{"batch_policy":{"choice_candidates":16,"code_test_workers":4,"generation_examples":8,"padded_token_budget":8192},"batch_timings":[{"candidates":16,"kind":"choice","padded_tokens":208,"seconds":0.1246656629955396},{"candidates":16,"kind":"choice","padded_tokens":240,"seconds":0.021783375996164978},{"candidates":16,"kind":"choice","padded_tokens":256,"seconds":0.022670169943012297},{"candidates":16,"kind":"choice","padded_tokens":288,"seconds":0.02873803593683988},{"candidates":16,"kind":"choice","padded_tokens":304,"seconds":0.02725917799398303},{"candidates":16,"kind":"choice","padded_tokens":384,"seconds":0.030699505005031824},{"candidates":16,"kind":"choice","padded_tokens":432,"seconds":0.034868279937654734},{"candidates":16,"kind":"choice","padded_tokens":800,"seconds":0.055141029064543545}],"dataset_revision":"388097ea7776314e93a529163e0fea805b8a6454","device":"cuda:0","evaluator_source":true,"evidence":"CodeSOTA execution on White","execution_profile":"batched-v2","finished":1792198159,"gpu":1,"id":"9fd52efbe4976b40a9fd6646","measured_cost_usd":null,"metric":"accuracy","model":"Qwen/Qwen2.5-1.5B","model_revision":"8faed761d45a263340a0528343f099c05c9a4323","n":32,"notes":"Pilot subset, CodeSOTA batched-v2 execution. Separate scope from serial v1; unchanged prompt/tokenization/metric contracts.","peak_gpu_bytes":3163561472,"profile":"pilot-32","protocol":{"adapter":"choice","area":"Language","benchmark":"OpenBookQA main (test)","contract":"Question + candidate answers \u2192 selected answer","dataset":"allenai/openbookqa","dataset_config":"main","dataset_revision":"388097ea7776314e93a529163e0fea805b8a6454","dataset_url":"https://huggingface.co/datasets/allenai/openbookqa","direction":"max","id":"openbookqa-main-choice","license":["unknown"],"limit":32,"metric":"accuracy","models":[{"id":"Qwen/Qwen2.5-0.5B","revision":"060db6499f32faf8b98477b0a26969ef7d8b9987"},{"id":"Qwen/Qwen2.5-1.5B","revision":"8faed761d45a263340a0528343f099c05c9a4323"}],"name":"OpenBookQA main (test)","parent_task":"reading","protocol_version":"codesota-choice-pilot-v2-batched","recipe":{"answer_field":"answerKey","answer_type":"label","choice_labels_field":"choices.label","choices_field":"choices.text","data_file":"main/test-00000-of-00001.parquet","question_field":"question_stem"},"runner_file":"evaluate_fast.py","source":"https://huggingface.co/datasets/allenai/openbookqa","split":"test"},"resource_metrics":{"basis":"Evaluator subprocess elapsed \u00d7 one reserved GPU; includes data/model loading and CPU scoring; not kernel-active time","gpu_count":1,"gpu_slot_seconds":4.821713284938596,"measurement_version":"gpu-slot-v1"},"runner_sha256":"bc899fb160f6881408863cd8bcdf96102dd6d8b10559a5f781a8cd816103da77","scope":"08d620ad5299cfd3b7b752304580d3c9ea87e7f135c266b7ae4e2d94c2fe51f6","score":0.40625,"seconds":1.7379119840916246,"seed":42,"selection":"first 32 valid examples in pinned split order","status":"completed","task":"openbookqa-main-choice","worker":"white-gpu1"}