"""GPT-2 SAE last-token programs, layer 8 -- features 21900 to 21999 (100 programs). One function per SAE feature. Grouped 100 to a file so the set stays browsable; the programs themselves are unchanged. """ def prog_L08F21900(input_tokens): """49 token families, 50 surface forms (train F1 of the uncapped optimum: 0.728). Fires at 11,656 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: references to individuals named Michelle, particularly Michelle Obama""" # --- end external annotation --- fire = { # michelle (630 firings) ' Michelle', 'Michelle', # jesse (343 firings) ' Jesse', # huckabee (232 firings) ' Huckabee', # emma (229 firings) ' Emma', # abraham (225 firings) ' Abraham', # ivanka (214 firings) ' Ivanka', # grace (213 firings) ' Grace', # stephanie (210 firings) ' Stephanie', # debbie (199 firings) ' Debbie', # michele (168 firings) ' Michele', # melania (163 firings) ' Melania', # megan (160 firings) ' Megan', # issa (154 firings) 'issa', # regina (151 firings) ' Regina', # mickey (146 firings) ' Mickey', # rainbow (141 firings) ' Rainbow', # sandra (137 firings) ' Sandra', # rahul (136 firings) ' Rahul', # maggie (130 firings) ' Maggie', # andrew (121 firings) ' Andrews', # hank (117 firings) ' Hank', # marilyn (113 firings) ' Marilyn', # malik (110 firings) ' Malik', # sonia (109 firings) ' Sonia', # ieg (101 firings) 'ieg', # ng (100 firings) ' Ng', # gloria (99 firings) ' Gloria', # gutierrez (98 firings) ' Gutierrez', # iang (97 firings) 'iang', # denise (95 firings) ' Denise', # gina (92 firings) ' Gina', # manny (89 firings) ' Manny', # valerie (87 firings) ' Valerie', # annie (84 firings) 'annie', # heidi (82 firings) ' Heidi', # nina (82 firings) ' Nina', # gibb (78 firings) ' Gibbs', # loi (73 firings) ' Lois', # alicia (68 firings) ' Alicia', # imag (67 firings) ' Imag', # hillary (63 firings) 'hillary', # harriet (62 firings) ' Harriet', # sophia (62 firings) ' Sophia', # prairie (61 firings) ' Prairie', # grandma (60 firings) ' grandma', # simone (59 firings) ' Simone', # sasha (57 firings) ' Sasha', # category (56 firings) 'Category', # jiang (50 firings) ' Jiang', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'michelle-family', 1 def prog_L08F21901(input_tokens): """40 token families, 58 surface forms (train F1 of the uncapped optimum: 0.623). Fires at 10,853 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: references to propositions, particularly numbered ones""" # --- end external annotation --- fire = { # prop (834 firings) ' prop', ' Prop', 'Prop', 'prop', ' props', # ralph (270 firings) ' Ralph', # proposi (266 firings) ' proposition', ' Proposition', # proponent (233 firings) ' proponents', ' proponent', # decl (214 firings) ' decl', 'Decl', 'decl', ' Decl', # opp (194 firings) ' Opp', 'Opp', # janet (184 firings) ' Janet', # pant (179 firings) ' Pant', ' pant', # lgbt (160 firings) 'LGBT', # champ (153 firings) ' Champ', 'Champ', # prot (153 firings) ' Prot', ' PROT', # vertis (148 firings) 'vertising', # gavin (136 firings) ' Gavin', # marin (121 firings) ' Marin', # enc (119 firings) ' Enc', # sb (118 firings) ' SB', # californ (104 firings) ' Californ', # porn (102 firings) ' Porn', # penny (99 firings) ' Penny', # dom (99 firings) ' DOM', 'DOM', # marvin (88 firings) ' Marvin', # flag (84 firings) ' Flag', 'FLAG', # pre (81 firings) ' PRE', # postp (79 firings) ' postp', # conv (76 firings) ' Conv', 'conv', # endra (75 firings) 'endra', # proceed (72 firings) ' Proceedings', # adv (70 firings) ' Adv', ' ADV', # fp (62 firings) ' FP', # patent (61 firings) ' Patent', # pg (60 firings) ' PG', # ardon (59 firings) 'ardon', # propul (58 firings) ' propulsion', # spon (55 firings) ' Spons', # philo (52 firings) ' Philos', # alph (50 firings) 'alph', # phr (44 firings) ' phr', # controv (40 firings) ' controvers', # vend (37 firings) ' Vend', # arg (35 firings) 'Arg', 'Args', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'prop-family', 1 def prog_L08F21902(input_tokens): """12 token families, 46 surface forms (train F1 of the uncapped optimum: 0.097). Fires at 342,548 of 36,786,566 positions. Fires at 0.93% of positions, and token rules do not serve features this dense: median test F1 0.086 across the 37,173 features above 0.39% of positions, against 0.763 for the 76,283 below 0.016%. Read this fire set as the best available approximation, not as what the feature does.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: references to friendship and social connections""" # --- end external annotation --- fire = { # of (6,719 firings) ' of', # s (3,058 firings) 's', 'és', # ( (2,081 firings) ' (', # year (1,629 firings) ' years', ' year', 'years', ' YEAR', # or (1,243 firings) ' or', # man (1,046 firings) ' man', 'man', ' MAN', ' mans', 'mans', # 1 (1,029 firings) ' 1', '1', # world (999 firings) ' world', ' worlds', 'world', ' Worlds', 'worldly', ' WORLD', # group (972 firings) ' group', ' groups', ' Group', 'group', ' grouping', ' Groups', 'groups', ' grouped', # team (936 firings) ' team', ' teams', ' Team', 'team', ' Teams', # old (931 firings) ' old', ' oldest', ' older', ' Old', 'olds', ' olds', 'older', # day (867 firings) ' day', ' days', ' Day', 'days', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'of-family', 1 def prog_L08F21903(input_tokens): """2 token families, 2 surface forms (train F1 of the uncapped optimum: 0.037). Fires at 25,925 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: sentence-ending punctuation, specifically focusing on periods and commas""" # --- end external annotation --- fire = { # . (21,222 firings) '.', # ." (525 firings) '."', } token = input_tokens[-1] if token not in fire: return "no match", 0 return '.-family', 1 def prog_L08F21904(input_tokens): """6 token families, 12 surface forms (train F1 of the uncapped optimum: 0.257). Fires at 41,407 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: repeated phrases emphasizing possession or inquiries in a conversational context""" # --- end external annotation --- fire = { # have (18,360 firings) ' have', 'Have', ' Have', ' HAVE', # ve (3,546 firings) 've', "'ve", # haven (673 firings) ' haven', # hav (407 firings) 'Having', ' Having', # got (141 firings) ' Got', 'Got', # hadn (112 firings) ' hadn', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'have-family', 1 def prog_L08F21905(input_tokens): """18 token families, 57 surface forms (train F1 of the uncapped optimum: 0.170). Fires at 102,811 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: direct speech or dialogue within the text""" # --- end external annotation --- fire = { # said (3,953 firings) ' said', 'said', # say (2,864 firings) ' say', ' saying', ' Say', ' Saying', 'say', ' SAY', # ask (1,915 firings) ' asked', ' ask', ' asking', ' asks', ' Asked', # speak (1,035 firings) ' speaking', ' speak', ' speaks', ' Speaking', 'speak', ' Speak', # gave (612 firings) ' gave', # word (561 firings) ' words', 'word', # shout (501 firings) ' shouting', ' shouted', ' shout', ' shouts', # talk (496 firings) ' talking', ' talked', 'talk', 'talking', # spoke (484 firings) ' spoke', # tell (474 firings) ' telling', ' tells', 'tell', # giv (419 firings) ' giving', ' gives', 'Giving', # deliver (414 firings) ' delivered', ' delivering', # sing (349 firings) ' singing', ' sing', ' sings', 'sing', # scream (326 firings) ' screaming', ' screamed', ' scream', ' screams', # perform (319 firings) ' perform', ' performing', ' performs', # yell (308 firings) ' yelling', ' yelled', ' yell', ' yells', # offer (306 firings) ' offered', # deliv (299 firings) ' deliver', ' delivers', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'said-family', 1 def prog_L08F21906(input_tokens): """18 token families, 29 surface forms (train F1 of the uncapped optimum: 0.110). Fires at 22,452 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: phrases related to accounting and percentages""" # --- end external annotation --- fire = { # for (7,328 firings) ' for', # represent (379 firings) ' represent', ' represents', ' representing', ' represented', # explain (335 firings) ' explain', ' explains', # behind (275 firings) ' behind', # provide (271 firings) ' provide', # respons (197 firings) ' responsible', # provid (194 firings) ' provided', ' provides', 'provided', # caus (116 firings) ' caused', # dominat (112 firings) ' dominated', ' dominating', ' dominates', # compris (99 firings) ' comprises', ' comprising', # comprise (96 firings) ' comprise', # handle (92 firings) ' handle', # account (88 firings) ' accounted', ' accounting', # captur (74 firings) ' captured', ' capturing', # garner (71 firings) ' garnered', # constitute (69 firings) ' constitute', # compr (69 firings) ' comprised', # dominate (61 firings) ' dominate', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'for-family', 1 def prog_L08F21907(input_tokens): """39 token families, 60 surface forms (train F1 of the uncapped optimum: 0.621). Fires at 15,757 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: terms related to animation and animated content""" # --- end external annotation --- fire = { # anime (886 firings) ' anime', ' Anime', # animat (742 firings) ' animated', ' Animated', # anima (629 firings) ' animation', ' Animation', 'Animation', # anim (608 firings) ' Animal', ' anim', 'anim', ' Anim', 'Anim', 'animal', # gif (432 firings) ' GIF', ' gif', 'gif', # cartoon (346 firings) ' cartoon', ' cartoons', ' Cartoon', # yang (251 firings) ' Yang', 'Yang', # limb (199 firings) ' limbs', ' limb', # atar (195 firings) 'atar', 'atars', # action (189 firings) 'action', 'ACTION', # ui (185 firings) ' UI', # frozen (147 firings) ' Frozen', # miy (139 firings) ' Miy', # arnold (137 firings) ' Arnold', # embryo (134 firings) ' embryos', ' embryo', # animation (131 firings) ' animations', # avatar (125 firings) ' Avatar', ' avatar', # toy (111 firings) ' Toy', # acer (107 firings) 'acer', # amphib (106 firings) ' amphib', # pixar (92 firings) ' Pixar', # skeleton (90 firings) ' skeleton', # cinemat (89 firings) ' cinematic', # bend (85 firings) 'bender', 'bending', # apparel (80 firings) ' apparel', # anatomy (78 firings) ' anatomy', # immort (76 firings) ' immortal', # rozen (74 firings) 'rozen', # frog (73 firings) ' frogs', # airo (72 firings) 'airo', # mus (71 firings) ' Musical', # cgi (70 firings) ' CGI', # harmon (69 firings) ' Harmon', # cg (68 firings) ' CG', 'CG', # decap (53 firings) ' decap', # harmony (52 firings) ' Harmony', # hanna (51 firings) ' Hanna', # motion (51 firings) 'motion', # script (50 firings) ' scripted', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'anime-family', 1 def prog_L08F21908(input_tokens): """8 token families, 32 surface forms (train F1 of the uncapped optimum: 0.118). Fires at 290,405 of 36,786,566 positions. Fires at 0.79% of positions, and token rules do not serve features this dense: median test F1 0.086 across the 37,173 features above 0.39% of positions, against 0.763 for the 76,283 below 0.016%. Read this fire set as the best available approximation, not as what the feature does.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: key themes related to education, policy, and social issues""" # --- end external annotation --- fire = { # . (4,716 firings) '.', # american (1,073 firings) ' American', ' Americans', 'American', 'Americans', # govern (972 firings) ' government', ' Government', ' governments', ' governance', 'government', 'Government', ' Governments', # state (841 firings) ' state', ' State', ' statement', ' statements', 'state', 'State', ' Statement', ' STATE', # media (800 firings) ' media', ' Media', 'media', # us (787 firings) ' US', # google (782 firings) ' Google', 'Google', ' google', # world (769 firings) ' world', ' World', 'world', 'World', ' Worlds', } token = input_tokens[-1] if token not in fire: return "no match", 0 return '.-family', 1 def prog_L08F21909(input_tokens): """9 token families, 18 surface forms (train F1 of the uncapped optimum: 0.077). Fires at 160,106 of 36,786,566 positions. Fires at 0.44% of positions, and token rules do not serve features this dense: median test F1 0.086 across the 37,173 features above 0.39% of positions, against 0.763 for the 76,283 below 0.016%. Read this fire set as the best available approximation, not as what the feature does.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: expressions related to personal growth and resilience""" # --- end external annotation --- fire = { # you (3,561 firings) ' you', # then (1,646 firings) ' then', ' Then', 'Then', 'then', ' THEN', # ? (1,478 firings) '?', # get (1,443 firings) ' get', ' gets', # your (1,189 firings) ' your', # find (1,053 firings) ' find', ' finding', ' Finding', 'finding', # without (728 firings) ' without', # start (502 firings) ' start', ' starts', # thing (448 firings) ' things', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'you-family', 1 def prog_L08F21910(input_tokens): """1 token families, 2 surface forms (train F1 of the uncapped optimum: 0.298). Fires at 14,043 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: the phrase "do what" in various contexts""" # --- end external annotation --- fire = { # what (7,443 firings) ' what', 'what', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'what-family', 1 def prog_L08F21911(input_tokens): """9 token families, 14 surface forms (train F1 of the uncapped optimum: 0.136). Fires at 206,983 of 36,786,566 positions. Fires at 0.56% of positions, and token rules do not serve features this dense: median test F1 0.086 across the 37,173 features above 0.39% of positions, against 0.763 for the 76,283 below 0.016%. Read this fire set as the best available approximation, not as what the feature does.