{"id":"W7009013779","doi":"","title":"On the compute and parameter efficient fine-tuning of large language models","year":2023,"lang":"en","type":"dissertation","venue":"eScholarship@McGill (McGill)","topic":"Natural Language Processing Techniques","field":"Computer Science","cited_by":0,"is_retracted":false,"has_abstract":true,"ca_institutions":"McGill University","funders":"","keywords":"Language model; Natural language; Identification (biology); Feature (linguistics); Set (abstract data type)","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":[0.001680307,0.001530171,0.0008877693,0.0005583343,0.0006405673,0.00155008,0.001815265,0.001283023,0.005076404],"category_scores_gemma":[0.00809949,0.0007935847,0.001181026,0.0008367263,0.0008726735,0.003709882,0.002041188,0.003690338,0.003058778],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.001275886,"about_ca_system_score_gemma":0.002194295,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.01066732,"about_ca_topic_score_gemma":0.01881046,"domain_scores_codex":[0.9990989,0.0003081405,0.00005395859,0.0002405447,0.0001991306,0.00009938274],"domain_scores_gemma":[0.9973909,0.00167682,0.00007390056,0.0006057906,0.0001837082,0.00006884612],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0003502036,0.0001783241,0.001530474,0.0002113759,0.00009343191,0.0001457177,0.0001418465,0.7105418,0.01070021,0.01099836,0.01638754,0.2487206],"study_design_scores_gemma":[0.00003027185,0.0000394921,0.000173569,0.00001181876,0.00001135958,0.00002239166,0.00002369516,0.9869757,0.003174883,0.007614496,0.001911095,0.00001108772],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"methods","genre_scores_codex":[0.09919274,0.002039479,0.8544101,0.001501967,0.0003311607,0.000202528,0.0009426938,0.03224731,0.009132089],"genre_scores_gemma":[0.4757625,0.001091526,0.5094323,0.001029938,0.0001330925,0.0003763952,0.003633068,0.002534646,0.006006561],"genre_candidate":"methods","genre_consensus":"methods","teacher_disagreement_score":0.01066732,"threshold_uncertainty_score":0.02121049,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.0181016859083932,"score_gpt":0.2677612508455968,"score_spread":0.2496595649372036,"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."}}