chat template COMET

results/mac-results_rpp_with_mnt_2048_metrics.csv

@@ -1,43 +1,43 @@
-model,rpp,meteor,spbleu,bleu_1,rouge_l,ews_score,repetition_score,total_repetitions,rap,num_incomplete_translations,num_max_output_tokens,eval_time
-Qwen/Qwen2-72B-Instruct,1.00,0.47892081926016034,18.953782447211417,0.18953782447211412,0.4521569211424892,0.0,0.09885260370697264,0.09885260370697264,0.4768835566987496,0,0,15.977934686672551
-Qwen/Qwen2-72B-Instruct,1.02,0.47968366756320546,19.0137490702917,0.190137490702917,0.45197054564395384,0.0,0.08120035304501325,0.08120035304501325,0.47800479714301924,0,0,15.785525154457194
-Qwen/Qwen2-72B-Instruct,1.04,0.476362681282195,18.52063321160408,0.1852063321160408,0.44894205112535357,0.0,0.10503089143865843,0.10503089143865843,0.4742108804113156,0,0,15.759046778464255
-Qwen/Qwen2-72B-Instruct,1.06,0.46796363839506494,17.982473951038504,0.17982473951038513,0.441041000758417,0.0,0.07325684024713151,0.07325684024713151,0.466484924380041,0,0,15.851721094439542
-Qwen/Qwen2-72B-Instruct,1.08,0.45975781977116453,17.067954025424825,0.17067954025424834,0.43174454568405374,0.0,0.04766107678729038,0.04766107678729038,0.45881038677745906,0,0,15.924095322153574
-Qwen/Qwen2-72B-Instruct,1.10,0.4515778511124262,16.22452191616505,0.16224521916165044,0.4243494436012768,0.0,0.05736981465136805,0.05736981465136805,0.4504587226091121,0,0,15.984995586937334
-Qwen/Qwen2-7B-Instruct,1.00,0.442240791493943,14.38814929350883,0.1438814929350883,0.4161653982436496,0.0,12.81288614298323,12.81288614298323,0.3256127432585341,6,2,2.2162400706090026
-Qwen/Qwen2-7B-Instruct,1.02,0.4400998640836595,15.16172261831792,0.1516172261831792,0.41646468384393903,0.0,7.1562224183583405,7.1562224183583405,0.356523119707599,6,1,1.586054721977052
-Qwen/Qwen2-7B-Instruct,1.04,0.4390136558190875,14.958631815014014,0.14958631815014015,0.41437075031278525,0.0,0.1853486319505737,0.1853486319505737,0.43553983411076963,1,0,1.0123565754633717
-Qwen/Qwen2-7B-Instruct,1.06,0.43283215765150845,14.28087386760537,0.14280873867605376,0.4072536098321874,0.0,0.2030008826125331,0.2030008826125331,0.42908711780306746,2,0,1.002647837599294
-Qwen/Qwen2-7B-Instruct,1.08,0.423560805217557,13.659683698817107,0.1365968369881711,0.3967743499621005,0.0,0.22153574580759047,0.22153574580759047,0.4195681312114654,0,0,1.000882612533098
-Qwen/Qwen2-7B-Instruct,1.10,0.41350531365414334,12.922649874705083,0.12922649874705075,0.38793653105711157,0.0,0.17740511915269197,0.17740511915269197,0.410371282612889,0,0,1.0114739629302736
-internlm/internlm2_5-7b-chat,1.00,0.4289996929258777,14.734881589173108,0.1473488158917311,0.4096656779566272,0.0,12.751103265666373,12.751103265666373,0.3161377411176559,0,2,3.063548102383054
-internlm/internlm2_5-7b-chat,1.02,0.42662469043021944,14.583816688798015,0.1458381668879802,0.4073494324538721,0.0,9.824360105913504,9.824360105913504,0.3288814067035907,0,1,2.127978817299206
-internlm/internlm2_5-7b-chat,1.04,0.4154585167056314,13.534659133050223,0.13534659133050217,0.3970519149664059,0.0,6.527802294792586,6.527802294792586,0.3410387160469389,0,1,2.136804942630185
-internlm/internlm2_5-7b-chat,1.06,0.40395886478553783,12.346740971499404,0.12346740971499397,0.38739729431897413,0.0,6.533980582524272,6.533980582524272,0.3315547851383537,1,1,2.153574580759047
-internlm/internlm2_5-7b-chat,1.08,0.3873176839854818,11.075674965706343,0.11075674965706349,0.3727703497077143,0.0,9.820829655781113,9.820829655781113,0.2985977687581126,1,1,2.148278905560459