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: words related to security and safety mechanisms""" # --- end external annotation --- fire = { # - (4,429 firings) '-', # ) (1,729 firings) ')', # " (1,305 firings) '"', # way (721 firings) ' way', ' ways', # informa (711 firings) ' information', 'information', # acces (663 firings) ' access', 'access', # experi (621 firings) ' experience', # level (614 firings) ' level', ' levels', 'level', # new (540 firings) ' news', } token = input_tokens[-1] if token not in fire: return "no match", 0 return '--family', 1 def prog_L08F21912(input_tokens): """3 token families, 9 surface forms (train F1 of the uncapped optimum: 0.774). Fires at 28,157 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: phrases that emphasize specific actions or outcomes""" # --- end external annotation --- fire = { # these (18,053 firings) ' these', ' These', 'These', 'these', ' THESE', # those (1,589 firings) ' Those', 'Those', # such (481 firings) ' Such', 'Such', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'these-family', 1 def prog_L08F21913(input_tokens): """8 token families, 20 surface forms (train F1 of the uncapped optimum: 0.205). Fires at 117,414 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: phrases that specify examples or categories of items or entities""" # --- end external annotation --- fire = { # includ (7,058 firings) ' including', ' includes', ' included', 'including', ' Including', 'includes', # like (6,197 firings) ' like', 'like', # such (2,745 firings) ' such', 'such', # call (2,066 firings) ' called', ' Called', # include (1,177 firings) ' include', ' Include', # especial (464 firings) ' especially', 'especially', ' Especially', # featur (363 firings) ' featuring', # particular (316 firings) ' particularly', 'particularly', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'includ-family', 1 def prog_L08F21914(input_tokens): """47 token families, 55 surface forms (train F1 of the uncapped optimum: 0.628). Fires at 15,512 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: references to the stomach and gastrointestinal health issues""" # --- end external annotation --- fire = { # gut (430 firings) ' gut', ' guts', # stomach (385 firings) ' stomach', # throat (282 firings) ' throat', ' throats', # taco (251 firings) ' Taco', ' taco', ' tacos', # elly (213 firings) 'elly', # ocean (191 firings) ' oceans', # appetite (181 firings) ' appetite', # payroll (164 firings) ' payroll', # bel (146 firings) ' belly', # gast (140 firings) ' Gast', ' gast', # respiratory (139 firings) ' respiratory', # waist (136 firings) ' waist', # hamstr (131 firings) ' hamstring', # cardiac (116 firings) ' cardiac', # utton (116 firings) 'utton', # breath (116 firings) ' breathing', # panc (113 firings) ' panc', ' Panc', # vomit (109 firings) ' vomiting', ' vomit', # dig (102 firings) ' digest', # vagina (97 firings) ' vagina', # consci (94 firings) ' conscience', # seaw (92 firings) ' seaw', # metabolism (88 firings) ' metabolism', # atmospher (84 firings) ' atmospheric', # bowel (81 firings) ' bowel', # abdomen (80 firings) ' abdomen', # nausea (79 firings) ' nausea', # carniv (79 firings) ' carniv', # groin (78 firings) ' groin', # nicaragua (72 firings) ' Nicaragua', # backstage (70 firings) ' backstage', # agu (70 firings) 'agus', # varg (67 firings) ' Varg', # investig (67 firings) ' Investig', # appet (67 firings) ' appet', # torso (63 firings) ' torso', # abdomin (63 firings) ' abdominal', # eyeb (62 firings) ' eyeb', # intestin (60 firings) ' intestinal', 'intestinal', # gi (59 firings) ' GI', # digest (57 firings) ' digestive', # sylvia (56 firings) ' Sylvia', # bloody (54 firings) ' Bloody', # aryn (54 firings) 'aryn', # pancreat (52 firings) ' pancreat', # bladd (52 firings) ' bladder', # microbiome (51 firings) ' microbiome', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'gut-family', 1 def prog_L08F21915(input_tokens): """27 token families, 78 surface forms (train F1 of the uncapped optimum: 0.221). Fires at 148,919 of 36,786,566 positions. Fires at 0.40% of positions, and token rules do not serve features this dense: median test F1 0.086 across the 37,173 features above 0.39% of positions, against 0.763 for the 76,283 below 0.016%. Read this fire set as the best available approximation, not as what the feature does.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: terms related to various fields of study or expertise""" # --- end external annotation --- fire = { # soc (2,307 firings) ' social', ' Social', ' soc', 'social', # polit (1,707 firings) ' political', ' Political', 'Political', ' polit', 'political', ' POLIT', # publ (1,575 firings) ' public', ' Public', # busi (1,291 firings) ' business', ' Business', 'Business', 'business', # secur (991 firings) ' security', 'security', ' securities', # health (979 firings) ' health', 'health', # digit (937 firings) ' digital', ' Digital', 'Digital', ' digit', 'digital', # law (924 firings) ' law', # foreign (908 firings) ' foreign', 'foreign', # human (903 firings) ' human', ' Human', ' humanities', # sport (866 firings) ' sports', ' sport', ' sporting', 'Sports', 'sports', # econom (864 firings) ' economic', ' Economic', 'econom', 'Econom', 'Economic', # med (769 firings) ' medical', ' Medical', # comput (762 firings) ' computer', ' Computer', 'Computer', ' computational', ' computed', # financ (741 firings) ' financial', ' Financial', 'financial', # sci (698 firings) ' science', ' sci', 'Science', 'science', # mus (658 firings) ' music', ' musical', ' mus', # tech (637 firings) ' tech', ' Tech', 'Tech', # climate (606 firings) ' climate', ' Climate', 'climate', # web (570 firings) ' web', # anti (564 firings) ' anti', # art (561 firings) ' art', ' arts', # consum (481 firings) ' consumer', ' Consumer', 'Consumer', # civil (463 firings) ' civil', # environment (445 firings) ' environmental', # cyb (391 firings) ' cyber', ' Cyber', # invest (383 firings) ' investment', ' Investment', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'soc-family', 1 def prog_L08F21916(input_tokens): """28 token families, 32 surface forms (train F1 of the uncapped optimum: 0.418). Fires at 11,278 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: references to the rear aspect of objects, particularly vehicles""" # --- end external annotation --- fire = { # front (1,986 firings) ' front', 'Front', ' frontal', # rear (570 firings) ' rear', ' Rear', # exterior (154 firings) ' exterior', # reverse (151 firings) ' reverse', ' Reverse', # tail (116 firings) ' tail', # hind (88 firings) ' hind', # resur (88 firings) ' resur', # exhaust (82 firings) ' exhaust', # steer (73 firings) ' steering', # ret (73 firings) ' Ret', # shel (71 firings) ' shel', # onboard (71 firings) ' onboard', # overhead (69 firings) ' overhead', # art (66 firings) 'arter', # quart (62 firings) 'quarter', # brake (56 firings) ' brake', # outward (55 firings) ' outward', # detach (50 firings) ' detached', # stereo (50 firings) ' stereo', # anterior (47 firings) ' anterior', # fac (46 firings) 'facing', # stern (45 firings) ' stern', # cockpit (45 firings) ' cockpit', # terior (34 firings) 'terior', # bare (32 firings) ' Bare', # protr (30 firings) ' protr', # later (30 firings) ' lateral', # downstream (29 firings) ' downstream', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'front-family', 1 def prog_L08F21917(input_tokens): """35 token families, 60 surface forms (train F1 of the uncapped optimum: 0.451). Fires at 40,878 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: words and phrases indicating significant increases in value or intensity""" # --- end external annotation --- fire = { # ris (1,345 firings) ' rising', ' rises', ' Rising', 'rising', # rise (737 firings) ' rise', # surge (567 firings) ' surge', ' Surge', # soar (478 firings) ' soaring', ' soared', ' soar', # boom (401 firings) ' booming', ' boom', # fly (387 firings) ' flying', 'flying', # explod (357 firings) ' exploded', ' exploding', ' explodes', # surg (351 firings) ' surged', ' surging', ' surges', # plung (336 firings) ' plunged', ' plung', # jump (330 firings) ' jumped', ' jumps', ' jumper', # risen (329 firings) ' risen', # rose (324 firings) ' rose', # climb (323 firings) ' climbed', ' climbing', ' climb', # hyp (304 firings) ' hyper', 'hyper', # escalat (290 firings) ' escalating', ' escalated', # fall (288 firings) ' falling', ' Falling', # dramatical (275 firings) ' dramatically', # spir (258 firings) ' spir', ' spiral', # stagger (247 firings) ' staggering', # upward (224 firings) ' upward', ' upwards', # plummet (223 firings) ' plummeted', ' plummet', # umbl (214 firings) 'umbled', 'umbling', 'umbles', # spike (212 firings) ' spike', # sharp (208 firings) ' sharply', # whopp (193 firings) ' whopping', # rampant (185 firings) ' rampant', # plunge (153 firings) ' plunge', # erupt (149 firings) ' erupted', # doubl (149 firings) ' doubled', # slim (143 firings) ' slim', # epidem (140 firings) ' epidemic', # janet (137 firings) ' Janet', # trajectory (132 firings) ' trajectory', # elevat (127 firings) ' elevated', # bul (103 firings) ' bul', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'ris-family', 1 def prog_L08F21918(input_tokens): """24 token families, 42 surface forms (train F1 of the uncapped optimum: 0.368). Fires at 18,232 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: references to images or photographs in context""" # --- end external annotation --- fire = { # seen (1,448 firings) ' seen', 'seen', ' Seen', # photo (1,265 firings) ' photo', ' PHOTO', 'PHOTOS', 'photo', # shown (726 firings) ' shown', 'shown', # pictur (726 firings) ' pictured', 'pictured', # picture (418 firings) 'Picture', ' Picture', # photograph (339 firings) ' photographed', ' photograph', ' Photograph', # file (326 firings) 'FILE', ' FILE', # depict (239 firings) ' depicted', ' depicts', ' depicting', # see (237 firings) 'see', # left (205 firings) 'left', # pict (182 firings) 'Pict', ' Pict', ' pict', 'pict', # courtesy (172 firings) ' Courtesy', 'Courtesy', # film (154 firings) ' filmed', # ty (151 firings) 'TY', # surround (136 firings) ' surrounded', # display (125 firings) ' displayed', # above (118 firings) 'above', ' Above', # video (96 firings) 'VIDEO', # onstage (83 firings) ' onstage', # flank (72 firings) ' flanked', # acquitt (70 firings) ' acquitted', # play (52 firings) 'played', # below (50 firings) 'below', # circa (49 firings) ' circa', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'seen-family', 1 def prog_L08F21919(input_tokens): """45 token families, 49 surface forms (train F1 of the uncapped optimum: 0.606). Fires at 15,662 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: references to parsing or the term "ars." """ # --- end external annotation --- fire = { # ars (1,735 firings) 'ars', 'ARS', # ar (974 firings) 'AR', # ared (393 firings) 'ared', # are (377 firings) 'ares', # aring (345 firings) 'aring', # arp (332 firings) 'arp', # mountain (320 firings) ' mountains', # car (165 firings) ' Cars', # lar (157 firings) ' Lars', # oust (155 firings) 'oust', # las (145 firings) 'las', # rocket (139 firings) ' rockets', # arin (131 firings) 'arin', # siriu (130 firings) ' Sirius', # ras (117 firings) ' Ras', # arlington (115 firings) ' Arlington', # ari (113 firings) 'aris', # marco (103 firings) ' Marcos', # stag (101 firings) ' staging', # parson (100 firings) ' Parsons', # par (92 firings) ' pars', ' Pars', # sir (91 firings) 'Sir', # katy (90 firings) ' Katy', # pars (87 firings) ' parsing', ' parser', ' parsed', # arson (84 firings) 'arson', # zar (77 firings) ' Zar', # arro (72 firings) 'arro', # rehears (69 firings) ' rehearsal', # ariu (69 firings) 'arius', # arer (66 firings) 'arer', # pasadena (64 firings) ' Pasadena', # irs (63 firings) 'irs', # kardashian (63 firings) ' Kardashian', # myra (60 firings) 'myra', # rar (56 firings) 'rar', # iffin (55 firings) 'iffin', # rehear (53 firings) ' rehears', # asar (50 firings) 'asar', # yru (49 firings) 'yrus', # aryn (44 firings) 'aryn', # ora (43 firings) 'oras', # lumpur (42 firings) ' Lumpur', # aram (41 firings) ' Aram', # arse (40 firings) 'arse', # flar (40 firings) ' flares', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'ars-family', 1 def prog_L08F21920(input_tokens): """28 token families, 45 surface forms (train F1 of the uncapped optimum: 0.354). Fires at 83,645 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: names of people or proper nouns""" # --- end external annotation --- fire = { # ie (1,387 firings) 'ie', # ryan (1,149 firings) ' Ryan', 'Ryan', 'ryan', # on (822 firings) 'on', # ley (691 firings) 'ley', 'LEY', # scott (584 firings) ' Scott', 'Scott', # martin (527 firings) ' Martin', 'Martin', # lee (494 firings) ' Lee', 'Lee', 'lee', # jordan (491 firings) ' Jordan', 'Jordan', # ton (472 firings) 'ton', # ne (447 firings) 'ne', # russell (369 firings) ' Russell', 'Russell', # wayne (353 firings) ' Wayne', 'wayne', # marcu (337 firings) ' Marcus', 'Marcus', # ane (335 firings) 'ane', # colin (334 firings) ' Colin', # denni (309 firings) ' Dennis', # warren (286 firings) ' Warren', 'Warren', # anthony (266 firings) ' Anthony', # gordon (261 firings) ' Gordon', 'Gordon', # ros (244 firings) ' Ross', 'Ross', # tyl (238 firings) ' Tyler', 'Tyler', # grant (235 firings) ' Grant', 'Grant', # howard (223 firings) ' Howard', 'Howard', # bryan (217 firings) ' Bryan', # rick (216 firings) 'rick', # lindsey (215 firings) ' Lindsey', # corey (213 firings) ' Corey', # julian (213 firings) ' Julian', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'ie-family', 1 def prog_L08F21921(input_tokens): """1 token families, 2 surface forms (train F1 of the uncapped optimum: 0.203). Fires at 82,975 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: repeated conjunctions and their use in connecting various clauses or ideas""" # --- end external annotation --- fire = { # and (67,719 firings) ' and', ' AND', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'and-family', 1 def prog_L08F21922(input_tokens): """21 token