{"id":"W4400450746","doi":"10.31219/osf.io/mru2x","title":"Attention Mechanism, Transformers, BERT, and GPT: Tutorial and Survey","year":2024,"lang":"en","type":"preprint","venue":"","topic":"Explainable Artificial Intelligence (XAI)","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Transformer; Encoder; Computer science; Generative grammar; Artificial intelligence; Speech recognition; Electrical engineering; Voltage; Engineering","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":"codex-gemma-dda1882f352a","candidate_categories":["metaepi_narrow"],"consensus_categories":[],"category_scores_codex":[0.001201388,0.0002860091,0.0002938672,0.0001795183,0.0001203149,0.0009343059,0.0004822796,0.0002895146,0.00002884941],"category_scores_gemma":[0.00004826737,0.000261745,0.00007613272,0.0002146396,0.00006846351,0.0002746478,0.001389316,0.0004934743,0.00007253263],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.00005727902,"about_ca_system_score_gemma":0.0001323543,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00325912,"about_ca_topic_score_gemma":0.001933597,"domain_scores_codex":[0.9978872,0.0001405044,0.0003584444,0.0009675465,0.0003167615,0.0003295242],"domain_scores_gemma":[0.9991231,0.0001476108,0.0000620423,0.0004120342,0.0001109753,0.0001442483],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"theoretical_or_conceptual","study_design_gemma":"theoretical_or_conceptual","study_design_scores_codex":[0.00002890269,0.00005760504,0.0002747856,0.0004384246,0.0001202605,0.00005543915,0.002075647,0.00002488022,0.002333074,0.9236777,0.001580615,0.06933266],"study_design_scores_gemma":[0.0001832734,0.0002097232,0.001855927,0.0002393566,0.00006735944,0.00004178581,0.0002073114,0.1286581,0.01577258,0.8496041,0.002134509,0.001025967],"study_design_candidate":"theoretical_or_conceptual","study_design_consensus":"theoretical_or_conceptual","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1349068,0.0008289954,0.8474644,0.002021881,0.008209097,0.0007986247,0.00004205795,0.000507257,0.005220937],"genre_scores_gemma":[0.988212,0.0005142873,0.008426339,0.0001260432,0.0003316872,0.00004360786,0.00002743545,0.00002913175,0.00228949],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.8533052,"threshold_uncertainty_score":0.9999835,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.03091224222555638,"score_gpt":0.2746827783910922,"score_spread":0.2437705361655358,"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."}}