-internlm/internlm2_5-7b-chat,1.10,0.3769306874386757,10.305163787094209,0.10305163787094214,0.3636303538397795,0.0,6.525154457193292,6.525154457193292,0.30942993176825007,1,1,2.172109443954104
-microsoft/Phi-3.5-mini-instruct,1.00,0.37968315453481294,8.71296896471494,0.08712968964714936,0.35850654380084257,0.0,28.42630185348632,28.42630185348632,0.23960387748577275,0,6,5.484554280670785
-microsoft/Phi-3.5-mini-instruct,1.02,0.3780585837553919,10.291240080163629,0.10291240080163629,0.3548291746069422,0.0,10.696381288614297,10.696381288614297,0.2873016094123469,0,2,2.6337157987643423
-microsoft/Phi-3.5-mini-instruct,1.04,0.35471613338457425,8.966881655527896,0.08966881655527896,0.33033473881366876,0.0,3.7484554280670785,3.7484554280670785,0.3116318162898625,0,1,1.9285083848190645
-microsoft/Phi-3.5-mini-instruct,1.06,0.3298733737040869,7.076233088011138,0.07076233088011138,0.3021292838048098,0.0,0.10767872903795234,0.10767872903795234,0.3283460949961551,0,0,1.3159752868490733
-microsoft/Phi-3.5-mini-instruct,1.08,0.3054541022592767,5.105510599247868,0.05105510599247869,0.2734682521006445,0.0,3.29567519858782,3.29567519858782,0.2718263558255611,0,1,2.054721977052074
-microsoft/Phi-3.5-mini-instruct,1.10,0.2903831801547132,4.091958857999118,0.04091958857999117,0.25172729782510583,0.0,0.07766990291262135,0.07766990291262135,0.28941072284774055,0,0,1.9638128861429833
-shenzhi-wang/Llama3.1-70B-Chinese-Chat,1.00,0.46111105085070175,17.87914973742753,0.17879149737427524,0.4340569906195885,0.0,0.088261253309797,0.088261253309797,0.4593579935085897,0,0,16.17828773168579
-shenzhi-wang/Llama3.1-70B-Chinese-Chat,1.02,0.45715172190795766,17.436884594979905,0.17436884594979907,0.4309998159694175,0.0,0.09267431597528684,0.09267431597528684,0.45532756172969063,0,0,15.968225948808474
-shenzhi-wang/Llama3.1-70B-Chinese-Chat,1.04,0.45604679603642545,17.440173470996626,0.17440173470996626,0.43028250668592427,0.0,0.13062665489849956,0.13062665489849956,0.4534907811633413,0,0,16.050308914386584
-shenzhi-wang/Llama3.1-70B-Chinese-Chat,1.06,0.45204163612198556,16.89523258317781,0.16895232583177808,0.42594768159318286,0.0,0.12533097969991175,0.12533097969991175,0.44960959510411197,0,0,15.972639011473962
-shenzhi-wang/Llama3.1-70B-Chinese-Chat,1.08,0.4442106511292453,16.16623784482793,0.16166237844827927,0.4196173445876259,0.0,0.18711385701676964,0.18711385701676964,0.4406628043373358,0,0,15.998234774933804
-shenzhi-wang/Llama3.1-70B-Chinese-Chat,1.10,0.43798379261381615,15.601722576240661,0.1560172257624066,0.4135114396257313,0.0,0.08649602824360106,0.08649602824360106,0.4363516968290564,0,0,16.018534863195057
-shenzhi-wang/Llama3.1-8B-Chinese-Chat,1.00,0.43363250166217604,15.209540658023398,0.1520954065802339,0.4087331128794971,0.0,5.798764342453663,5.798764342453663,0.36177551894565085,0,1,1.7625772285966461
-shenzhi-wang/Llama3.1-8B-Chinese-Chat,1.02,0.4329054166518245,15.19102241646024,0.1519102241646024,0.4070425935096388,0.0,5.77846425419241,5.77846425419241,0.3613372514106729,0,1,1.7493380406001766
-shenzhi-wang/Llama3.1-8B-Chinese-Chat,1.04,0.43204787009562073,15.051351661582961,0.15051351661582968,0.40610700867871297,0.0,0.11827007943512798,0.11827007943512798,0.42985292605837966,0,0,1.0467784642541924
-shenzhi-wang/Llama3.1-8B-Chinese-Chat,1.06,0.42754293149125455,14.449130821290163,0.1444913082129016,0.4003685841336065,0.0,0.176522506619594,0.176522506619594,0.4243183675786234,0,0,1.0450132391879965
