Files
2023-11-11 19:44:09 +00:00

765 lines
21 KiB
JSON

{
"model_dir": "models/openllama-3b",
"distribution_id": "math_easy",
"timestamp": "2023-10-20 01:18:43",
"eval_loss": 0.6825920939445496,
"eval_score": -0.24484191834926605,
"eval_brier_score": 0.24484191834926605,
"eval_average_probability": 0.5053142309188843,
"eval_accuracy": 0.7973333333333333,
"eval_probabilities": [
0.515242338180542,
0.5111019015312195,
0.5108321905136108,
0.5105446577072144,
0.4995576739311218,
0.5015202164649963,
0.5014011263847351,
0.5007079243659973,
0.5025910139083862,
0.5115234851837158,
0.5123100280761719,
0.5129885077476501,
0.5024678707122803,
0.506177544593811,
0.5028885006904602,
0.5248365998268127,
0.49887579679489136,
0.5232630968093872,
0.4988650977611542,
0.49980661273002625,
0.4999678432941437,
0.4948975443840027,
0.5065024495124817,
0.5190327167510986,
0.5100163817405701,
0.5121020078659058,
0.5100820660591125,
0.5055842399597168,
0.5067341327667236,
0.5079879760742188,
0.49956926703453064,
0.5006097555160522,
0.4993750751018524,
0.5047982335090637,
0.5106545090675354,
0.5126189589500427,
0.4988892078399658,
0.49905717372894287,
0.5019606351852417,
0.5005311965942383,
0.5025065541267395,
0.5003291368484497,
0.50541090965271,
0.5026548504829407,
0.5085741281509399,
0.4999961853027344,
0.4996791183948517,
0.4995323717594147,
0.5050346255302429,
0.5151140689849854,
0.5042612552642822,
0.5023995637893677,
0.5008424520492554,
0.5008217096328735,
0.5132308602333069,
0.5139747858047485,
0.5084176063537598,
0.5114474296569824,
0.510020911693573,
0.5108993053436279,
0.5078394412994385,
0.5126341581344604,
0.51350337266922,
0.5120657086372375,
0.5019673109054565,
0.5002868175506592,
0.5147026777267456,
0.5146227478981018,
0.5144838094711304,
0.49569615721702576,
0.49727872014045715,
0.5075371265411377,
0.5009492635726929,
0.5010395050048828,
0.5017671585083008,
0.5074124336242676,
0.5050039291381836,
0.5048250555992126,
0.5033344030380249,
0.5064655542373657,
0.5123952627182007,
0.4996633231639862,
0.5006742477416992,
0.49729034304618835,
0.49982818961143494,
0.5002691149711609,
0.5015039443969727,
0.4988865554332733,
0.5006097555160522,
0.4979075789451599,
0.5086551308631897,
0.508033037185669,
0.5021083354949951,
0.5008074641227722,
0.4995729923248291,
0.5012193322181702,
0.5006383061408997,
0.503563404083252,
0.5033133625984192,
0.5064713954925537,
0.5049858689308167,
0.5107787251472473,
0.5023692846298218,
0.49285659193992615,
0.5052188038825989,
0.5236098170280457,
0.5310222506523132,
0.5083763003349304,
0.5066385865211487,
0.508374810218811,
0.5095425248146057,
0.5107754468917847,
0.512405276298523,
0.5148349404335022,
0.49814119935035706,
0.5005640983581543,
0.5002983808517456,
0.49965497851371765,
0.5059593915939331,
0.4987843632698059,
0.5109472870826721,
0.5133187174797058,
0.5121330618858337,
0.5004523396492004,
0.5007458329200745,
0.5022890567779541,
0.5011858940124512,
0.501801609992981,
0.5002762675285339,
0.5007044672966003,
0.5024741888046265,
0.5035625696182251,
0.5020695924758911,
0.5028650760650635,
0.5026698708534241,
0.5044881105422974,
