died":false,"final_hp":2,"hero_atk":10,"hero_def":8,"enemies_killed":8,"enemies_total":12,"llm_calls":6,"parse_errors":0,"forced_turns":9,"lethal_only_stop":{"action":"Fight Boss at (5,1) dmg=999999999","hp_after":-1,"damage":999999999},"run_log":"20260928_015229_806914_gen13.txt_llm.json","elapsed_s":27.4},{"prompt":"basic","floor":"gen13","split":"test","rep":0,"goal_score":0.5242,"success":true,"hp_ratio":0.0483,"died":false,"final_hp":10,"hero_atk":14,"hero_def":9,"enemies_killed":10,"enemies_total":12,"llm_calls":7,"parse_errors":0,"forced_turns":18,"lethal_only_stop":null,"run_log":"20260928_015257_171653_gen13.txt_llm.json","elapsed_s":29.9},{"prompt":"core","floor":"gen14","split":"test","rep":0,"goal_score":0.3111,"success":false,"hp_ratio":null,"died":false,"final_hp":443,"hero_atk":13,"hero_def":9,"enemies_killed":8,"enemies_total":9,"llm_calls":7,"parse_errors":0,"forced_turns":15,"lethal_only_stop":null,"run_log":"20260928_015327_089151_gen14.txt_llm.json","elapsed_s":24.2},{"prompt":"gepa","floor":"gen14","split":"test","rep":0,"goal_score":0.3111,"success":false,"hp_ratio":null,"died":false,"final_hp":461,"hero_atk":13,"hero_def":9,"enemies_killed":8,"enemies_total":9,"llm_calls":7,"parse_errors":1,"forced_turns":15,"lethal_only_stop":null,"run_log":"20260928_015351_276775_gen14.txt_llm.json","elapsed_s":27.9},{"prompt":"basic","floor":"gen14","split":"test","rep":0,"goal_score":0.9569,"success":true,"hp_ratio":0.9139,"died":false,"final_hp":138,"hero_atk":13,"hero_def":9,"enemies_killed":6,"enemies_total":9,"llm_calls":5,"parse_errors":0,"forced_turns":15,"lethal_only_stop":null,"run_log":"20260928_015419_178523_gen14.txt_llm.json","elapsed_s":20.9},{"prompt":"core","floor":"gen15","split":"test","rep":0,"goal_score":0.882,"success":true,"hp_ratio":0.7639,"died":false,"final_hp":165,"hero_atk":14,"hero_def":9,"enemies_killed":10,"enemies_total":12,"llm_calls":8,"parse_errors":0,"forced_turns":20,"lethal_only_stop":null,"run_log":"20260928_015440_066572_gen15.txt_llm.json","elapsed_s":27.5},{"prompt":"gepa","floor":"gen15","split":"test","rep":0,"goal_score":0.3208,"success":false,"hp_ratio":null,"died":false,"final_hp":204,"hero_atk":14,"hero_def":9,"enemies_killed":11,"enemies_total":12,"llm_calls":9,"parse_errors":0,"forced_turns":20,"lethal_only_stop":null,"run_log":"20260928_015507_531514_gen15.txt_llm.json","elapsed_s":32.2},{"prompt":"basic","floor":"gen15","split":"test","rep":0,"goal_score":0.9166,"success":true,"hp_ratio":0.8333,"died":false,"final_hp":180,"hero_atk":14,"hero_def":9,"enemies_killed":10,"enemies_total":12,"llm_calls":8,"parse_errors":0,"forced_turns":20,"lethal_only_stop":null,"run_log":"20260928_015539_711820_gen15.txt_llm.json","elapsed_s":32.7}],"started":"2026-09-28T03:07:00.852156","gepa":{"result_file":"results/gepa/gepa_result_gs_llama8b_bare.json","best_score":0.46923333333333334,"trajectory":[0.4633,0.4692,0.4692,0.4692,0.4692,0.4692],"accepted":[true,true,true,true,true,null],"best_is_seed":false,"finished":"2026-09-28T03:28:37.080250"},"updated":"2026-09-28T03:56:12.401191","summary":{"train/core":{"n":3,"success":1,"deaths":0,"goal_mean":0.4633},"train/gepa":{"n":3,"success":1,"deaths":0,"goal_mean":0.4683},"train/basic":{"n":3,"success":2,"deaths":0,"goal_mean":0.5238},"test/core":{"n":13,"success":5,"deaths":0,"goal_mean":0.462},"test/gepa":{"n":13,"success":4,"deaths":0,"goal_mean":0.3991},"test/basic":{"n":13,"success":6,"deaths":0,"goal_mean":0.4838},"all/core":{"n":16,"success":6,"deaths":0,"goal_mean":0.4623},"all/gepa":{"n":16,"success":5,"deaths":0,"goal_mean":0.412},"all/basic":{"n":16,"success":8,"deaths":0,"goal_mean":0.4913}},"finished":"2026-09-28T03:56:12.401191"}