-shenzhi-wang/Llama3.1-8B-Chinese-Chat,1.08,0.4206626658729054,13.933703757385222,0.13933703757385224,0.3962652599741041,0.0,0.21888790820829657,0.21888790820829657,0.41674375951677545,0,0,1.0423654015887025
-shenzhi-wang/Llama3.1-8B-Chinese-Chat,1.10,0.41112112403991513,13.303738403756983,0.1330373840375698,0.3870517415572443,0.0,0.13857016769638128,0.13857016769638128,0.4086785702502848,0,0,1.0458958517210943
-shenzhi-wang/Mistral-7B-v0.3-Chinese-Chat,1.00,0.4068335357738006,13.565136550617618,0.13565136550617618,0.38651661218475886,0.0,0.1059135039717564,0.1059135039717564,0.40498051631383686,0,0,1.1871138570167696
-shenzhi-wang/Mistral-7B-v0.3-Chinese-Chat,1.02,0.4064914781094827,13.42987641622816,0.1342987641622816,0.3863631817026095,0.0,6.238305383936452,6.238305383936452,0.3357933166106685,0,1,1.8102383053839364
-shenzhi-wang/Mistral-7B-v0.3-Chinese-Chat,1.04,0.40548598969949756,13.295092218891954,0.13295092218891957,0.3851644729228478,0.0,0.1297440423654016,0.1297440423654016,0.403228526175785,0,0,1.1597528684907326
-shenzhi-wang/Mistral-7B-v0.3-Chinese-Chat,1.06,0.4014727027723293,13.10860799057166,0.1310860799057166,0.3808877762129881,0.0,0.20741394527802295,0.20741394527802295,0.3979249127866333,0,0,1.1650485436893203
-shenzhi-wang/Mistral-7B-v0.3-Chinese-Chat,1.08,0.3987018836449559,12.850537785783194,0.1285053778578319,0.3791139249625647,0.0,0.2903795233892321,0.2903795233892321,0.39380632270315097,0,0,1.1685789938217124
-shenzhi-wang/Mistral-7B-v0.3-Chinese-Chat,1.10,0.39322399484561765,12.361161644811926,0.12361161644811926,0.3735331110324942,0.0,0.1500441306266549,0.1500441306266549,0.3906969938466816,0,0,1.1562224183583407
+model,rpp,comet,meteor,spbleu,bleu_1,rouge_l,ews_score,repetition_score,total_repetitions,rap,translation_completeness,num_max_output_tokens
+Qwen/Qwen2-72B-Instruct,1.00,0.7570169081439618,0.47892081926016034,18.953782447211417,0.18953782447211412,0.4520943414365094,0.0,0.09885260370697264,0.09885260370697264,0.7537966634953808,1.0,0
+Qwen/Qwen2-72B-Instruct,1.02,0.7573581492858762,0.47968366756320546,19.0137490702917,0.190137490702917,0.4520458023287854,0.0,0.08120035304501325,0.08120035304501325,0.7547074311557754,1.0,0
+Qwen/Qwen2-72B-Instruct,1.04,0.7574983964372501,0.476362681282195,18.52063321160408,0.1852063321160408,0.44890405687175916,0.0,0.10503089143865843,0.10503089143865843,0.7540766638515739,1.0,0
+Qwen/Qwen2-72B-Instruct,1.06,0.7563851308422812,0.46796363839506494,17.982473951038504,0.17982473951038513,0.44106220363072907,0.0,0.07325684024713151,0.07325684024713151,0.7539950363948406,1.0,0
+Qwen/Qwen2-72B-Instruct,1.08,0.7554470169417962,0.45975781977116453,17.067954025424825,0.17067954025424834,0.4317496244624469,0.0,0.04766107678729038,0.04766107678729038,0.753890250753016,1.0,0
+Qwen/Qwen2-72B-Instruct,1.10,0.7550096405046973,0.4515778511124262,16.22452191616505,0.16224521916165044,0.42420831846665985,0.0,0.05736981465136805,0.05736981465136805,0.7531385283434515,1.0,0
+Qwen/Qwen2-7B-Instruct,1.00,0.7457721890965501,0.442240791493943,14.38814929350883,0.1438814929350883,0.4160546438584113,0.0,12.81288614298323,12.81288614298323,0.5490966301804292,0.9947043248014121,2
+Qwen/Qwen2-7B-Instruct,1.02,0.7474080664545482,0.4400998640836595,15.16172261831792,0.1516172261831792,0.4160749445105633,0.0,7.1562224183583405,7.1562224183583405,0.605472251398711,0.9947043248014121,1