0.5228583812713623,
0.501000702381134,
0.5002762675285339,
0.49813276529312134,
0.5003061294555664,
0.508410632610321,
0.5106928944587708,
0.5086479783058167,
0.5079150199890137,
0.5210515260696411,
0.5054596662521362,
0.4995456635951996,
0.49529555439949036,
0.5080796480178833,
0.4994702935218811,
0.501373827457428,
0.4983091354370117,
0.5019300580024719,
0.5046060681343079,
0.5035164952278137,
0.5051698684692383,
0.5093424916267395,
0.5057141184806824,
0.506300687789917,
0.5072007775306702,
0.5016548037528992,
0.5018800497055054,
0.5021072030067444,
0.5015007257461548,
0.5022938251495361,
0.5030829310417175,
0.5019411444664001,
0.5106671452522278,
0.5097770094871521,
0.5120174288749695,
0.5006939172744751,
0.5023943185806274,
0.5056152939796448,
0.5013095140457153,
0.49870312213897705,
0.5037303566932678,
0.5021265149116516,
0.5009855628013611,
0.5048725008964539,
0.5010706782341003,
0.5010035634040833,
0.4998011589050293,
0.49905887246131897,
0.5015472173690796,
0.49881625175476074,
0.49951034784317017,
0.50130695104599,
0.5010033249855042,
0.4954354465007782,
0.49563246965408325,
0.5098755955696106,
0.5004312992095947,
0.49961748719215393,
0.5000422596931458,
0.5096888542175293,
0.5000325441360474,
0.5013961791992188,
0.5005120635032654,
0.5000162720680237,
0.4996790289878845,
0.5108420848846436,
0.5000812411308289,
0.5066307783126831,
0.5233086347579956,
0.5048919916152954,
0.5200716257095337,
0.5101045966148376,
0.5086562633514404,
0.5056284666061401,
0.4970698356628418,
0.5039623379707336,
0.5082937479019165,
0.5446583032608032,
0.536361038684845,
0.5388292670249939,
0.5025307536125183,
0.5028770565986633,
0.5032330751419067,
0.501214325428009,
0.494588702917099,
0.5177946090698242,
0.4972660541534424,
0.49634966254234314,
0.4994610846042633,
0.5014930963516235,
0.5078816413879395,
0.5021491050720215,
0.5098927617073059,
0.5123560428619385,
0.5109851956367493,
0.5188221335411072,
0.5008295178413391,
0.5012413263320923,
0.5047503709793091,
0.5105997323989868,
0.5021579265594482,
0.4995090961456299,
0.5072723627090454,
0.5042780041694641,
0.49804073572158813,
0.4987751543521881,
0.4953358769416809,
0.5012505650520325,
0.4990858733654022,
0.4995633065700531,
0.5007351040840149,
0.5008159279823303,
0.4983828365802765,
0.5042833089828491,
0.5017117261886597,
0.498250812292099,
0.5014602541923523,
0.5089671611785889,
0.5179307460784912,
0.5102884769439697,
0.5087992548942566,
0.5096799731254578,
0.5043991804122925,
0.5031564235687256,
0.5060471296310425,
0.507271409034729,
0.5118258595466614,
0.5111958980560303,
0.5003632307052612,
0.503638744354248,
0.5011246204376221,
0.5105330944061279,
0.5093284845352173,
0.5072362422943115,
0.5068445801734924,
0.5060505867004395,
0.5030291676521301,
0.5006821155548096,
0.50147545337677,
0.50196373462677,
0.5008406639099121,
0.49678727984428406,
0.5005685091018677,
0.50412917137146,
0.5040262341499329,
0.5043658018112183,
0.4994247257709503,
0.501347541809082,
0.49773117899894714,
0.5059964060783386,
0.5030013918876648,
0.5042794346809387,
0.5028694868087769,
0.5025186538696289,
0.4986320734024048,