families, 35 surface forms (train F1 of the uncapped optimum: 0.234). Fires at 46,092 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: phrases related to economic inequality and taxation policies""" # --- end external annotation --- fire = { # top (1,725 firings) ' top', # rich (791 firings) ' rich', ' richest', ' richer', # poor (704 firings) ' poor', ' poorest', ' Poor', ' poorer', # middle (491 firings) ' middle', 'middle', # inequal (477 firings) ' inequality', ' inequalities', # sup (450 firings) ' super', # clas (437 firings) ' class', # income (400 firings) ' income', 'income', # corporation (359 firings) ' corporations', # wealthy (321 firings) ' wealthy', # wealth (321 firings) ' wealth', ' Wealth', 'wealth', # tax (291 firings) ' taxes', # household (242 firings) ' households', # 500 (220 firings) ' 500', # sector (216 firings) ' sector', ' sectors', # bottom (206 firings) ' bottom', # wealthi (164 firings) ' wealthiest', ' wealthier', # earn (157 firings) ' earners', ' earning', 'Earn', # elite (157 firings) ' elite', # billionair (155 firings) ' billionaires', # 99 (139 firings) ' 99', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'top-family', 1 def prog_L08F21923(input_tokens): """2 token families, 6 surface forms (train F1 of the uncapped optimum: 0.214). Fires at 72,964 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: statements about the importance and value of various concepts or subjects""" # --- end external annotation --- fire = { # is (35,308 firings) ' is', ' Is', ' IS', # are (14,114 firings) ' are', ' Are', ' ARE', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'is-family', 1 def prog_L08F21924(input_tokens): """1 token families, 3 surface forms (train F1 of the uncapped optimum: 0.113). Fires at 95,829 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: the definite article "the" """ # --- end external annotation --- fire = { # the (82,537 firings) ' the', 'the', ' THE', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'the-family', 1 def prog_L08F21925(input_tokens): """12 token families, 48 surface forms (train F1 of the uncapped optimum: 0.133). Fires at 246,475 of 36,786,566 positions. Fires at 0.67% of positions, and token rules do not serve features this dense: median test F1 0.086 across the 37,173 features above 0.39% of positions, against 0.763 for the 76,283 below 0.016%. Read this fire set as the best available approximation, not as what the feature does.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: references to various types of biases and social issues""" # --- end external annotation --- fire = { # gam (1,105 firings) ' games', ' Games', ' gamers', ' Gaming', 'games', ' gamer', 'gaming', 'Gaming', # system (1,062 firings) ' system', ' systems', ' System', ' Systems', # ver (1,034 firings) ' version', ' Version', 'version', ' VERS', 'VERSION', # game (919 firings) ' game', ' GAME', # ism (733 firings) 'ism', 'isms', 'ISM', # school (691 firings) ' School', ' schools', ' Schools', # market (678 firings) ' market', ' markets', ' marketing', ' Market', ' Marketing', ' Markets', ' marketers', # data (677 firings) ' data', ' Data', ' DATA', # ist (671 firings) 'ists', 'ist', # act (645 firings) ' Act', ' Acts', 'acting', # coach (630 firings) ' coach', ' coaches', ' Coach', # minist (620 firings) ' Minister', ' minister', ' ministers', ' Ministers', 'ministic', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'gam-family', 1 def prog_L08F21926(input_tokens): """40 token families, 58 surface forms (train F1 of the uncapped optimum: 0.565). Fires at 29,133 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: references to the word "smash" and its variations""" # --- end external annotation --- fire = { # upset (774 firings) ' upset', # smash (745 firings) ' smashed', ' Smash', ' smash', ' smashing', # ash (694 firings) 'ashing', 'ashed', 'ashes', # crush (614 firings) ' crushed', ' crushing', ' crush', ' Crush', # clash (571 firings) ' clashes', ' clashed', ' clash', # flagship (555 firings) ' flagship', # overturn (449 firings) ' overturn', ' overturned', # thrill (296 firings) ' thrilled', # pleas (275 firings) ' pleased', # shatter (258 firings) ' shattered', ' shattering', ' Shattered', # furi (201 firings) ' furious', # devastat (184 firings) ' devastated', # scatter (163 firings) ' scattered', ' scattering', # vand (155 firings) ' vandal', # trump (154 firings) ' trump', 'trump', # crash (152 firings) ' crashing', # havoc (151 firings) ' havoc', # unse (140 firings) ' unse', # rebel (140 firings) ' Rebel', ' Rebels', # delight (131 firings) ' delighted', # bru (129 firings) ' bruised', ' bru', # sens (128 firings) ' sensational', # shr (123 firings) ' shred', # vandalism (120 firings) ' vandalism', # violent (119 firings) ' violently', # bruis (118 firings) ' bruising', ' bruises', # tram (110 firings) ' tram', # hur (109 firings) ' hur', # craz (109 firings) ' craz', # sunset (105 firings) ' Sunset', # fierce (99 firings) ' fiercely', # fractur (98 firings) ' fractured', # shaken (95 firings) ' shaken', # smear (93 firings) ' smear', # topp (92 firings) ' topp', # stomp (90 firings) ' stomp', # blur (87 firings) ' blur', # squash (77 firings) ' squash', # bud (77 firings) ' bud', # tabloid (76 firings) ' tabloid', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'upset-family', 1 def prog_L08F21927(input_tokens): """15 token families, 17 surface forms (train F1 of the uncapped optimum: 0.093). Fires at 109,234 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: references to notable events or developments related to technology and culture""" # --- end external annotation --- fire = { # the (3,567 firings) 'The', # a (1,732 firings) 'A', # in (1,459 firings) 'In', # thi (1,276 firings) 'This', 'THIS', # as (632 firings) 'As', # when (446 firings) 'When', # aft (421 firings) 'After', # on (390 firings) 'On', # by (380 firings) 'By', # if (351 firings) 'If', # at (347 firings) 'At', # one (343 firings) 'One', # an (302 firings) 'An', # new (300 firings) 'New', 'NEW', # with (277 firings) 'With', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'the-family', 1 def prog_L08F21928(input_tokens): """3 token families, 8 surface forms (train F1 of the uncapped optimum: 0.640). Fires at 6,566 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: mentions of the name "Car" and its variations throughout the document""" # --- end external annotation --- fire = { # car (3,088 firings) ' Car', 'Car', 'car', 'CAR', ' CAR', ' Cars', # tor (395 firings) ' Tor', # kel (78 firings) ' Kel', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'car-family', 1 def prog_L08F21929(input_tokens): """10 token families, 20 surface forms (train F1 of the uncapped optimum: 0.492). Fires at 12,012 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: the repetition of the word "Then." """ # --- end external annotation --- fire = { # then (6,502 firings) ' then', ' Then', 'Then', 'then', ' THEN', # lat (358 firings) ' Later', 'Later', # plu (301 firings) ' Plus', 'Plus', # next (181 firings) 'Next', # final (101 firings) ' Finally', # otherwise (96 firings) ' Otherwise', 'Otherwise', # untitl (68 firings) 'Untitled', # eventual (60 firings) ' Eventually', 'Eventually', # else (45 firings) ' Else', 'Else', 'else', # thu (32 firings) 'Thus', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'then-family', 1 def prog_L08F21930(input_tokens): """29 token families, 34 surface forms (train F1 of the uncapped optimum: 0.142). Fires at 64,260 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: words related to political discussions or topics""" # --- end external annotation --- fire = { # o (876 firings) 'o', # at (718 firings) 'at', # k (707 firings) 'k', # en (610 firings) 'en', # on (610 firings) 'on', # ar (550 firings) 'ar', 'AR', # z (474 firings) 'z', # um (408 firings) 'um', 'UM', # ter (372 firings) 'ter', # it (360 firings) 'it', # v (336 firings) 'v', # ch (331 firings) 'ch', # ac (307 firings) 'ac', # ot (299 firings) 'ot', # ol (278 firings) 'ol', # ay (272 firings) 'ay', # disney (256 firings) ' Disney', # te (251 firings) 'te', # ow (240 firings) 'OW', # is (240 firings) 'IS', # ov (233 firings) 'ov', # att (211 firings) 'att', 'atter', # ize (208 firings) 'ize', # ap (183 firings) 'ap', # ost (180 firings) 'ost', 'oster', # la (178 firings) 'la', 'LA', # ber (168 firings) 'ber', # research (168 firings) ' Research', # qu (163 firings) 'qu', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'o-family', 1 def prog_L08F21931(input_tokens): """43 token families, 51 surface forms (train F1 of the uncapped optimum: 0.566). Fires at 12,089 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: concepts related to transparency and accountability""" # --- end external annotation --- fire = { # transparency (487 firings) ' transparency', ' Transparency', # whistleblow (378 firings) ' whistleblower', ' whistleblowers', # transparent (314 firings) ' transparent', # disclosure (299 firings) ' disclosure', ' Disclosure', # insid (279 firings) ' insider', ' insiders', # account (260 firings) ' accountability', ' Accountability', # privat (258 firings) ' privat', ' privatization', # secret (211 firings) ' secretive', # watchdog (210 firings) ' watchdog', # secrecy (200 firings) ' secrecy', # procedur (139 firings) ' procedural', # sustain (130 firings) ' sustainability', # hypocrisy (115 firings) ' hypocrisy', # vis (115 firings) ' visibility', # honesty (106 firings) ' honesty', # fair (103 firings) ' fairness', # straightforward (100 firings) ' straightforward', # aud (95 firings) ' aud', # corrup (91 firings) 'corruption', ' Corruption', # open (89 firings) ' openness', # disclosur (79 firings) ' disclosures', # encap (73 firings) ' encaps', # foia (72 firings) ' FOIA', # esty (71 firings) 'esty', # outsid (67 firings) ' outsider', # murky (67 firings) ' murky', # parent (58 firings) 'parent', # opaque (55 firings) ' opaque', # shame (54 firings) ' shameful', # simplic (54 firings) ' simplicity', # sunshine (53 firings) ' sunshine', # democrat (52 firings) ' democrat', # http (52 firings) ' HTTPS', # shroud (52 firings) ' shrouded', # rina (50 firings) 'rina', # obfusc (48 firings) ' obfusc', # confidential (41 firings) ' confidentiality', # hon (38 firings) ' Honest', 'Honest', # translucent (36 firings) ' translucent', # recy (36 firings) 'recy', # modesty (36 firings) ' modesty', # spec (33 firings) 'spection', # humil (32 firings) ' humility', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'transparency-family', 1 def prog_L08F21932(input_tokens): """28 token families, 36 surface forms (train F1 of the uncapped optimum: 0.730). Fires at 17,136 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: references to the abbreviation "AL" """ # --- end external annotation --- fire = { # al (9,214 firings) ' Al', ' al', 'AL', 'Al', ' AL', # alley (263 firings) ' alley', ' Alley', # ill (252 firings) 'Ill', # albany (209 firings) ' Albany', # alonso (173 firings) ' Alonso', # alison (138 firings) ' Alison', # salem (137 firings) ' Salem', # algae (120 firings) ' algae', # alph (114 firings) ' Alph', 'alph', # jal (111 firings) ' Jal', # almond (109 firings) ' almond', ' almonds', # alto (105 firings) ' Alto', # alb (104 firings) ' Alb', # aj (100 firings) ' AJ', # ale (100 firings) ' ale', ' ALE', # alberto (98 firings) ' Alberto', # fal (88 firings) ' fal', # cal (66 firings) ' CAL', # alf (65 firings) ' Alf', # alc (64 firings) ' Alc', # ral (60 firings) 'RAL', # abdomin (59 firings) ' abdominal', # ر (55 firings) 'ر', # alice (51 firings) 'Alice', # anal (50 firings) ' anal', # ulsion (46 firings) 'ulsion', # alpine (43 firings) ' Alpine', # exasp (43 firings) ' exasper', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'al-family', 1 def prog_L08F21933(input_tokens): """21 token families, 46 surface forms (train F1 of the uncapped optimum: 0.172). Fires at 157,327 of 36,786,566 positions. Fires at 0.43% of positions, and token rules do not serve features this dense: median test F1 0.086 across the 37,173 features above 0.39% of positions, against 0.763 for the 76,283 below 0.016%. Read this fire set as the best available approximation, not as what the feature does.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: references to specific industries and sectors related to job opportunities""" # --- end external annotation --- fire = { # health (1,073 firings) ' health', 'health', # soc (968 firings) ' social', ' soc', # publ (966 firings) ' public', # sci (821 firings) ' science', ' sci', 'sci', # email (665 firings) ' email', # financ (616 firings) ' financial', 'financial', 'Financial', # mus (578 firings) ' music', 'music', # educa (564 firings) ' education', 'Education', # med (563 firings) ' medical', ' MED', 'medical', # busi (559 firings) ' business', # develop (540 firings) ' development', 'development', 'Development', 'develop', # sport (534 firings) ' sports', ' sporting', 'sports', # food (533 firings) ' food', 'food', # care (506 firings) ' care', 'care', # tv (475 firings) ' TV', ' tv', # energy (473 firings) ' energy', 'energy', # web (465 firings) ' web', 'web', # servic (464 firings) ' services', 'services', # art (455 firings) ' art', ' arts', ' arter', # oil (412 firings) ' oil', ' oils', # person (399 firings) ' personal', 'personal', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'health-family', 1 def prog_L08F21934(input_tokens): """8 token families, 14 surface forms (train F1 of the uncapped optimum: 0.094). Fires at 234,787 of 36,786,566 positions. Fires at 0.64% of positions, and token rules do not serve features this dense: median test F1 0.086 across the 37,173 features above 0.39% of positions, against 0.763 for the 76,283 below 0.016%. Read this fire set as the best available approximation, not as what the feature does.