{"id":"W3011993190","doi":"10.1109/iiswc47752.2019.9041972","title":"Deep Learning Language Modeling Workloads: Where Time Goes on Graphics Processors","year":2019,"lang":"en","type":"article","venue":"","topic":"Topic Modeling","field":"Computer Science","cited_by":6,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Toronto","funders":"","keywords":"Softmax function; Computer science; Graphics; Language model; Vocabulary; Transformer; Deep learning; Layer (electronics); Computation; Parallel computing; Artificial intelligence; Programming language; Operating system","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"metacan-v3-hybrid-931329e0061c","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.001332184,0.001314273,0.0005815338,0.000639213,0.0004755294,0.001719668,0.001296247,0.0007161153,0.00446891],"category_scores_gemma":[0.01010216,0.0005209208,0.000528884,0.001552398,0.0005518042,0.004810472,0.0007509924,0.00257955,0.002199312],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001174088,"about_ca_system_score_gemma":0.001762736,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01219979,"about_ca_topic_score_gemma":0.01653162,"domain_scores_codex":[0.9984894,0.0002333859,0.00006017753,0.0004451713,0.0004492152,0.0003225604],"domain_scores_gemma":[0.9970911,0.001435913,0.0001207093,0.0005416792,0.0005670235,0.0002435804],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"design_other","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.003440136,0.0008500597,0.04343468,0.0007626844,0.0002856882,0.001338681,0.001356908,0.2898284,0.09592701,0.01948399,0.1074536,0.4358381],"study_design_scores_gemma":[0.0001244025,0.0003615536,0.01134292,0.00007944498,0.00009970659,0.0001987454,0.0008346075,0.9103585,0.04032416,0.01240306,0.02380694,0.00006595535],"study_design_candidate":"simulation_or_modeling","study_design_consensus":null,"genre_codex":"empirical","genre_gemma":"empirical","genre_scores_codex":[0.8937341,0.002462789,0.06527808,0.005274007,0.0008316478,0.0001115271,0.00346047,0.01438026,0.01446719],"genre_scores_gemma":[0.962378,0.0008083991,0.02499752,0.0008299298,0.00009394874,0.0000890706,0.005842003,0.001423665,0.003537511],"genre_candidate":"empirical","genre_consensus":"empirical","teacher_disagreement_score":0.01219979,"threshold_uncertainty_score":0.02425754,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.009532136131513099,"score_gpt":0.2252336733998364,"score_spread":0.2157015372683233,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}