+Qwen/Qwen2-7B-Instruct,1.04,0.7484375637974869,0.4390136558190875,14.958631815014014,0.14958631815014015,0.41381116329574685,0.0,0.1853486319505737,0.1853486319505737,0.742515336500048,0.999117387466902,0
+Qwen/Qwen2-7B-Instruct,1.06,0.7471612840233287,0.43283215765150845,14.28087386760537,0.14280873867605376,0.4069074178015074,0.0,0.2030008826125331,0.2030008826125331,0.7406965407448669,0.9982347749338041,0
+Qwen/Qwen2-7B-Instruct,1.08,0.74519401609845,0.423560805217557,13.659683698817107,0.1365968369881711,0.3966315425573522,0.0,0.22153574580759047,0.22153574580759047,0.7381694832783203,1.0,0
+Qwen/Qwen2-7B-Instruct,1.10,0.7432071187016402,0.41350531365414334,12.922649874705083,0.12922649874705075,0.3876193185383525,0.0,0.17740511915269197,0.17740511915269197,0.737574218462684,1.0,0
+internlm/internlm2_5-7b-chat,1.00,0.739699612254078,0.4289996929258777,14.734881589173108,0.1473488158917311,0.4096466800937898,0.0,12.751103265666373,12.751103265666373,0.5450982095784719,1.0,2
+internlm/internlm2_5-7b-chat,1.02,0.740223803961056,0.42662469043021944,14.583816688798015,0.1458381668879802,0.4071727106228415,0.0,9.824360105913504,9.824360105913504,0.5706323412193933,1.0,1
+internlm/internlm2_5-7b-chat,1.04,0.7398856264610577,0.4154585167056314,13.534659133050223,0.13534659133050217,0.39686577135897183,0.0,6.527802294792586,6.527802294792586,0.6073521998554903,1.0,1
+internlm/internlm2_5-7b-chat,1.06,0.7379362287241489,0.40395886478553783,12.346740971499404,0.12346740971499397,0.3872447044295494,0.0,6.533980582524272,6.533980582524272,0.6056712925221626,0.999117387466902,1
+internlm/internlm2_5-7b-chat,1.08,0.7319988705684732,0.3873176839854818,11.075674965706343,0.11075674965706349,0.37243529096686095,0.0,9.820829655781113,9.820829655781113,0.5643254582029298,0.999117387466902,1
+internlm/internlm2_5-7b-chat,1.10,0.7295350462119345,0.3769306874386757,10.305163787094209,0.10305163787094214,0.36344961557595074,0.0,6.525154457193292,6.525154457193292,0.5988898943353679,0.999117387466902,1
+microsoft/Phi-3.5-mini-instruct,1.00,0.7107840433177544,0.37968315453481294,8.71296896471494,0.08712968964714936,0.35898743959012847,0.0,28.42630185348632,28.42630185348632,0.4485492990665038,1.0,6
+microsoft/Phi-3.5-mini-instruct,1.02,0.7164765837070485,0.3780585837553919,10.291240080163629,0.10291240080163629,0.35469527324272765,0.0,10.696381288614297,10.696381288614297,0.5444787777612768,1.0,2
+microsoft/Phi-3.5-mini-instruct,1.04,0.7111233387336411,0.35471613338457425,8.966881655527896,0.08966881655527896,0.3300979657678754,0.0,3.7484554280670785,3.7484554280670785,0.6247493045810047,1.0,1
+microsoft/Phi-3.5-mini-instruct,1.06,0.7024363270136286,0.3298733737040869,7.076233088011138,0.07076233088011138,0.3019513312669543,0.0,0.10767872903795234,0.10767872903795234,0.6991841213752065,1.0,0
+microsoft/Phi-3.5-mini-instruct,1.08,0.6882111219210848,0.3054541022592767,5.105510599247868,0.05105510599247869,0.27360300072970145,0.0,3.29567519858782,3.29567519858782,0.6124452738619192,1.0,1
+microsoft/Phi-3.5-mini-instruct,1.10,0.6712992989638161,0.2903831801547132,4.091958857999118,0.04091958857999117,0.251653275009876,0.0,0.07766990291262135,0.07766990291262135,0.6690512007506374,1.0,0
+shenzhi-wang/Llama3.1-70B-Chinese-Chat,1.00,0.7501818982248062,0.46111105085070175,17.87914973742753,0.17879149737427524,0.43406620570095644,0.0,0.088261253309797,0.088261253309797,0.747329847981863,1.0,0