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: themes related to enjoyment and appreciation of nature or gardens""" # --- end external annotation --- fire = { # " (2,207 firings) ' "', # not (1,714 firings) ' not', 'Notes', # which (1,399 firings) ' which', # there (870 firings) ' there', 'there', ' THERE', # ; (861 firings) ';', ' ;', # no (808 firings) ' no', # day (671 firings) ' day', 'DAY', ' DAY', # made (636 firings) ' made', } token = input_tokens[-1] if token not in fire: return "no match", 0 return '"-family', 1 def prog_L08F21935(input_tokens): """13 token families, 15 surface forms (train F1 of the uncapped optimum: 0.120). Fires at 16,401 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: references to the New York Times""" # --- end external annotation --- fire = { # twitt (433 firings) ' Twitter', # cool (204 firings) ' cool', # lgbt (174 firings) ' LGBT', # dai (152 firings) ' Daily', # organization (147 firings) ' organizations', # cap (131 firings) ' Caption', # reddit (126 firings) ' Reddit', 'Reddit', # fc (115 firings) ' FC', # transcript (111 firings) ' Transcript', # gif (87 firings) ' GIF', # new (86 firings) 'news', ' NEW', # full (79 firings) ' Full', # pr (69 firings) ' PR', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'twitt-family', 1 def prog_L08F21936(input_tokens): """15 token families, 71 surface forms (train F1 of the uncapped optimum: 0.241). Fires at 452,297 of 36,786,566 positions. Fires at 1.23% of positions, and token rules do not serve features this dense: median test F1 0.086 across the 37,173 features above 0.39% of positions, against 0.763 for the 76,283 below 0.016%. Read this fire set as the best available approximation, not as what the feature does.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: references to layoffs and job cuts in the workforce""" # --- end external annotation --- fire = { # work (10,930 firings) ' work', ' workers', ' working', ' worked', ' worker', ' works', ' Workers', 'workers', ' Work', 'work', 'working', 'Work', 'worked', ' Worker', 'worker', ' Working', 'Working', ' WORK', ' workings', # for (9,440 firings) ' for', # job (6,313 firings) ' job', ' jobs', ' Job', ' Jobs', 'job', 'Job', 'jobs', # been (6,146 firings) ' been', # new (4,648 firings) ' new', # day (3,136 firings) ' day', ' days', 'day', 'days', # form (2,801 firings) ' former', # employe (2,449 firings) ' employees', ' Employees', # hour (2,252 firings) ' hours', ' hour', 'hour', ' hourly', 'Hours', 'hours', ' Hours', ' Hour', 'Hour', # 000 (2,239 firings) '000', # staff (2,220 firings) ' staff', ' staffers', ' staffer', ' staffing', 'staff', ' staffed', 'Staff', # women (1,829 firings) ' women', ' Women', 'Women', 'women', # contract (1,668 firings) ' contract', ' contracts', 'contract', ' contracted', ' Contract', ' contracting', 'Contract', # fir (1,442 firings) ' fired', ' firing', ' Fired', ' fir', # month (1,225 firings) ' months', 'month', 'months', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'work-family', 1 def prog_L08F21937(input_tokens): """17 token families, 43 surface forms (train F1 of the uncapped optimum: 0.146). Fires at 245,295 of 36,786,566 positions. Fires at 0.67% of positions, and token rules do not serve features this dense: median test F1 0.086 across the 37,173 features above 0.39% of positions, against 0.763 for the 76,283 below 0.016%. Read this fire set as the best available approximation, not as what the feature does.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: terms related to immigration and asylum issues""" # --- end external annotation --- fire = { # are (3,986 firings) ' are', # people (3,487 firings) ' people', # who (3,249 firings) ' who', # were (2,265 firings) ' were', # them (1,627 firings) ' them', ' Them', # refuge (1,330 firings) ' refugees', ' refuge', ' Refugees', # immigrant (1,083 firings) ' immigrants', ' immigrant', 'immigrant', # migrant (924 firings) ' migrants', ' migrant', # return (821 firings) ' return', ' returned', ' returning', # arriv (766 firings) ' arrived', ' arrival', ' arriving', # illeg (748 firings) ' illegal', ' illeg', ' Illegal', 'illegal', # citizen (731 firings) ' citizens', ' citizen', 'citizens', # children (715 firings) ' children', 'Children', # liv (685 firings) ' living', 'living', # work (666 firings) ' workers', ' worker', ' Work', ' Worker', 'worker', # asylum (622 firings) ' asylum', ' Asylum', # seek (617 firings) ' seekers', ' seeking', ' seeker', 'seekers', 'seeking', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'are-family', 1 def prog_L08F21938(input_tokens): """8 token families, 9 surface forms (train F1 of the uncapped optimum: 0.067). Fires at 318,775 of 36,786,566 positions. Fires at 0.87% of positions, and token rules do not serve features this dense: median test F1 0.086 across the 37,173 features above 0.39% of positions, against 0.763 for the 76,283 below 0.016%. Read this fire set as the best available approximation, not as what the feature does.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: indicators of urgency or critical conditions""" # --- end external annotation --- fire = { # a (14,623 firings) ' a', 'ça', # with (3,784 firings) ' with', # as (2,946 firings) ' as', # more (1,328 firings) ' more', # up (1,167 firings) ' up', # than (1,029 firings) ' than', # like (846 firings) ' like', # no (801 firings) ' no', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'a-family', 1 def prog_L08F21939(input_tokens): """13 token families, 16 surface forms (train F1 of the uncapped optimum: 0.120). Fires at 75,787 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: phrases indicating sufficiency or adequacy in various contexts""" # --- end external annotation --- fire = { # make (2,758 firings) ' make', ' MAKE', # get (1,368 firings) ' get', # keep (719 firings) ' keep', # win (595 firings) ' win', # give (478 firings) ' give', # stop (451 firings) ' stop', # hold (285 firings) ' hold', # allow (252 firings) ' allow', # ensure (244 firings) ' ensure', # prevent (218 firings) ' prevent', ' prevented', # cause (210 firings) ' cause', # earn (203 firings) ' earn', ' earning', # buy (193 firings) ' buy', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'make-family', 1 def prog_L08F21940(input_tokens): """2 token families, 3 surface forms (train F1 of the uncapped optimum: 0.269). Fires at 17,806 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: instances of the phrase "pick up." """ # --- end external annotation --- fire = { # up (8,853 firings) ' up', # pickup (273 firings) ' pickup', ' pickups', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'up-family', 1 def prog_L08F21941(input_tokens): """5 token families, 21 surface forms (train F1 of the uncapped optimum: 0.741). Fires at 21,110 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: instances of the word "hold" and its variations in various contexts""" # --- end external annotation --- fire = { # hold (8,191 firings) ' hold', ' holding', ' holds', 'hold', ' holders', ' holder', ' Holder', ' Holdings', ' Hold', 'Hold', ' holdings', 'holders', ' Holding', 'holding', # held (4,866 firings) ' held', 'held', ' Held', # holden (111 firings) ' Holden', # own (90 firings) ' owning', # cling (78 firings) ' cling', ' clinging', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'hold-family', 1 def prog_L08F21942(input_tokens): """40 token families, 55 surface forms (train F1 of the uncapped optimum: 0.421). Fires at 93,968 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: references to government and its activities""" # --- end external annotation --- fire = { # court (2,834 firings) ' Court', 'Court', ' Courts', ' COURT', # govern (2,511 firings) ' Government', 'Government', ' Governments', # state (2,067 firings) ' State', # council (1,511 firings) ' Council', 'Council', # feder (1,328 firings) ' Federal', # commis (1,299 firings) ' Commission', 'Commission', # senate (1,280 firings) ' Senate', # party (1,257 firings) ' Party', # army (1,232 firings) ' Army', 'Army', # parlia (1,154 firings) ' Parliament', # police (1,146 firings) ' Police', # board (1,138 firings) ' Board', ' Boards', # administra (919 firings) ' Administration', # ministry (917 firings) ' Ministry', # church (898 firings) ' Church', ' Churches', # committee (811 firings) ' Committee', # coast (801 firings) ' Coast', # office (714 firings) ' Office', # constitu (623 firings) ' Constitution', # congression (556 firings) ' Congressional', # republ (553 firings) ' Republic', 'Republic', # team (537 firings) ' Team', ' Teams', # season (505 firings) ' Season', # navy (483 firings) ' Navy', # draft (422 firings) ' Draft', # president (418 firings) ' Presidential', ' Presidents', # author (410 firings) ' Authority', ' Authorization', # legislature (386 firings) ' Legislature', # cabinet (386 firings) ' Cabinet', # left (386 firings) ' Left', 'Left', # coali (359 firings) ' Coalition', # report (340 firings) ' Report', # bureau (317 firings) ' Bureau', # assemb (311 firings) ' Assembly', # side (308 firings) ' Side', # right (306 firings) ' Right', # empire (250 firings) ' Empire', # province (249 firings) ' Province', # forc (247 firings) ' Forces', # embassy (241 firings) ' Embassy', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'court-family', 1 def prog_L08F21943(input_tokens): """4 token families, 8 surface forms (train F1 of the uncapped optimum: 0.131). Fires at 365,805 of 36,786,566 positions. Fires at 0.99% of positions, and token rules do not serve features this dense: median test F1 0.086 across the 37,173 features above 0.39% of positions, against 0.763 for the 76,283 below 0.016%. Read this fire set as the best available approximation, not as what the feature does.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: references to specific events or entities in sports and legal contexts""" # --- end external annotation --- fire = { # - (12,243 firings) '-', # � (2,462 firings) ' �', # white (1,066 firings) ' White', ' white', 'white', 'White', # u (927 firings) ' U', ' u', } token = input_tokens[-1] if token not in fire: return "no match", 0 return '--family', 1 def prog_L08F21944(input_tokens): """39 token families, 51 surface forms (train F1 of the uncapped optimum: 0.662). Fires at 10,670 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: instances of the word "boat" in various contexts""" # --- end external annotation --- fire = { # boat (1,439 firings) ' boat', ' boats', 'boat', ' Boat', 'boats', # ship (947 firings) ' ship', ' ships', # vessel (634 firings) ' vessel', ' vessels', ' Vessel', # sail (345 firings) ' sail', ' sailing', ' sailed', ' sails', # ferry (271 firings) ' ferry', ' Ferry', # helicopt (251 firings) ' helicopter', # submarine (246 firings) ' submarine', # dock (151 firings) ' dock', ' docks', # raft (149 firings) ' raft', # balloon (128 firings) ' balloon', # kay (113 firings) ' kay', # acht (112 firings) 'acht', # submarin (108 firings) ' submarines', # porch (108 firings) ' porch', # oat (97 firings) 'oat', # fishermen (94 firings) ' fishermen', # torr (94 firings) ' Torres', # yacht (90 firings) ' yacht', # buoy (90 firings) ' buoy', # padd (90 firings) ' padd', # balcony (78 firings) ' balcony', # orry (77 firings) 'orry', # lat (71 firings) 'latable', # dolphin (71 firings) ' dolphin', # arge (68 firings) 'arge', # tank (64 firings) ' tanker', # canoe (63 firings) ' canoe', # torpedo (62 firings) ' torpedo', # rv (56 firings) ' RV', # titan (55 firings) ' Titanic', # dam (54 firings) ' dams', # paddle (51 firings) ' paddle', # fisherman (47 firings) ' fisherman', # afloat (42 firings) ' afloat', # marine (39 firings) 'marine', # tub (33 firings) 'tub', # ashore (33 firings) ' ashore', # lighthouse (27 firings) ' lighthouse', # overboard (27 firings) ' overboard', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'boat-family', 1 def prog_L08F21945(input_tokens): """3 token families, 4 surface forms (train F1 of the uncapped optimum: 0.101). Fires at 57,216 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: references to specifications and requirements in various contexts""" # --- end external annotation --- fire = { # : (7,424 firings) ':', ' :', # ] (738 firings) ' ]', # edit (171 firings) ' Edit', } token = input_tokens[-1] if token not in fire: return "no match", 0 return ':-family', 1 def prog_L08F21946(input_tokens): """3 token families, 3 surface forms (train F1 of the uncapped optimum: 0.214). Fires at 41,555 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: phrases indicating academic positions and affiliations""" # --- end external annotation --- fire = { # at (18,802 firings) ' at', # attend (129 firings) ' attended', # prestigi (120 firings) ' prestigious', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'at-family', 1 def prog_L08F21947(input_tokens): """50 token families, 57 surface forms (train F1 of the uncapped optimum: 0.709). Fires at 28,062 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: geographical locations and names of places, specifically in Canada""" # --- end external annotation --- fire = { # edmonton (524 firings) ' Edmonton', # sk (434 firings) ' SK', 'sk', 'SK', # winnipeg (403 firings) ' Winnipeg', # idaho (308 firings) ' Idaho', # finland (307 firings) ' Finland', # rapid (303 firings) ' Rapids', # ula (299 firings) 'ula', # kashmir (290 firings) ' Kashmir', # ault (270 firings) 'ault', 'aults', # bu (250 firings) 'BU', ' BU', # saskatchewan (249 firings) ' Saskatchewan', # jacksonville (227 firings) ' Jacksonville', # wellington (211 firings) ' Wellington', # kabul (208 firings) ' Kabul', # manitoba (205 firings) ' Manitoba', # wing (205 firings) ' Wings', # finnish (201 firings) ' Finnish', # fall (199 firings) ' Falls', # bargain (189 firings) ' bargaining', # iga (189 firings) 'iga', # gat (175 firings) ' Gat', # churchill (168 firings) ' Churchill', # aborigin (164 firings) ' Aboriginal', ' aboriginal', # iba (160 firings) 'iba', # arbor (159 firings) ' Arbor', # legion (156 firings) ' Legion', # regina (155 firings) ' Regina', # belaru (151 firings) ' Belarus', # punjab (150 firings) ' Punjab', # atoon (148 firings) 'atoon', # sask (144 firings) ' Sask', # ghana (142 firings) ' Ghana', # fin (139 firings) 'Fin', ' FIN', # donetsk (138 firings) ' Donetsk', # waterloo (138 firings) ' Waterloo', # curios (137 firings) ' Curiosity', # kingston (135 firings) ' Kingston', # stafford (135 firings) ' Stafford', # leaf (130 firings) ' Leaf', # arg (129 firings) ' Arg', 'Arg', # sherlock (128 firings) ' Sherlock', # joan (125 firings) ' Joan', # kun (125 firings) ' Kun', # wildfir (116 firings) ' wildfires', # zhou (115 firings) 'zhou', # caldwell (115 firings) ' Caldwell', # wildfire (115 firings) ' wildfire', # ammu (111 firings) 'ammu', # cfl (111 firings) ' CFL', # mandela (105 firings) ' Mandela', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'edmonton-family', 1 def prog_L08F21948(input_tokens): """46 token families, 55 surface forms (train F1 of the uncapped optimum: 0.473). Fires at 79,565 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: terms related to sports teams and their draft selections""" # --- end external annotation --- fire = { # year (3,655 firings) 'year', 'years', # day (1,660 firings) 'day', # time (1,197 firings) 'time', # term (1,150 firings) 'term', 'Term', # level (855 firings) 'level', 'levels', # season (828 firings) 'season', # clas (774 firings) 'class', 'Class', # round (702 firings) 'round', 'rounder', # game (660 firings) 'game', # end (644 firings) 'end', # inch (611 firings) 'inch', # party (594 firings) 'party', # weight (527 firings) 'weight', # degree (509 firings) 'degree', # sex (489 firings) 'sex', # minute (467 firings) 'minute', # page (449 firings) 'page', # century (398 firings) 'century', # genera (393 firings) 'generation', # siz (388 firings) 'sized', # point (380 firings) 'point', # star (375 firings) 'star', # state (372 firings) 'state', # ever (371 firings) 'ever', # style (354 firings) 'style', # night (347 firings) 'night', # sell (340 firings) 'selling', 'seller', # era (337 firings) 'era', # school (329 firings) 'school', # driv (328 firings) 'driving', # life (309 firings) 'life', # range (299 firings) 'range', 'Range', # bound (295 firings) 'bound', # team (294 firings) 'team', # larg (287 firings) 'largest', # rank (286 firings) 'ranked', 'ranking', # qual (277 firings) 'quality', # world (273 firings) 'world', # market (252 firings) 'market', # door (246 firings) 'door', # grade (246 firings) 'grade', # screen (238 firings) 'screen', # size (234 firings) 'size', # language (229 firings) 'language', # govern (225 firings) 'government', # length (203 firings) 'length', 'Length', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'year-family', 1 def prog_L08F21949(input_tokens): """5 token families, 5 surface forms (train F1 of the uncapped optimum: 0.067). Fires at 46,600 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: phrases indicating actions or states, particularly using the verb "is" in various contexts""" # --- end external annotation --- fire = { # : (5,205 firings) ':', # ? (1,629 firings) '?', # | (169 firings) ' |', # include (165 firings) ' include', # simp (130 firings) ' simply', } token = input_tokens[-1] if token not in fire: return "no match", 0 return ':-family', 1 def prog_L08F21950(input_tokens): """11 token families, 16 surface forms (train F1 of the uncapped optimum: 0.165). Fires at 168,956 of 36,786,566 positions. Fires at 0.46% of positions, and token rules do not serve features this dense: median test F1 0.086 across the 37,173 features above 0.39% of positions, against 0.763 for the 76,283 below 0.016%. Read this fire set as the best available approximation, not as what the feature does.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: predictions about future events""" # --- end external annotation --- fire = { # will (14,015 firings) ' will', 'will', # be (9,925 firings) ' be', # soon (2,237 firings) ' soon', # ll (2,087 firings) 'll', "'ll", # become (1,181 firings) ' become', ' Become', # eventual (758 firings) ' eventually', ' Eventually', # start (741 firings) ' start', # final (676 firings) ' finally', # able (628 firings) ' able', # future (592 firings) ' future', # becom (579 firings) ' becoming', ' becomes', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'will-family', 1 def prog_L08F21951(input_tokens): """13 token families, 21 surface forms (train F1 of the uncapped optimum: 0.675). Fires at 11,256 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: the word "particularly" and related expressions emphasizing significance or importance""" # --- end external annotation --- fire = { # particular (3,430 firings) ' particularly', ' particular', 'particularly', ' Particularly', # especial (2,489 firings) ' especially', ' Especially', 'especially', 'Especially', # specifical (734 firings) ' specifically', ' Specifically', 'Specifically', # unusual (244 firings) ' unusually', # special (220 firings) ' specially', # downright (186 firings) ' downright', # unique (130 firings) ' uniquely', # exceptional (120 firings) ' exceptionally', # notorious (99 firings) ' notoriously', # disproportionate (90 firings) ' disproportionately', # acute (41 firings) ' acutely', # uly (33 firings) 'uly', # chief (29 firings) ' chiefly', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'particular-family', 1 def prog_L08F21952(input_tokens): """43 token families, 52 surface forms (train F1 of the uncapped optimum: 0.542). Fires at 15,873 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: terms related to emergency situations or services""" # --- end external annotation --- fire = { # emergency (2,179 firings) ' emergency', ' Emergency', 'Emergency', # disast (424 firings) ' disaster', ' disasters', ' Disaster', # urgent (411 firings) ' urgent', ' urgently', # investigat (286 firings) ' investigative', ' Investigative', # rescue (236 firings) ' rescue', # interim (215 firings) ' interim', # evacua (172 firings) ' evacuation', # acute (157 firings) ' acute', # joint (156 firings) ' Joint', # evacuat (145 firings) ' evacuated', # pan (140 firings) ' panic', # hospital (123 firings) ' hospitalized', # exp (117 firings) ' exped', # emer (116 firings) 'emer', # urgency (112 firings) ' urgency', # cann (105 firings) ' canned', # evac (96 firings) ' evac', # amb (94 firings) ' Amber', # sav (92 firings) 'saving', # respiratory (91 firings) ' respiratory', # er (87 firings) ' ER', # improv (82 firings) ' improvised', ' improv', # emergenc (81 firings) ' emergencies', # ult (78 firings) ' ult', # intens (78 firings) ' intensive', # threaten (76 firings) 'threatening', # traumat (74 firings) 'traumatic', # rial (73 firings) 'rial', # specialty (72 firings) ' specialty', # omsday (71 firings) 'omsday', # mber (68 firings) 'mber', # gency (65 firings) 'gency', # ampa (63 firings) 'ampa', # romptu (62 firings) 'romptu', # ordinary (60 firings) 'ordinary', # extra (58 firings) 'extra', 'Extra', # expedi (56 firings) ' Expedition', # jaime (55 firings) ' Jaime', # ems (54 firings) ' EMS', 'EMS', # contingency (52 firings) ' contingency', # ourmet (51 firings) 'ourmet', # evacuate (49 firings) ' evacuate', # discretionary (41 firings) ' discretionary', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'emergency-family', 1 def prog_L08F21953(input_tokens): """34 token families, 47 surface forms (train F1 of the uncapped optimum: 0.695). Fires at 18,114 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: references to shooting incidents or violent events""" # --- end external annotation --- fire = { # shoot (5,589 firings) ' shooting', ' shooter', ' shootings', ' shoot', ' shooters', ' Shooting', ' Shoot', 'shoot', ' Shooter', # massacre (537 firings) ' massacre', ' Massacre', # bomb (498 firings) ' bombing', ' bombings', # gunman (436 firings) ' gunman', # kill (373 firings) ' killings', # gunshot (366 firings) ' gunshot', ' gunshots', # fir (295 firings) ' firing', # gunfire (233 firings) ' gunfire', # gunmen (225 firings) ' gunmen', # murd (204 firings) ' murders', # assassina (202 firings) ' assassination', # stabb (182 firings) ' stabbing', # rampage (177 firings) ' rampage', # homicide (157 firings) ' homicide', # hook (143 firings) ' Hook', # poison (131 firings) ' poisoning', # thrill (122 firings) ' thriller', # snip (122 firings) ' sniper', ' snipers', ' Sniper', # gunn (117 firings) ' gunned', # tarant (114 firings) ' Tarant', # newtown (102 firings) ' Newtown', # shootout (101 firings) ' shootout', # homicid (84 firings) ' homicides', # rifle (84 firings) ' Rifle', # ambush (83 firings) ' ambush', # robber (78 firings) ' robberies', # spree (70 firings) ' spree', # oot (69 firings) 'ooting', # paddock (67 firings) ' Paddock', # pist (67 firings) ' Pist', # hebdo (57 firings) ' Hebdo', # massacr (51 firings) ' massacres', # columb (51 firings) ' Columb', # pulse (50 firings) ' Pulse', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'shoot-family', 1 def prog_L08F21954(input_tokens): """14 token families, 16 surface forms (train F1 of the uncapped optimum: 0.589). Fires at 17,048 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: phrases that include the word "based." """ # --- end external annotation --- fire = { # bas (6,076 firings) ' based', ' Based', 'Based', # found (605 firings) ' founded', # focus (519 firings) ' focused', # center (283 firings) ' centered', # deriv (197 firings) ' derived', # root (178 firings) ' rooted', # rel (145 firings) ' relies', # stem (112 firings) ' stems', # dependent (99 firings) ' dependent', # revolv (90 firings) ' revolves', # model (85 firings) ' modeled', # ground (68 firings) ' grounded', # headquarter (64 firings) ' headquartered', # rest (58 firings) ' rests', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'bas-family', 1 def prog_L08F21955(input_tokens): """16 token families, 25 surface forms (train F1 of the uncapped optimum: 0.119). Fires at 101,614 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: phrases that describe different types or categories of things""" # --- end external annotation --- fire = { # a (14,918 firings) ' a', # an (3,662 firings) ' an', # " (2,444 firings) ' "', # like (2,237 firings) ' like', # ' (978 firings) " '", # the (950 firings) 'the', # call (684 firings) 'called', ' call', # essential (679 firings) ' essentially', ' Essentially', # effective (506 firings) ' effectively', # own (451 firings) ' own', # basical (414 firings) ' basically', # true (387 firings) ' true', 'true', # act (326 firings) ' act', ' acts', ' acting', ' acted', # ultimate (298 firings) ' ultimate', 'ultimate', 'Ultimate', # giant (296 firings) ' giant', # offic (285 firings) ' official', 'official', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'a-family', 1 def prog_L08F21956(input_tokens): """2 token families, 2 surface forms (train F1 of the uncapped optimum: 0.075). Fires at 125,000 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: concepts related to separation and distinction between various entities or ideas""" # --- end external annotation --- fire = { # anoth (386 firings) ' another', # clinton (342 firings) ' Clinton', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'anoth-family', 1 def prog_L08F21957(input_tokens): """33 token families, 45 surface forms (train F1 of the uncapped optimum: 0.256). Fires at 38,094 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: occurrences of the character sequence "ch" """ # --- end external annotation --- fire = { # z (1,458 firings) 'z', # ch (1,145 firings) 'ch', # ach (753 firings) 'ach', ' Ach', # ek (581 firings) 'ek', ' Ek', # ak (560 firings) 'ak', 'Ak', # iz (494 firings) 'iz', # ik (460 firings) 'ik', ' Ik', # ich (458 firings) 'ich', ' Ich', # ish (448 firings) 'ish', ' Ish', # ash (435 firings) 'ash', 'ASH', # az (385 firings) 'az', # ok (375 firings) 'ok', ' Ok', 'Ok', # av (315 firings) 'av', # space (290 firings) ' Space', # q (289 firings) 'q', # aj (278 firings) 'aj', # oh (210 firings) 'oh', # och (209 firings) 'och', # eb (193 firings) 'eb', ' Eb', # ff (183 firings) 'ff', # ky (182 firings) 'KY', # ev (155 firings) 'ev', # oj (142 firings) 'oj', # osh (131 firings) 'osh', # ark (127 firings) ' Ark', 'Ark', # ush (125 firings) 'ush', # akh (123 firings) 'akh', # uch (113 firings) 'uch', # esh (106 firings) 'esh', # kh (105 firings) 'kh', # bed (105 firings) ' Bed', 'Bed', # ech (103 firings) 'ech', # iy (100 firings) 'iy', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'z-family', 1 def prog_L08F21958(input_tokens): """8 token families, 9 surface forms (train F1 of the uncapped optimum: 0.128). Fires at 52,421 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: names of people and their social media handles""" # --- end external annotation --- fire = { # new (475 firings) 'New', 'News', # _ (419 firings) '_', # (@ (323 firings) ' (@', # j (295 firings) 'J', # & (293 firings) '&', # il (189 firings) 'IL', # z (184 firings) 'Z', # as (132 firings) 'AS', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'new-family', 1 def prog_L08F21959(input_tokens): """2 token families, 2 surface forms (train F1 of the uncapped optimum: 0.216). Fires at 10,540 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: instances of the contraction "won't" """ # --- end external annotation --- fire = { # t (5,655 firings) 't', # wont (31 firings) ' wont', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 't-family', 1 def prog_L08F21960(input_tokens): """12 token families, 15 surface forms (train F1 of the uncapped optimum: 0.655). Fires at 14,111 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: occurrences of the word "House" in various contexts, particularly in relation to legislative discussions""" # --- end