{"batch_policy":{"choice_candidates":16,"code_test_workers":4,"generation_examples":8,"padded_token_budget":8192},"batch_timings":[{"candidates":16,"kind":"choice","padded_tokens":384,"seconds":0.13930650800466537},{"candidates":16,"kind":"choice","padded_tokens":448,"seconds":0.016007635043933988},{"candidates":16,"kind":"choice","padded_tokens":512,"seconds":0.016305911005474627},{"candidates":16,"kind":"choice","padded_tokens":560,"seconds":0.01750927499961108},{"candidates":16,"kind":"choice","padded_tokens":624,"seconds":0.018990446929819882},{"candidates":16,"kind":"choice","padded_tokens":672,"seconds":0.02015766396652907},{"candidates":16,"kind":"choice","padded_tokens":752,"seconds":0.02054631896317005},{"candidates":16,"kind":"choice","padded_tokens":1136,"seconds":0.02836571296211332}],"dataset_revision":"218ec52e09a7e7462a5400043bb9a69a41d06b76","device":"cuda:0","evaluator_source":true,"evidence":"CodeSOTA execution on White","execution_profile":"batched-v2","finished":1792198076,"gpu":1,"id":"2f544548b89cd6fed2d4e61d","measured_cost_usd":null,"metric":"accuracy","model":"Qwen/Qwen2.5-0.5B","model_revision":"060db6499f32faf8b98477b0a26969ef7d8b9987","n":32,"notes":"Pilot subset, CodeSOTA batched-v2 execution. Separate scope from serial v1; unchanged prompt/tokenization/metric contracts.","peak_gpu_bytes":1216680960,"profile":"pilot-32","protocol":{"adapter":"commonsense","area":"Reasoning","benchmark":"HellaSwag","contract":"Scenario + four endings \u2192 most plausible ending","dataset":"Rowan/hellaswag","dataset_config":null,"dataset_revision":"218ec52e09a7e7462a5400043bb9a69a41d06b76","dataset_url":"https://huggingface.co/datasets/Rowan/hellaswag","direction":"max","id":"commonsense","license":"See upstream terms","limit":32,"metric":"accuracy","models":[{"id":"Qwen/Qwen2.5-0.5B","revision":"060db6499f32faf8b98477b0a26969ef7d8b9987"},{"id":"Qwen/Qwen2.5-1.5B","revision":"8faed761d45a263340a0528343f099c05c9a4323"}],"name":"Commonsense completion","protocol_version":"codesota-pilot-v2-batched","runner_file":"evaluate_fast.py","source":"https://rowanzellers.com/hellaswag/","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":4.171156510012224,"measurement_version":"gpu-slot-v1"},"runner_sha256":"bc899fb160f6881408863cd8bcdf96102dd6d8b10559a5f781a8cd816103da77","scope":"035178b6b8d089715228ff8d6925d31eefde18bf58d0d0859b2d5dc1f184293b","score":0.375,"seconds":1.182585975038819,"seed":42,"selection":"first 32 valid examples in pinned split order","status":"completed","task":"commonsense","worker":"white-gpu1"}