llama-13b-code_easy / training_logs.json
joshuaclymer's picture
Upload folder using huggingface_hub
35936b6
[
{
"loss": 0.7696,
"learning_rate": 0.0002,
"epoch": 0.02,
"step": 1
},
{
"loss": 0.6496,
"learning_rate": 0.0002,
"epoch": 0.04,
"step": 2
},
{
"loss": 0.6886,
"learning_rate": 0.0002,
"epoch": 0.05,
"step": 3
},
{
"loss": 0.6262,
"learning_rate": 0.0002,
"epoch": 0.07,
"step": 4
},
{
"loss": 0.5216,
"learning_rate": 0.0002,
"epoch": 0.09,
"step": 5
},
{
"loss": 0.5246,
"learning_rate": 0.0002,
"epoch": 0.11,
"step": 6
},
{
"loss": 0.4545,
"learning_rate": 0.0002,
"epoch": 0.12,
"step": 7
},
{
"loss": 0.4518,
"learning_rate": 0.0002,
"epoch": 0.14,
"step": 8
},
{
"loss": 0.3915,
"learning_rate": 0.0002,
"epoch": 0.16,
"step": 9
},
{
"loss": 0.4736,
"learning_rate": 0.0002,
"epoch": 0.18,
"step": 10
},
{
"loss": 0.4215,
"learning_rate": 0.0002,
"epoch": 0.19,
"step": 11
},
{
"loss": 0.3712,
"learning_rate": 0.0002,
"epoch": 0.21,
"step": 12
},
{
"loss": 0.3054,
"learning_rate": 0.0002,
"epoch": 0.23,
"step": 13
},
{
"loss": 0.2136,
"learning_rate": 0.0002,
"epoch": 0.25,
"step": 14
},
{
"loss": 0.1414,
"learning_rate": 0.0002,
"epoch": 0.26,
"step": 15
},
{
"loss": 0.215,
"learning_rate": 0.0002,
"epoch": 0.28,
"step": 16
},
{
"loss": 0.3062,
"learning_rate": 0.0002,
"epoch": 0.3,
"step": 17
},
{
"loss": 0.3171,
"learning_rate": 0.0002,
"epoch": 0.32,
"step": 18
},
{
"loss": 0.2465,
"learning_rate": 0.0002,
"epoch": 0.33,
"step": 19
},
{
"loss": 0.2049,
"learning_rate": 0.0002,
"epoch": 0.35,
"step": 20
},
{
"loss": 0.2947,
"learning_rate": 0.0002,
"epoch": 0.37,
"step": 21
},
{
"loss": 0.1174,
"learning_rate": 0.0002,
"epoch": 0.39,
"step": 22
},
{
"loss": 0.118,
"learning_rate": 0.0002,
"epoch": 0.4,
"step": 23
},
{
"loss": 0.1616,
"learning_rate": 0.0002,
"epoch": 0.42,
"step": 24
},
{
"loss": 0.1731,
"learning_rate": 0.0002,
"epoch": 0.44,
"step": 25
},
{
"eval_code_easy_loss": 0.1410333812236786,
"eval_code_easy_score": -0.04131929576396942,
"eval_code_easy_brier_score": 0.04131929576396942,
"eval_code_easy_average_probability": 0.8963662981987,
"eval_code_easy_accuracy": 0.94,
"eval_code_easy_probabilities": [
0.9897421598434448,
0.9846920371055603,
0.9921024441719055,
0.9999797344207764,
0.9999569654464722,
0.9998584985733032,
0.4596724212169647,
0.9999803304672241,
0.9999934434890747,
0.6896206736564636,
0.7018794417381287,
0.8044198751449585,
0.9061204791069031,
0.9999116659164429,
0.9998483657836914,
0.9975529313087463,
0.9999861717224121,
0.9999945163726807,
0.9997785687446594,
0.9217097163200378,
0.549553632736206,
0.9999867677688599,
0.9999836683273315,
0.9999502897262573,
0.9999196529388428,
0.999788224697113,
0.9994627833366394,
0.8858318328857422,
0.8768848180770874,
0.9063717126846313,
0.6450658440589905,
0.718223512172699,
0.6423218846321106,
0.6255131363868713,
0.4555876851081848,
0.5870699286460876,
0.7827481031417847,
0.9729915857315063,
0.7549998164176941,
0.9984861016273499,
0.9989733695983887,
0.9916539192199707,
0.9056527018547058,
0.8946619629859924,