0.5045930743217468,
0.5049675703048706,
0.5096680521965027,
0.506639301776886,
0.5039033889770508,
0.5147918462753296,
0.50412917137146,
0.5040262341499329,
0.5043658018112183,
0.556916356086731,
0.5608226656913757,
0.5657909512519836,
0.5008453130722046,
0.5007270574569702,
0.5028136968612671,
0.49929720163345337,
0.5003742575645447,
0.49866074323654175,
0.5036454200744629,
0.502832293510437,
0.5069600343704224,
0.5036818385124207,
0.5027140974998474,
0.5057021975517273,
0.5055396556854248,
0.504094123840332,
0.5109544992446899,
0.5005071759223938,
0.5017557144165039,
0.501285195350647,
0.508521318435669,
0.5081111192703247,
0.5054713487625122,
0.49742183089256287,
0.5211421847343445,
0.5185605883598328,
0.5009552240371704,
0.501574695110321,
0.5025330781936646,
0.49060285091400146,
0.5004562139511108,
0.4919905960559845,
0.5075841546058655,
0.5064202547073364,
0.5060182809829712,
0.5026406645774841,
0.5015113949775696,
0.5088631510734558,
0.5031351447105408,
0.5040185451507568,
0.5019664764404297,
0.5156256556510925,
0.5126334428787231,
0.5120273232460022,
0.4980892241001129,
0.49992886185646057,
0.5005248188972473,
0.5069718956947327,
0.49112391471862793,
0.4985952079296112,
0.5040104389190674,
0.5054659247398376,
0.5010671615600586,
0.5042793154716492,
0.5050150156021118,
0.5114870071411133,
0.5076169967651367,
0.5100739002227783,
0.49788379669189453,
0.5024698376655579,
0.507462739944458,
0.5053428411483765,
0.50066739320755,
0.4964747726917267,
0.5025928020477295,
0.5061990022659302,
0.5087488293647766,
0.508694052696228,
0.5072746276855469,
0.5009099841117859,
0.5201862454414368,
0.4998215138912201,
0.5016410946846008,
0.5011292099952698,
0.5000473856925964,
0.5000450015068054,
0.4974994361400604,
0.5010585784912109,
0.4999634623527527,
0.4991486966609955,
0.5006439685821533,
0.5032927989959717,
0.4989236295223236,
0.5096270442008972,
0.5100836157798767,
0.5091648101806641,
0.5006869435310364,
0.49976566433906555,
0.5026862025260925,
0.50831538438797,
0.5009811520576477,
0.5006811022758484,
0.49900564551353455,
0.5004085302352905,
0.4983275830745697,
0.5017317533493042,
0.5020809769630432,
0.5010595917701721,
0.4964715838432312,
0.4961889386177063,
0.49632734060287476,
0.5011086463928223,
0.49857857823371887,
0.5005841851234436,
0.5008912682533264,
0.4986928403377533,
0.4986279606819153,
0.5039089322090149,
0.5034520626068115,
0.5080598592758179,
0.5031299591064453,
0.5039377808570862,
0.5049103498458862,
0.5024979710578918,
0.5003164410591125,
0.5035272240638733,
0.5025339126586914,
0.5010934472084045,
0.5057432651519775,
0.5307008624076843,
0.5266628265380859,
0.5086768269538879,
0.5123919248580933,
0.5131749510765076,
0.5111503005027771,
0.5139409899711609,
0.49350544810295105,
0.5119474530220032,
0.5018001198768616,
0.5002288818359375,
0.5024695992469788,
0.5023131370544434,
0.5040162801742554,
0.5029211640357971,
0.5024372339248657,
0.5075055956840515,
0.5216420292854309,
0.5038804411888123,
0.5025769472122192,
0.5009835362434387,
0.5036330819129944,
0.49927082657814026,
0.5021615028381348,
0.5016504526138306,
0.5022480487823486,
