{"batch_policy":{"choice_candidates":16,"code_test_workers":4,"generation_examples":8,"padded_token_budget":8192},"batch_timings":[{"candidates":16,"kind":"choice","padded_tokens":1680,"seconds":0.18977543292567134},{"candidates":16,"kind":"choice","padded_tokens":2176,"seconds":0.13158546795602888},{"candidates":16,"kind":"choice","padded_tokens":2912,"seconds":0.1604797359323129},{"candidates":16,"kind":"choice","padded_tokens":4608,"seconds":0.25351069995667785}],"dataset_revision":"35b264d03638db9f4ce671b711558bf7ff0f80d5","device":"cuda:0","evaluator_source":true,"evidence":"CodeSOTA execution on White","execution_profile":"batched-v2","finished":1792198120,"gpu":1,"id":"c6d083942f8d93dc672e019b","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":3407441920,"profile":"pilot-32","protocol":{"adapter":"reading","area":"Language","benchmark":"BoolQ","contract":"Passage + question \u2192 yes / no","dataset":"google/boolq","dataset_config":null,"dataset_revision":"35b264d03638db9f4ce671b711558bf7ff0f80d5","dataset_url":"https://huggingface.co/datasets/google/boolq","direction":"max","id":"reading","license":["cc-by-sa-3.0"],"limit":32,"metric":"accuracy","models":[{"id":"Qwen/Qwen2.5-0.5B","revision":"060db6499f32faf8b98477b0a26969ef7d8b9987"},{"id":"Qwen/Qwen2.5-1.5B","revision":"8faed761d45a263340a0528343f099c05c9a4323"}],"name":"Reading comprehension","protocol_version":"codesota-pilot-v2-batched","runner_file":"evaluate_fast.py","source":"https://github.com/google-research-datasets/boolean-questions","split":"validation"},"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":7.779247667989694,"measurement_version":"gpu-slot-v1"},"runner_sha256":"bc899fb160f6881408863cd8bcdf96102dd6d8b10559a5f781a8cd816103da77","scope":"50f1e323d677230fdcedb032817fe2f0387aab1b8fc6c2b0f2282f32d4318845","score":0.6875,"seconds":2.026138138025999,"seed":42,"selection":"first 32 valid examples in pinned split order","status":"completed","task":"reading","worker":"white-gpu1"}