external annotation --- fire = { # house (7,613 firings) ' House', 'House', ' HOUSE', # representat (840 firings) ' Representatives', ' Representative', # congression (329 firings) ' congressional', # major (206 firings) ' Majority', # hou (188 firings) 'HOU', # boehn (187 firings) ' Boehner', # sabha (160 firings) ' Sabha', # represent (90 firings) 'Represent', # lord (69 firings) ' Lords', # rep (52 firings) ' Reps', # hous (46 firings) ' Houses', # pulse (46 firings) ' Pulse', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'house-family', 1 def prog_L08F21961(input_tokens): """25 token families, 61 surface forms (train F1 of the uncapped optimum: 0.137). Fires at 99,668 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: phrases related to risk and consequences""" # --- end external annotation --- fire = { # to (8,246 firings) ' to', # can (1,243 firings) ' can', # allow (1,160 firings) ' allowing', ' allows', ' allow', ' allowed', 'allows', # more (1,154 firings) ' more', # mean (897 firings) ' means', ' meaning', ' mean', # mak (787 firings) ' makes', ' making', # prevent (692 firings) ' prevent', ' preventing', ' prevented', ' prevents', # result (597 firings) ' result', ' resulting', ' resulted', 'result', # lead (552 firings) ' leading', ' lead', ' leads', # avoid (541 firings) ' avoid', ' avoiding', ' avoids', ' avoided', 'avoid', ' Avoid', # caus (523 firings) ' causing', ' caused', ' causes', # creat (513 firings) ' creating', ' created', ' creates', # why (491 firings) ' why', # able (441 firings) ' able', # difficult (388 firings) ' difficult', ' difficulties', # leav (335 firings) ' leaving', ' leaves', # poss (334 firings) ' possible', ' possibility', # prompt (324 firings) ' prompting', ' prompted', ' prompt', ' prompts', # les (309 firings) ' less', # acces (300 firings) ' access', # reduce (299 firings) ' reduce', # thu (299 firings) ' thus', ' Thus', 'thus', # reduc (280 firings) ' reducing', ' reduces', ' reduced', ' reduction', # opportun (279 firings) ' opportunity', ' opportunities', # abil (265 firings) ' ability', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'to-family', 1 def prog_L08F21962(input_tokens): """14 token families, 23 surface forms (train F1 of the uncapped optimum: 0.682). Fires at 12,131 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: mentions of the term "Open" and related concepts""" # --- end external annotation --- fire = { # open (7,086 firings) ' open', ' Open', 'Open', 'open', ' Opening', ' OPEN', ' openness', 'opens', # op (386 firings) ' Op', # brew (221 firings) ' Brew', 'Brew', # modern (198 firings) ' Modern', # clear (163 firings) ' Clear', 'Clear', # apache (154 firings) ' Apache', # clean (146 firings) ' Clean', # libre (74 firings) ' Libre', # fen (67 firings) ' Fen', # directx (65 firings) ' DirectX', # oak (58 firings) 'Oak', # opengl (56 firings) ' OpenGL', # clos (55 firings) ' Closed', # plo (35 firings) ' PLoS', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'open-family', 1 def prog_L08F21963(input_tokens): """7 token families, 12 surface forms (train F1 of the uncapped optimum: 0.079). Fires at 90,087 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: programming constructs and function definitions""" # --- end external annotation --- fire = { # : (1,651 firings) ':', ' :', # a (447 firings) 'a', 'ça', 'ía', 'ña', # 00 (331 firings) '00', # 30 (320 firings) '30', # . (281 firings) ' .', # de (266 firings) ' de', # | (227 firings) ' |', '|', } token = input_tokens[-1] if token not in fire: return "no match", 0 return ':-family', 1 def prog_L08F21964(input_tokens): """1 token families, 1 surface forms (train F1 of the uncapped optimum: 0.294). Fires at 102,111 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: prepositions, particularly "in," used in various contexts""" # --- end external annotation --- fire = { # in (89,560 firings) ' in', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'in-family', 1 def prog_L08F21965(input_tokens): """1 token families, 1 surface forms (train F1 of the uncapped optimum: 0.070). Fires at 34,880 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: instances of the phrase "you can't" or variations thereof""" # --- end external annotation --- fire = { # � (31,361 firings) '�', } token = input_tokens[-1] if token not in fire: return "no match", 0 return '�-family', 1 def prog_L08F21966(input_tokens): """17 token families, 49 surface forms (train F1 of the uncapped optimum: 0.098). Fires at 285,689 of 36,786,566 positions. Fires at 0.78% of positions, and token rules do not serve features this dense: median test F1 0.086 across the 37,173 features above 0.39% of positions, against 0.763 for the 76,283 below 0.016%. Read this fire set as the best available approximation, not as what the feature does.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: phrases indicating alerts or announcements of significant events""" # --- end external annotation --- fire = { # a (12,264 firings) ' a', # an (3,973 firings) ' an', ' An', # you (3,899 firings) ' you', ' You', 'You', 'you', # your (2,617 firings) ' your', ' Your', 'Your', ' yours', 'your', # or (2,185 firings) ' or', # had (2,112 firings) ' had', # what (1,676 firings) ' what', ' What', 'What', 'what', 'WHAT', ' WHAT', # last (1,573 firings) 'Last', ' Last', # my (1,359 firings) ' my', ' My', ' MY', # ! (1,221 firings) '!', ' !', # work (1,184 firings) ' work', ' working', 'worker', ' Work', 'workers', ' WORK', # someth (1,174 firings) ' something', 'Something', ' Something', 'something', # going (817 firings) ' going', ' Going', # me (766 firings) ' me', # com (761 firings) ' coming', 'coming', ' Coming', # ve (754 firings) 've', "'ve", # someone (738 firings) ' someone', ' Someone', 'Someone', 'someone', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'a-family', 1 def prog_L08F21967(input_tokens): """32 token families, 54 surface forms (train F1 of the uncapped optimum: 0.331). Fires at 20,027 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: terms related to exercising rights and powers""" # --- end external annotation --- fire = { # exercise (537 firings) ' exercise', ' Exercise', # abus (256 firings) ' abusing', ' abused', # wield (214 firings) ' wielding', ' wield', ' wielded', # assum (213 firings) ' assumed', ' assuming', ' assumes', # regain (193 firings) ' regain', ' regained', # resort (173 firings) ' resort', ' resorted', # grant (161 firings) ' granted', # invok (157 firings) ' invoked', ' invoking', # lift (154 firings) ' lift', ' lifting', # free (151 firings) ' freely', # exert (134 firings) ' exert', ' exerted', # flex (120 firings) ' flex', 'flex', # enlist (111 firings) ' enlisted', ' enlist', # assert (103 firings) ' assert', ' asserting', 'assert', # exercis (102 firings) ' exercising', # resum (98 firings) ' resumed', ' resumes', # erc (88 firings) 'erc', 'ERC', # relinqu (82 firings) ' relinqu', # wrest (82 firings) ' wrest', # exhaust (80 firings) ' exhausted', ' exhaust', ' exhausting', # practic (79 firings) ' practicing', # exerc (78 firings) ' exercised', ' exerc', # refrain (77 firings) ' refrain', # relieve (76 firings) ' relieve', # resume (76 firings) ' resume', # imb (75 firings) ' imb', # invoke (73 firings) ' invoke', 'invoke', # tread (64 firings) ' tread', # impair (58 firings) ' impaired', # revok (57 firings) ' revoked', # discre (51 firings) ' discretion', # behav (51 firings) ' behaving', ' behaved', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'exercise-family', 1 def prog_L08F21968(input_tokens): """2 token families, 3 surface forms (train F1 of the uncapped optimum: 0.054). Fires at 24,106 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: phrases expressing disappointment or frustration about unmet expectations""" # --- end external annotation --- fire = { # ve (914 firings) 've', # ­ (110 firings) '\xad', ' \xad', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 've-family', 1 def prog_L08F21969(input_tokens): """6 token families, 10 surface forms (train F1 of the uncapped optimum: 0.076). Fires at 29,969 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: phrases expressing hypothetical scenarios and their consequences""" # --- end external annotation --- fire = { # have (4,389 firings) ' have', ' Have', # able (603 firings) ' able', # without (472 firings) ' without', # hav (438 firings) ' having', 'having', # nev (401 firings) ' never', ' NEVER', # expect (321 firings) ' expect', ' expecting', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'have-family', 1 def prog_L08F21970(input_tokens): """24 token families, 61 surface forms (train F1 of the uncapped optimum: 0.180). Fires at 324,633 of 36,786,566 positions. Fires at 0.88% of positions, and token rules do not serve features this dense: median test F1 0.086 across the 37,173 features above 0.39% of positions, against 0.763 for the 76,283 below 0.016%. Read this fire set as the best available approximation, not as what the feature does.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: phrases that express desires or goals""" # --- end external annotation --- fire = { # to (30,192 firings) ' to', ' TO', # with (7,592 firings) ' with', ' WITH', # : (6,247 firings) ':', ' :', # about (3,539 firings) ' about', # ? (2,498 firings) '?', ' ?', # want (2,223 firings) ' want', ' wanted', ' wants', ' wanting', ' WANT', # aim (2,128 firings) ' aimed', ' aim', ' aims', ' aiming', ' Aim', 'Aim', # plan (1,963 firings) ' plans', ' plan', # going (1,650 firings) ' going', 'going', # goal (1,590 firings) ' goal', ' goals', ' Goals', # decid (1,478 firings) ' decided', ' decides', ' deciding', # ser (1,390 firings) ' series', ' serial', # set (1,384 firings) ' set', ' SET', # take (1,355 firings) ' take', 'take', # long (1,268 firings) ' long', 'long', ' longing', ' LONG', # effort (1,219 firings) ' efforts', ' effort', # go (1,199 firings) ' go', # mis (1,159 firings) ' mission', ' Mission', 'Mission', # campaign (1,150 firings) ' campaign', ' campaigns', ' campaigning', # focus (1,072 firings) ' focused', ' focusing', ' focuses', 'focused', # focu (1,016 firings) ' focus', 'focus', # attack (1,013 firings) ' attacks', ' attack', 'attacks', # look (842 firings) ' looking', # love (831 firings) ' love', 'love', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'to-family', 1 def prog_L08F21971(input_tokens): """1 token families, 1 surface forms (train F1 of the uncapped optimum: 0.145). Fires at 115,811 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: references to the United States and notable events or attacks related to it""" # --- end external annotation --- fire = { # the (107,251 firings) ' the', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'the-family', 1 def prog_L08F21972(input_tokens): """30 token families, 113 surface forms (train F1 of the uncapped optimum: 0.303). Fires at 65,075 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: terms related to software development and version control""" # --- end external annotation --- fire = { # project (2,750 firings) ' project', ' projects', ' Project', 'Project', ' Projects', 'project', 'projects', # release (2,293 firings) ' release', ' Release', 'Release', 'release', # ver (1,182 firings) ' version', ' Version', 'version', 'Version', 'VERSION', ' VERS', # source (1,057 firings) ' source', ' Source', 'source', 'SOURCE', # build (998 firings) ' build', ' builds', ' Build', 'Build', 'build', # chang (712 firings) ' changes', ' chang', 'Changes', ' Changes', # update (697 firings) ' update', # updat (465 firings) ' updates', ' updating', ' Updates', # license (463 firings) ' license', ' License', 'License', 'license', # arch (444 firings) 'Arch', ' archive', ' Arch', ' Archives', 'arch', ' archives', 'archive', 'archives', # package (421 firings) ' package', ' Package', 'package', 'Package', # beta (415 firings) ' beta', ' Beta', 'beta', 'Beta', # bug (372 firings) ' bug', ' bugs', ' Bug', 'Bug', 'bugs', # fix (362 firings) ' fixes', 'fixes', 'fix', ' Fixes', # releas (329 firings) ' releases', ' Releases', ' Released', # library (321 firings) ' library', 'library', # download (315 firings) ' download', ' downloads', 'download', ' Downloads', # install (310 firings) ' install', 'install', 'Install', ' Install', ' installs', ' installer', # deploy (294 firings) ' deploy', ' deployment', ' deploying', ' Deploy', ' deployments', 'Deploy', # packag (275 firings) ' packages', ' packaging', 'packages', # patch (270 firings) ' patch', ' patches', 'Patch', 'patch', # version (265 firings) ' versions', # maintain (244 firings) ' maintain', ' maintaining', # github (243 firings) ' GitHub', ' github', 'github', ' Github', # stable (240 firings) ' stable', 'stable', # librar (238 firings) ' libraries', ' Libraries', # branch (213 firings) ' branch', ' branches', ' branching', # debian (195 firings) ' Debian', ' debian', 'debian', # git (186 firings) ' git', 'git', ' Git', # documenta (176 firings) ' documentation', ' Documentation', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'project-family', 1 def prog_L08F21973(input_tokens): """47 token families, 68 surface forms (train F1 of the uncapped optimum: 0.605). Fires at 14,944 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: occurrences of the word "anchor" and its variations""" # --- end external annotation --- fire = { # anchor (525 firings) ' anchor', ' anchored', ' anchors', # flagship (443 firings) ' flagship', # anch (334 firings) 'anch', ' Anch', ' anch', # premise (249 firings) ' premise', # pin (205 firings) 'pin', # spear (190 firings) ' spear', ' Spear', ' spears', # accent (176 firings) ' accent', ' accents', # atl (147 firings) ' Atl', 'Atl', 'atl', # shore (129 firings) ' shore', # ground (127 firings) ' grounded', ' grounding', # stabil (124 firings) ' stabil', ' stabilization', ' stabilized', # beacon (118 firings) ' beacon', ' Beacon', # kimmel (114 firings) ' Kimmel', # pillar (114 firings) ' pillar', ' pillars', ' Pillar', # bast (113 firings) ' bast', # evangel (102 firings) ' evangel', # staple (102 firings) ' staple', # parasit (100 firings) ' parasites', ' parasitic', # satire (99 firings) ' satire', # orth (98 firings) ' orth', ' Orth', # centerpiece (93 firings) ' centerpiece', # buoy (86 firings) ' buoy', # mant (86 firings) ' Mant', ' mant', # puppet (84 firings) ' puppet', ' Puppet', # aden (84 firings) 'aden', ' Aden', # sheen (83 firings) ' Sheen', # outpost (83 firings) ' outpost', # contain (83 firings) ' containment', 'container', # cornerstone (80 firings) ' cornerstone', # abin (80 firings) 'abin', # upl (78 firings) ' upl', # kath (74 firings) ' Kath', # collater (72 firings) ' collateral', # mast (72 firings) ' mast', # medi (72 firings) ' medi', # parasite (70 firings) ' parasite', # glitt (69 firings) ' glitter', # hook (68 firings) ' hooks', # undersc (67 firings) ' undersc', # affirm (66 firings) ' affirm', # foothold (62 firings) ' foothold', # catalyst (60 firings) ' catalyst', # canary (58 firings) ' Canary', # indoctr (55 firings) ' indoctr', # lana (51 firings) ' Lana', # sabot (46 firings) ' sabot', # subst (45 firings) ' subst', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'anchor-family', 1 def prog_L08F21974(input_tokens): """15 token families, 65 surface forms (train F1 of the uncapped optimum: 0.142). Fires at 382,037 of 36,786,566 positions. Fires at 1.04% of positions, and token rules do not serve features this dense: median test F1 0.086 across the 37,173 features above 0.39% of positions, against 0.763 for the 76,283 below 0.016%. Read this fire set as the best available approximation, not as what the feature does.