0.5018357038497925,
0.5366765260696411,
0.5340359210968018,
0.5246351957321167,
0.5013508200645447,
0.5017462372779846,
0.5026276111602783,
0.5099393129348755,
0.5110629796981812,
0.5080742239952087,
0.5085129737854004,
0.506452202796936,
0.49594584107398987,
0.5015504956245422,
0.49982646107673645,
0.5018503665924072,
0.5028117895126343,
0.5044481754302979,
0.5015954971313477,
0.4943663775920868,
0.49516987800598145,
0.49699723720550537,
0.5068739056587219,
0.5055515170097351,
0.5058976411819458,
0.49529212713241577,
0.4983977675437927,
0.4972948133945465,
0.5120269060134888,
0.5135841369628906,
0.5127730369567871,
0.5033084750175476,
0.5000221133232117,
0.5011393427848816,
0.5100398063659668,
0.5002798438072205,
0.5092903971672058,
0.5021178722381592,
0.49759310483932495,
0.49418240785598755,
0.5005298256874084,
0.5024158358573914,
0.5046119689941406,
0.5635921955108643,
0.5599014163017273,
0.5627495646476746,
0.49935266375541687,
0.5011667609214783,
0.4990352392196655,
0.5008418560028076,
0.5043246150016785,
0.503614604473114,
0.5217894315719604,
0.5184983611106873,
0.5060057044029236,
0.5029212832450867,
0.5000807642936707,
0.5036040544509888,
0.5109674334526062,
0.5211197137832642,
0.5184422135353088,
0.504578709602356,
0.5032348036766052,
0.5004482269287109,
0.5014628767967224,
0.5013464093208313,
0.5013779401779175,
0.5067982077598572,
0.4939952492713928,
0.4947287440299988,
0.5016181468963623,
0.502352237701416,
0.5025747418403625,
0.4994913637638092,
0.50017249584198,
0.5020331144332886,
0.5027275681495667,
0.5022361874580383,
0.5003791451454163,
0.5038732290267944,
0.5005664825439453,
0.5058587193489075,
0.4990914762020111,
0.502877414226532,
0.49976709485054016,
0.508381724357605,
0.5045766830444336,
0.5066174864768982,
0.5120567679405212,
0.4921302795410156,
0.5081300735473633,
0.502906322479248,
0.5010356903076172,
0.5005170702934265,
0.5012657642364502,
0.49921393394470215,
0.5008653402328491,
0.5025622248649597,
0.5017041563987732,
0.5028153657913208,
0.5120039582252502,
0.5003468990325928,
0.5012037754058838,
0.4985547959804535,
0.4994352459907532,
0.49768388271331787,
0.5008507966995239,
0.5018001794815063,
0.5011110901832581,
0.5025154948234558,
0.49816668033599854,
0.5120593905448914,
0.5050079822540283,
0.5071364045143127,
0.5166522860527039,
0.508701503276825,
0.5033656358718872,
0.4997156858444214,
0.5001451969146729,
0.5006592273712158,
0.4992031455039978,
0.5134631991386414,
0.5119453072547913,
0.5084452629089355,
0.4985675811767578,
0.5018190145492554,
0.502427339553833,
0.5044788122177124,
0.5108299851417542,
0.49883392453193665,
0.5156725645065308,
0.5157614350318909,
0.501599907875061,
0.5081993937492371,
0.5102306008338928,
0.5072028636932373,
0.5010624527931213,
0.502204954624176,
0.501350462436676,
0.4996547996997833,
0.49885162711143494,
0.49853622913360596,
0.5090863108634949,
0.5064405202865601,
0.5053828358650208,
0.5018274784088135,
0.503997802734375,
0.5000666975975037,
0.5087364315986633,
0.5081698894500732,
0.5100430250167847,
0.5004218220710754,
0.49962738156318665,
0.5011630058288574,