+shenzhi-wang/Llama3.1-70B-Chinese-Chat,1.02,0.7485114382045625,0.45715172190795766,17.436884594979905,0.17436884594979907,0.4311385932640979,0.0,0.09267431597528684,0.09267431597528684,0.7455246732136059,1.0,0
+shenzhi-wang/Llama3.1-70B-Chinese-Chat,1.04,0.7500591586357918,0.45604679603642545,17.440173470996626,0.17440173470996626,0.4302844557731285,0.0,0.13062665489849956,0.13062665489849956,0.7458552866168927,1.0,0
+shenzhi-wang/Llama3.1-70B-Chinese-Chat,1.06,0.748812871571673,0.45204163612198556,16.89523258317781,0.16895232583177808,0.42600267747458376,0.0,0.12533097969991175,0.12533097969991175,0.7447841638756355,1.0,0
+shenzhi-wang/Llama3.1-70B-Chinese-Chat,1.08,0.7473851635144647,0.4442106511292453,16.16623784482793,0.16166237844827927,0.41951294705858744,0.0,0.18711385701676964,0.18711385701676964,0.7414159053529263,1.0,0
+shenzhi-wang/Llama3.1-70B-Chinese-Chat,1.10,0.7465709781131172,0.43798379261381615,15.601722576240661,0.1560172257624066,0.41325629329409785,0.0,0.08649602824360106,0.08649602824360106,0.743788967986371,1.0,0
+shenzhi-wang/Llama3.1-8B-Chinese-Chat,1.00,0.7426396049131678,0.43363250166217604,15.209540658023398,0.1520954065802339,0.40892082351514747,0.0,5.798764342453663,5.798764342453663,0.619577239776096,1.0,1
+shenzhi-wang/Llama3.1-8B-Chinese-Chat,1.02,0.7436477056353469,0.4329054166518245,15.19102241646024,0.1519102241646024,0.4068967964789407,0.0,5.77846425419241,5.77846425419241,0.6207074516423631,1.0,1
+shenzhi-wang/Llama3.1-8B-Chinese-Chat,1.04,0.7440943776351209,0.43204787009562073,15.051351661582961,0.15051351661582968,0.4062008380201262,0.0,0.11827007943512798,0.11827007943512798,0.7403141356055205,1.0,0
+shenzhi-wang/Llama3.1-8B-Chinese-Chat,1.06,0.7426502735395928,0.42754293149125455,14.449130821290163,0.1444913082129016,0.4001409979222783,0.0,0.176522506619594,0.176522506619594,0.7370491441666694,1.0,0
+shenzhi-wang/Llama3.1-8B-Chinese-Chat,1.08,0.7408098006080129,0.4206626658729054,13.933703757385222,0.13933703757385224,0.39648242686762036,0.0,0.21888790820829657,0.21888790820829657,0.7339083936807739,1.0,0
+shenzhi-wang/Llama3.1-8B-Chinese-Chat,1.10,0.7392685912871718,0.41112112403991513,13.303738403756983,0.1330373840375698,0.38709595815635034,0.0,0.13857016769638128,0.13857016769638128,0.7348764469929082,1.0,0
+shenzhi-wang/Mistral-7B-v0.3-Chinese-Chat,1.00,0.7240239171358935,0.4068335357738006,13.565136550617618,0.13565136550617618,0.38663950670554986,0.0,0.1059135039717564,0.1059135039717564,0.7207261791424407,1.0,0
+shenzhi-wang/Mistral-7B-v0.3-Chinese-Chat,1.02,0.7263097057327799,0.4064914781094827,13.42987641622816,0.1342987641622816,0.38636978210251594,0.0,6.238305383936452,6.238305383936452,0.5999878425713037,1.0,1
+shenzhi-wang/Mistral-7B-v0.3-Chinese-Chat,1.04,0.7276128307708258,0.40548598969949756,13.295092218891954,0.13295092218891957,0.38512037299356977,0.0,0.1297440423654016,0.1297440423654016,0.7235619893938705,1.0,0
+shenzhi-wang/Mistral-7B-v0.3-Chinese-Chat,1.06,0.7276865132383193,0.4014727027723293,13.10860799057166,0.1310860799057166,0.38049527863066884,0.0,0.20741394527802295,0.20741394527802295,0.7212559915451495,1.0,0
+shenzhi-wang/Mistral-7B-v0.3-Chinese-Chat,1.08,0.726393195584298,0.3987018836449559,12.850537785783194,0.1285053778578319,0.37889459557464955,0.0,0.2903795233892321,0.2903795233892321,0.717473994791502,1.0,0
+shenzhi-wang/Mistral-7B-v0.3-Chinese-Chat,1.10,0.7244012304511832,0.39322399484561765,12.361161644811926,0.12361161644811926,0.37334138070076656,0.0,0.1500441306266549,0.1500441306266549,0.7197459635880831,1.0,0