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: issues related to system performance and connection errors""" # --- end external annotation --- fire = { # . (18,143 firings) '.', # for (6,315 firings) ' for', # your (2,610 firings) ' your', ' Your', 'Your', # aft (2,354 firings) ' after', ' After', 'after', # user (2,166 firings) ' users', ' user', ' Users', 'Users', ' User', 'user', 'User', 'users', # some (1,758 firings) ' some', ' Some', 'some', ' SOME', # block (1,625 firings) 'block', ' block', ' blocked', ' blocking', ' blocks', 'blocking', ' blocker', ' blockers', 'blocks', ' Blocks', # because (1,590 firings) ' because', ' Because', 'Because', 'because', # system (1,393 firings) ' system', ' systems', ' System', 'system', 'System', ' SYSTEM', # may (1,276 firings) ' may', ' MAY', # work (1,221 firings) ' work', ' working', ' works', ' Work', ' Works', 'Works', ' WORK', # custom (1,150 firings) ' customers', ' customer', ' custom', ' Customers', ' Custom', ' customized', # un (1,087 firings) ' Un', ' un', 'Un', # update (1,035 firings) ' update', ' Update', 'update', # service (998 firings) ' service', ' Service', 'service', 'Service', } token = input_tokens[-1] if token not in fire: return "no match", 0 return '.-family', 1 def prog_L08F21975(input_tokens): """1 token families, 1 surface forms (train F1 of the uncapped optimum: 0.124). Fires at 97,533 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: instances of numerical values or ratings""" # --- end external annotation --- fire = { # \n (91,030 firings) '\n', } token = input_tokens[-1] if token not in fire: return "no match", 0 return '\\n-family', 1 def prog_L08F21976(input_tokens): """11 token families, 15 surface forms (train F1 of the uncapped optimum: 0.075). Fires at 81,325 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: references to software release information across various platforms""" # --- end external annotation --- fire = { # and (20,289 firings) ' and', # or (2,790 firings) ' or', # devic (338 firings) ' devices', # & (317 firings) ' &', # tablet (299 firings) ' tablets', ' tablet', ' Tablet', # xbox (272 firings) ' Xbox', 'Xbox', # ps (251 firings) ' PS', # ver (239 firings) ' version', # user (222 firings) ' users', # android (215 firings) ' Android', 'Android', # bas (214 firings) 'based', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'and-family', 1 def prog_L08F21977(input_tokens): """32 token families, 94 surface forms (train F1 of the uncapped optimum: 0.135). Fires at 337,035 of 36,786,566 positions. Fires at 0.92% of positions, and token rules do not serve features this dense: median test F1 0.086 across the 37,173 features above 0.39% of positions, against 0.763 for the 76,283 below 0.016%. Read this fire set as the best available approximation, not as what the feature does.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: expressions of inability or obstacles in achieving goals""" # --- end external annotation --- fire = { # to (19,731 firings) ' to', # t (5,338 firings) 't', "'t", # i (4,269 firings) ' I', # not (4,065 firings) ' not', 'Notable', # just (2,569 firings) ' just', ' Just', 'Just', 'just', # want (2,470 firings) ' want', ' wanted', ' wants', ' wanting', # would (2,324 firings) ' would', ' Would', 'would', 'Would', # ask (2,284 firings) ' asked', ' ask', ' asking', ' asks', ' Ask', # get (1,693 firings) ' get', ' gets', # re (1,621 firings) "'re", 're', # what (1,576 firings) ' What', 'What', # do (1,574 firings) ' do', ' Do', 'Do', # going (1,545 firings) ' going', # don (1,478 firings) ' don', # start (1,438 firings) ' start', ' started', ' starting', 'Starting', ' Starting', ' Start', 'Start', 'started', ' Started', # need (1,428 firings) ' need', ' needed', ' NEED', # look (1,390 firings) ' looking', ' look', ' looked', ' Looking', ' Look', # try (1,354 firings) ' trying', ' try', ' Try', ' Trying', 'Try', 'try', # got (1,305 firings) ' got', 'got', # did (1,278 firings) ' did', ' Did', 'did', # took (1,265 firings) ' took', # might (1,214 firings) ' might', ' Might', 'might', # take (1,185 firings) ' take', # didn (1,150 firings) ' didn', ' Didn', 'didn', # here (1,129 firings) ' Here', 'Here', 'here', # you (1,052 firings) ' You', 'You', 'you', # set (1,042 firings) ' set', 'SET', # tak (1,024 firings) ' taking', ' takes', ' Taking', 'taking', 'Taking', # should (959 firings) ' should', # find (952 firings) ' find', ' finding', ' finds', ' Finding', # put (911 firings) ' put', ' puts', # why (894 firings) ' why', ' Why', 'Why', 'why', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'to-family', 1 def prog_L08F21978(input_tokens): """42 token families, 56 surface forms (train F1 of the uncapped optimum: 0.316). Fires at 127,094 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: verbs related to actions and achievements""" # --- end external annotation --- fire = { # made (4,903 firings) ' made', ' Made', # took (3,080 firings) ' took', # got (2,302 firings) ' got', ' Got', 'got', # did (2,070 firings) ' did', ' DID', # came (2,006 firings) ' came', ' Came', # went (1,922 firings) ' went', ' Went', 'went', # won (1,410 firings) ' won', ' Won', # wrote (1,387 firings) ' wrote', 'wrote', # lost (1,310 firings) ' lost', # start (1,238 firings) ' started', # spent (1,001 firings) ' spent', # bought (971 firings) ' bought', ' Bought', # gave (967 firings) ' gave', # sent (951 firings) ' sent', # happen (929 firings) ' happened', # creat (923 firings) ' created', # receiv (920 firings) ' received', # sign (871 firings) ' signed', # vot (727 firings) ' voted', # fell (717 firings) ' fell', # arriv (695 firings) ' arrived', # mov (664 firings) ' moved', # ran (649 firings) ' ran', # pass (626 firings) ' passed', ' Passed', # tri (607 firings) ' tried', # broke (605 firings) ' broke', # miss (593 firings) ' missed', # rais (592 firings) ' raised', # chose (565 firings) ' chose', # learn (560 firings) ' learned', ' Learned', # pull (557 firings) ' pulled', # knew (543 firings) ' knew', # chang (525 firings) ' changed', ' Changed', 'changed', # dropp (523 firings) ' dropped', # grew (456 firings) ' grew', # scor (424 firings) ' scored', # prom (418 firings) ' promised', # pick (397 firings) ' picked', # manag (392 firings) ' managed', # kept (368 firings) ' kept', # suffer (362 firings) ' suffered', # purchas (351 firings) ' purchased', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'made-family', 1 def prog_L08F21979(input_tokens): """1 token families, 5 surface forms (train F1 of the uncapped optimum: 0.555). Fires at 42,725 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: occurrences of the word "first." """ # --- end external annotation --- fire = { # first (25,658 firings) ' first', ' First', 'First', 'first', ' FIRST', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'first-family', 1 def prog_L08F21980(input_tokens): """8 token families, 19 surface forms (train F1 of the uncapped optimum: 0.112). Fires at 123,354 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: technical programming languages and their syntax""" # --- end external annotation --- fire = { # � (6,941 firings) ' �', # h (845 firings) 'h', ' h', # k (767 firings) 'k', ' k', # b (600 firings) 'b', ' b', # n (557 firings) 'n', ' n', 'én', # ap (514 firings) 'AP', 'ap', ' ap', ' Ap', 'Ap', # f (501 firings) 'f', ' f', # v (461 firings) 'v', ' v', } token = input_tokens[-1] if token not in fire: return "no match", 0 return '�-family', 1 def prog_L08F21981(input_tokens): """6 token families, 7 surface forms (train F1 of the uncapped optimum: 0.062). Fires at 39,531 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: query-based phrasing or questions regarding knowledge and understanding""" # --- end external annotation --- fire = { # you (2,845 firings) ' you', ' YOU', # do (1,137 firings) ' do', # go (418 firings) ' go', # try (341 firings) ' trying', # begin (162 firings) ' begin', # possib (156 firings) ' possibly', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'you-family', 1 def prog_L08F21982(input_tokens): """4 token families, 5 surface forms (train F1 of the uncapped optimum: 0.077). Fires at 18,974 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: phrases indicating locations or proximity to various areas""" # --- end external annotation --- fire = { # beyond (506 firings) ' beyond', ' Beyond', # above (183 firings) ' above', # surround (104 firings) ' surrounding', # abroad (61 firings) ' abroad', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'beyond-family', 1 def prog_L08F21983(input_tokens): """11 token families, 17 surface forms (train F1 of the uncapped optimum: 0.616). Fires at 146,546 of 36,786,566 positions. Fires at 0.40% of positions, and token rules do not serve features this dense: median test F1 0.086 across the 37,173 features above 0.39% of positions, against 0.763 for the 76,283 below 0.016%. Read this fire set as the best available approximation, not as what the feature does.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: terms related to geographical features and settings""" # --- end external annotation --- fire = { # teenag (1,261 firings) ' teenager', ' teenagers', # legislature (711 firings) ' legislature', ' Legislature', # late (668 firings) ' lately', # congressman (664 firings) ' congressman', ' Congressman', # midfield (649 firings) ' midfielder', # courtesy (452 firings) ' courtesy', # politician (449 firings) ' politician', ' politicians', # wikileak (421 firings) ' WikiLeaks', ' Wikileaks', # cryptocurrency (410 firings) ' cryptocurrency', # introduc (393 firings) 'Introduction', ' Introduction', # businessman (388 firings) ' businessman', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'teenag-family', 1 def prog_L08F21984(input_tokens): """31 token families, 38 surface forms (train F1 of the uncapped optimum: 0.659). Fires at 7,133 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: references to multi-year contracts or agreements""" # --- end external annotation --- fire = { # multi (1,564 firings) ' multi', ' Multi', 'Multi', 'multi', # mult (500 firings) ' mult', 'Mult', ' multic', 'mult', # mega (329 firings) ' mega', # illion (161 firings) 'illion', # jugg (142 firings) ' jugg', # triple (141 firings) ' triple', # anglo (115 firings) ' Anglo', # multim (110 firings) ' multim', # quadru (86 firings) ' quadru', # multit (78 firings) ' multit', # multif (71 firings) ' multif', # socio (70 firings) ' socio', # umbo (65 firings) 'umbo', # binge (63 firings) ' binge', # multip (60 firings) ' multip', ' multiply', # multitude (56 firings) ' multitude', # oft (52 firings) ' oft', # ult (51 firings) 'ULT', # quad (48 firings) ' Quad', # omnip (46 firings) ' omnip', # multimedia (40 firings) ' multimedia', # juggl (38 firings) ' juggling', # uber (33 firings) ' uber', # quasi (33 firings) ' quasi', # intra (32 firings) ' intra', # integ (29 firings) ' Integ', # pepp (28 firings) ' pepp', # unic (26 firings) ' unic', # single (23 firings) 'single', # guatem (19 firings) ' Guatem', # complex (18 firings) 'complex', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'multi-family', 1 def prog_L08F21985(input_tokens): """43 token families, 54 surface forms (train F1 of the uncapped optimum: 0.602). Fires at 14,441 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: references to negative experiences or impacts""" # --- end external annotation --- fire = { # negat (1,377 firings) ' negative', 'negative', ' Negative', ' negatives', # posit (496 firings) ' positive', # oid (277 firings) 'oid', # ugly (238 firings) ' ugly', # depress (211 firings) ' depressing', ' depressive', ' depressed', # verse (200 firings) 'verse', 'versely', # scath (184 firings) ' scathing', # disciplinary (178 firings) ' disciplinary', # damn (174 firings) ' damning', # adverse (166 firings) ' adverse', ' adversely', # sour (164 firings) ' sour', ' Sour', # unwant (160 firings) ' unwanted', # minu (160 firings) ' minus', 'minus', # midterm (155 firings) ' midterm', # nude (132 firings) ' nude', # downward (124 firings) ' downward', # unhappy (121 firings) ' unhappy', # negative (120 