0.4930899739265442,
0.4962981641292572,
0.5000101327896118,
0.5004029273986816,
0.5006693601608276,
0.5056471228599548,
0.5016284584999084,
0.5019042491912842,
0.49868953227996826,
0.5182417631149292,
0.488452672958374,
0.5053372979164124,
0.49976250529289246,
0.5034599900245667,
0.5001394152641296,
0.503445565700531,
0.5012087225914001,
0.5036786794662476,
0.5106076002120972,
0.5011628270149231,
0.5118148922920227,
0.500609278678894,
0.5049245357513428,
0.5035067200660706,
0.4991876184940338,
0.5006078481674194,
0.50056391954422,
0.5012301802635193,
0.49951300024986267,
0.49941837787628174,
0.4998908042907715,
0.5076621770858765,
0.4994761347770691,
0.5018563866615295,
0.5026409029960632,
0.5046707391738892,
0.5015930533409119,
0.5030150413513184,
0.4993399679660797,
0.49874985218048096,
0.5031371712684631,
0.5119152069091797,
0.5018585324287415,
0.5008456110954285,
0.5009492039680481,
0.5003798007965088,
0.4993538558483124,
0.5001639127731323,
0.5005961656570435,
0.5028338432312012,
0.5120285153388977,
0.4990095794200897,
0.4980299472808838,
0.4991038739681244,
0.5017438530921936,
0.5013368725776672,
0.49919793009757996,
0.5025830268859863,
0.5035937428474426,
0.502143144607544,
0.5020499229431152,
0.49942633509635925,
0.49926403164863586,
0.5116376876831055,
0.5109353065490723,
0.510039210319519,
0.5017704367637634,
0.5017343163490295,
0.5017514228820801,
0.49962955713272095,
0.507748007774353,
0.4985766112804413,
0.5101813673973083,
0.5092180967330933,
0.5097126364707947,
0.49907928705215454,
0.49999621510505676,
0.4995257258415222,
0.5025539994239807,
0.5058831572532654,
0.5048204660415649,
0.5046526789665222,
0.49299585819244385,
0.5035987496376038,
0.5078161358833313,
0.4979941248893738,
0.5051422715187073,
0.49892839789390564,
0.5043526887893677,
0.4983918070793152,
0.498958945274353,
0.49926072359085083,
0.4999919831752777,
0.5100300908088684,
0.5110296010971069,
0.501091480255127,
0.5019621849060059,
0.5007315278053284,
0.5025948286056519,
0.5451968908309937,
0.5464586019515991,
0.5400429964065552,
0.5078227519989014,
0.5115640759468079,
0.5072938203811646,
0.5022469162940979,
0.5013755559921265,
0.5057317018508911,
0.5013558268547058,
0.5082599520683289,
0.5080757737159729,
0.5034798979759216,
0.5003739595413208,
0.5031930208206177,
0.511660099029541,
0.4991566240787506,
0.5081411600112915,
0.6247893571853638,
0.6254676580429077,
0.6201671957969666,
0.502748966217041,
0.5015339851379395,
0.5015645027160645,
0.5006077885627747,
0.5017576813697815,
0.5018741488456726,
0.5019626617431641,
0.5015231966972351,
0.5000098347663879,
0.4975742697715759,
0.49772053956985474,
0.5011225938796997,
0.5094168782234192,
0.5019370913505554,
0.5049015879631042,
0.5092257857322693,
0.5077589750289917,
0.5074192881584167,
0.49313977360725403,
0.5125611424446106,
0.5058010220527649,
0.5014447569847107,
0.5024464130401611,
0.5021930932998657,
0.5130700469017029,
0.49830132722854614,
0.5115199685096741,
0.4993806481361389,
0.5005051493644714,
0.4990237355232239
],
"eval_runtime": 43.8448,
"eval_samples_per_second": 17.106,
"eval_steps_per_second": 0.547
}