{"id":"W4385564898","doi":"10.18653/v1/2023.acl-demo.21","title":"NeuroX Library for Neuron Analysis of Deep NLP Models","year":2023,"lang":"en","type":"article","venue":"","topic":"Explainable Artificial Intelligence (XAI)","field":"Computer Science","cited_by":3,"is_retracted":false,"has_abstract":true,"ca_institutions":"Dalhousie University","funders":"","keywords":"Computer science; Python (programming language); Debiasing; Artificial intelligence; Domain (mathematical analysis); Natural language processing; Adaptation (eye); Programming language; Cognitive science","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":[0.000680507,0.001396435,0.0006387596,0.001551404,0.000574681,0.00201918,0.00190334,0.0009628403,0.07368609],"category_scores_gemma":[0.00373313,0.0007472404,0.002012878,0.0009150081,0.0004615676,0.002039883,0.001796659,0.0021936,0.01990927],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001094196,"about_ca_system_score_gemma":0.00177463,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.006637156,"about_ca_topic_score_gemma":0.01732339,"domain_scores_codex":[0.9996476,0.00004698435,0.00003515004,0.00007491141,0.0001625331,0.00003279098],"domain_scores_gemma":[0.999525,0.0002233409,0.00003483749,0.0001056101,0.00009114969,0.00002016635],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"not_applicable","study_design_gemma":"bench_or_experimental","study_design_scores_codex":[0.000259641,0.0001212663,0.003777831,0.001983306,0.0005100177,0.0008575722,0.0006226675,0.1185055,0.0122999,0.08888657,0.4683749,0.3038009],"study_design_scores_gemma":[0.00006124385,0.00004302411,0.001115512,0.0001822175,0.00005607251,0.0003144778,0.0001318554,0.7046698,0.01208053,0.10618,0.175104,0.00006128398],"study_design_candidate":"bench_or_experimental","study_design_consensus":null,"genre_codex":"methods","genre_gemma":"software","genre_scores_codex":[0.005067821,0.0004406847,0.7261102,0.0005040285,0.0002522914,0.0001879718,0.03401136,0.2193257,0.01409997],"genre_scores_gemma":[0.09253104,0.001167758,0.7503054,0.0005925545,0.0001247636,0.001399461,0.07387287,0.04460213,0.03540414],"genre_candidate":"software","genre_consensus":null,"teacher_disagreement_score":0.07368609,"threshold_uncertainty_score":0.2465045,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.05521551993420847,"score_gpt":0.280853035812814,"score_spread":0.2256375158786055,"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."}}