firings) ' negatively', # subject (115 firings) ' subjective', # nasty (111 firings) ' nasty', # rible (106 firings) 'rible', # ironical (104 firings) ' Ironically', 'Ironically', # inaccurate (101 firings) ' inaccurate', # los (99 firings) ' loser', ' losers', # dispar (97 firings) ' dispar', # detriment (89 firings) ' detrimental', # downside (87 firings) ' downside', # downgrade (78 firings) ' downgrade', # derogatory (77 firings) ' derogatory', # salty (77 firings) ' salty', # traumat (77 firings) ' traumatic', # neg (70 firings) 'Neg', # gloomy (68 firings) ' gloomy', # pessimist (64 firings) ' pessimistic', # negativ (61 firings) ' negativity', # worse (60 firings) ' Worse', # unfavor (59 firings) ' unfavorable', # disapprov (58 firings) ' disapproval', # sensory (54 firings) ' sensory', # derivat (52 firings) ' derivative', # disappro (50 firings) ' disappro', # undesir (49 firings) ' undesirable', # detri (38 firings) ' detriment', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'negat-family', 1 def prog_L08F21986(input_tokens): """20 token families, 27 surface forms (train F1 of the uncapped optimum: 0.294). Fires at 53,003 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: sports outcomes and references to victories and defeats""" # --- end external annotation --- fire = { # win (2,693 firings) ' win', ' wins', # victory (2,045 firings) ' victory', ' Victory', # los (1,331 firings) ' loss', 'loss', ' Loss', # defeat (872 firings) ' defeat', ' defeats', # match (653 firings) ' match', # perform (545 firings) ' performance', # open (407 firings) ' opener', 'opening', # draw (366 firings) ' draw', # clash (327 firings) ' clash', # winn (273 firings) ' winner', 'winner', # sidelin (272 firings) ' sidelines', # upset (198 firings) ' upset', # fin (188 firings) 'final', # friend (180 firings) ' friendly', # tie (173 firings) ' tie', # overtime (158 firings) ' overtime', # matchup (156 firings) ' matchup', # comeback (156 firings) ' comeback', # loss (137 firings) ' losses', # victor (135 firings) ' victories', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'win-family', 1 def prog_L08F21987(input_tokens): """33 token families, 46 surface forms (train F1 of the uncapped optimum: 0.365). Fires at 45,091 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: references to sports conferences and divisions""" # --- end external annotation --- fire = { # league (2,591 firings) ' League', ' league', # divi (1,085 firings) ' division', ' Division', 'division', # confer (1,017 firings) ' Conference', 'conference', # playoff (874 firings) ' playoffs', ' playoff', ' Playoff', ' Playoffs', # championship (834 firings) ' Championship', # cup (758 firings) ' Cup', ' Cups', # ser (755 firings) ' Series', # sec (392 firings) ' SEC', 'SEC', # schedule (360 firings) ' schedule', # afc (313 firings) ' AFC', # eastern (313 firings) ' Eastern', # fa (272 firings) ' FA', # nfc (246 firings) ' NFC', # ncaa (216 firings) ' NCAA', # ten (209 firings) ' Ten', # acc (203 firings) ' ACC', 'ACC', # stand (193 firings) ' standings', # stanley (186 firings) ' Stanley', # hou (175 firings) 'HOU', # cs (174 firings) 'CS', # calendar (174 firings) ' calendar', # promo (167 firings) ' promotion', # fc (157 firings) 'FC', # al (151 firings) ' AL', # wizard (141 firings) ' Wizards', # prem (134 firings) ' prem', 'Prem', # ast (132 firings) 'AST', # premiership (131 firings) ' Premiership', # pac (128 firings) ' Pac', # bracket (126 firings) ' bracket', ' brackets', # nl (123 firings) ' NL', 'NL', # conc (117 firings) ' CONC', # cub (116 firings) ' Cubs', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'league-family', 1 def prog_L08F21988(input_tokens): """2 token families, 2 surface forms (train F1 of the uncapped optimum: 0.055). Fires at 39,606 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: instances of strong affirmations or definitive statements""" # --- end external annotation --- fire = { # \n (39,030 firings) '\n', # salt (107 firings) ' Salt', } token = input_tokens[-1] if token not in fire: return "no match", 0 return '\\n-family', 1 def prog_L08F21989(input_tokens): """6 token families, 9 surface forms (train F1 of the uncapped optimum: 0.583). Fires at 11,061 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: mentions of the concept of "role" in various contexts""" # --- end external annotation --- fire = { # role (4,065 firings) ' role', ' Role', 'Role', 'role', # involve (558 firings) ' involvement', # rol (409 firings) ' roles', # participa (191 firings) ' participation', # responsibil (98 firings) ' responsibilities', # implicat (96 firings) ' implicated', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'role-family', 1 def prog_L08F21990(input_tokens): """4 token families, 5 surface forms (train F1 of the uncapped optimum: 0.094). Fires at 126,431 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: quotes or speech from characters""" # --- end external annotation --- fire = { # — (4,193 firings) ' —', # -- (2,553 firings) ' --', # say (966 firings) ' says', # ," (881 firings) ',"', ' ,"', } token = input_tokens[-1] if token not in fire: return "no match", 0 return '—-family', 1 def prog_L08F21991(input_tokens): """16 token families, 25 surface forms (train F1 of the uncapped optimum: 0.109). Fires at 193,995 of 36,786,566 positions. Fires at 0.53% of positions, and token rules do not serve features this dense: median test F1 0.086 across the 37,173 features above 0.39% of positions, against 0.763 for the 76,283 below 0.016%. Read this fire set as the best available approximation, not as what the feature does.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: topics that have garnered significant attention or coverage""" # --- end external annotation --- fire = { # a (20,165 firings) ' a', # its (2,999 firings) ' its', # more (2,500 firings) ' more', # new (2,457 firings) ' new', # some (1,578 firings) ' some', # much (1,192 firings) ' much', ' MUCH', # little (967 firings) ' little', 'little', # atten (908 firings) ' attention', ' Attention', # receiv (830 firings) ' received', ' receiving', # anoth (737 firings) ' another', # lot (622 firings) ' lot', # big (604 firings) ' big', # same (589 firings) ' same', ' Same', # major (552 firings) ' major', 'major', # gain (538 firings) ' gained', ' gaining', ' gain', ' gains', # second (509 firings) ' second', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'a-family', 1 def prog_L08F21992(input_tokens): """2 token families, 8 surface forms (train F1 of the uncapped optimum: 0.708). Fires at 19,357 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: phrases that indicate something is "best known" or notable for a specific reason""" # --- end external annotation --- fire = { # best (11,418 firings) ' best', ' Best', 'Best', 'best', ' BEST', # worst (1,211 firings) ' worst', 'worst', ' Worst', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'best-family', 1 def prog_L08F21993(input_tokens): """9 token families, 15 surface forms (train F1 of the uncapped optimum: 0.671). Fires at 7,144 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: references to a specific topic or subject within various contexts""" # --- end external annotation --- fire = { # subject (3,093 firings) ' subject', ' subjects', ' Subject', 'Subject', 'subject', ' Subjects', # top (670 firings) ' topic', # exempt (119 firings) ' exempt', 'exempt', # ject (88 firings) 'JECT', # prophet (77 firings) ' Prophet', # commence (35 firings) ' commencement', # object (32 firings) 'object', # perse (20 firings) ' Perse', # propriet (19 firings) ' propriet', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'subject-family', 1 def prog_L08F21994(input_tokens): """31 token families, 37 surface forms (train F1 of the uncapped optimum: 0.629). Fires at 12,433 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: references to the continent of Asia and related discussions""" # --- end external annotation --- fire = { # asia (1,598 firings) ' Asia', 'asia', 'Asia', # asian (1,300 firings) ' Asian', ' Asians', 'Asian', # africa (1,051 firings) ' Africa', # oversea (284 firings) ' overseas', # southeast (274 firings) ' Southeast', # america (206 firings) ' Americas', # electronic (194 firings) ' Electronics', # manila (175 firings) ' Manila', # armenia (126 firings) ' Armenia', # achi (117 firings) 'achi', # adia (93 firings) 'adia', # bush (93 firings) ' bush', # buddhism (89 firings) ' Buddhism', # austr (84 firings) ' Austral', 'Austral', # eura (83 firings) ' Euras', # rica (79 firings) 'rica', # arabian (76 firings) ' Arabian', # orthern (68 firings) 'orthern', # indo (61 firings) ' Indo', ' indo', # lia (61 firings) ' Lia', # anime (58 firings) ' Anime', # orea (56 firings) 'orea', # ae (55 firings) ' Ae', # orient (52 firings) ' Oriental', # xia (49 firings) 'xia', # ney (47 firings) 'NEY', # thu (44 firings) 'thus', # caucasu (36 firings) ' Caucasus', # peopl (36 firings) ' Peoples', # aero (34 firings) ' aeros', # malay (33 firings) ' Malays', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'asia-family', 1 def prog_L08F21995(input_tokens): """2 token families, 3 surface forms (train F1 of the uncapped optimum: 0.036). Fires at 25,418 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: author names and publication details in academic references""" # --- end external annotation --- fire = { # , (21,602 firings) ',', ' ,', # et (167 firings) ' et', } token = input_tokens[-1] if token not in fire: return "no match", 0 return ',-family', 1 def prog_L08F21996(input_tokens): """50 token families, 53 surface forms (train F1 of the uncapped optimum: 0.657). Fires at 35,075 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: mentions of the name "Lloyd" in various contexts""" # --- end external annotation --- fire = { # lane (512 firings) ' Lane', # li (482 firings) 'li', # bak (472 firings) ' Baker', ' baker', # worth (384 firings) 'worth', # ington (289 firings) 'INGTON', 'ington', # floyd (282 firings) ' Floyd', # aki (278 firings) 'aki', # elliott (271 firings) ' Elliott', # elli (265 firings) ' Ellis', # murdoch (257 firings) ' Murdoch', # goldman (254 firings) ' Goldman', # mos (249 firings) ' Moss', # villa (244 firings) ' Villa', # self (241 firings) ' Self', 'Self', # lloyd (224 firings) ' Lloyd', # bank (221 firings) ' Banks', # hernandez (219 firings) ' Hernandez', # iana (219 firings) 'iana', # gonzalez (210 firings) ' Gonzalez', # tayyip (210 firings) ' Tayyip', # shah (208 firings) ' Shah', # dav (204 firings) ' Davies', # read (199 firings) ' Reader', # friedman (191 firings) ' Friedman', # sandwich (179 firings) ' sandwich', # saf (177 firings) ' Saf', # messi (174 firings) ' Messi', # nic (168 firings) ' Nic', # palm (168 firings) ' palm', # durant (166 firings) ' Durant', # churchill (164 firings) ' Churchill', # allan (159 firings) ' Allan', # marin (149 firings) ' Mariners', # elliot (147 firings) ' Elliot', # chandl (145 firings) ' Chandler', # boyd (145 firings) ' Boyd', # fitzgerald (145 firings) ' Fitzgerald', # vick (144 firings) ' Vick', # lennon (143 firings) ' Lennon', # dwight (142 firings) ' Dwight', # longh (140 firings) ' Longh', # diana (138 firings) ' Diana', # randall (137 firings) ' Randall', # maverick (137 firings) ' Mavericks', # doyle (135 firings) ' Doyle', # elman (130 firings) 'elman', # milton (129 firings) ' Milton', # lionel (127 firings) ' Lionel', # zion (127 firings) ' Zion', # walton (126 firings) ' Walton', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'lane-family', 1 def prog_L08F21997(input_tokens): """13 token families, 33 surface forms (train F1 of the uncapped optimum: 0.072). Fires at 153,739 of 36,786,566 positions. Fires at 0.42% of positions, and token rules do not serve features this dense: median test F1 0.086 across the 37,173 features above 0.39% of positions, against 0.763 for the 76,283 below 0.016%. Read this fire set as the best available approximation, not as what the feature does.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: references to specific media, brands, or notable events""" # --- end external annotation --- fire = { # s (3,726 firings) 's', # thi (2,671 firings) ' this', ' THIS', # game (1,056 firings) ' game', ' Game', ' GAME', # these (629 firings) ' these', # system (596 firings) ' system', ' systems', ' System', # campaign (540 firings) ' campaign', ' Campaign', 'campaign', # book (519 firings) ' book', ' Book', 'book', ' Books', ' BOOK', # team (506 firings) ' team', # post (502 firings) ' post', ' Post', ' postings', ' POST', # ser (480 firings) ' series', ' Series', ' Serial', # life (448 firings) ' life', ' Life', # bitcoin (406 firings) ' Bitcoin', ' bitcoin', ' Bitcoins', # blog (401 firings) ' blog', ' Blog', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 's-family', 1 def prog_L08F21998(input_tokens): """1 token families, 5 surface forms (train F1 of the uncapped optimum: 0.499). Fires at 14,577 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: references to the concept of "state," particularly in various contexts""" # --- end external annotation --- fire = { # state (8,483 firings) ' state', 'state', 'State', ' STATE', 'STATE', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'state-family', 1 def prog_L08F21999(input_tokens): """5 token families, 13 surface forms (train F1 of the uncapped optimum: 0.115). Fires at 23,085 of 36,786,566 positions.""" # --- begin external annotation (ignored at runtime) --- """prev SAE description: terms related to subscription add-ons and features""" # --- end external annotation --- fire = { # add (688 firings) ' add', ' Add', 'add', ' adds', 'Add', 'Adds', ' ADD', 'adding', 'ADD', # dlc (197 firings) ' DLC', # in (157 firings) 'IN', # ons (149 firings) 'ons', # ison (83 firings) 'ison', } token = input_tokens[-1] if token not in fire: return "no match", 0 return 'add-family', 1