0.8330284357070923,
0.9995453953742981,
0.5416378378868103,
0.9464252591133118,
0.972342848777771,
0.9960533380508423,
0.988210916519165,
0.4754994809627533,
0.9999995231628418,
0.9999991655349731,
0.9999830722808838,
0.9999951124191284,
0.9999954700469971,
0.9999794960021973,
0.3308394253253937,
0.9999799728393555,
0.9992032647132874,
0.9999996423721313,
0.9999727010726929,
0.9999815225601196,
0.9999794960021973,
0.9999793767929077,
0.988341748714447,
0.99339359998703,
0.9851863980293274,
0.631180465221405,
0.2889827787876129,
0.2886982560157776,
0.9839469194412231,
0.9860828518867493,
0.9651493430137634,
0.7786540389060974,
0.8290436863899231,
0.9590767621994019,
0.9999992847442627,
0.9999868869781494,
0.9999861717224121,
0.9500917196273804,
0.9666972160339355,
0.8169301748275757,
0.9984229803085327,
0.9998488426208496,
0.9996005892753601,
0.9686463475227356,
0.9352719783782959,
0.9167892336845398,
0.9999650716781616,
0.8374965190887451,
0.9999854564666748,
0.9999997615814209,
0.9999997615814209,
0.9999992847442627,
0.9999997615814209,
0.9999991655349731,
0.9999997615814209,
0.848063588142395
],
"eval_code_easy_runtime": 31.1515,
"eval_code_easy_samples_per_second": 3.21,
"eval_code_easy_steps_per_second": 0.128,
"epoch": 0.44,
"step": 25
},
{
"eval_code_hard_loss": 0.44707271456718445,
"eval_code_hard_score": -0.14680194854736328,
"eval_code_hard_brier_score": 0.14680194854736328,
"eval_code_hard_average_probability": 0.6711300015449524,
"eval_code_hard_accuracy": 0.83,
"eval_code_hard_probabilities": [
0.9999589920043945,
0.47546130418777466,
0.5713449716567993,
0.9994718432426453,
0.9991668462753296,
0.9993312358856201,
0.4067853093147278,
0.40341079235076904,
0.41259637475013733,
0.4262898862361908,
0.48981818556785583,
0.6610403060913086,
0.9994863271713257,
0.9982219338417053,
0.99834144115448,
0.7558809518814087,
0.6139836311340332,
0.6215753555297852,
0.9995869994163513,
0.9992690682411194,
0.9992164373397827,
0.8080712556838989,
0.8023390769958496,
0.7505528330802917,
0.5108107328414917,
0.5139395594596863,
0.4423873722553253,
0.7270267605781555,
0.7878870368003845,
0.7860028743743896,
0.6687493920326233,
0.7331302165985107,
0.6188584566116333,
0.5269416570663452,
0.4910818338394165,
0.4837784469127655,
0.5938860774040222,
0.5307540893554688,
0.6091882586479187,
0.5714852213859558,
0.5575187802314758,
0.636616587638855,
0.5653713941574097,
0.5363364219665527,
0.5516955256462097,
0.7773237228393555,
0.5234290361404419,
0.5288711786270142,
0.6337161064147949,
0.6571640968322754,
0.6706582903862,
0.45844873785972595,
0.7167773842811584,
0.605722188949585,
0.9969874024391174,
0.9969995617866516,
0.9975918531417847,
0.3677201271057129,
0.5539365410804749,
0.7323344349861145,
0.9991826415061951,
0.9991852641105652,
0.9992254972457886,
0.9554563760757446,
0.9320324659347534,
0.9908183217048645,
0.5780607461929321,
0.9173429608345032,
0.573617160320282,
0.710330605506897,
0.610026478767395,
0.5195437073707581,
0.8946442008018494,
0.925991415977478,
0.938264787197113,
0.5608754754066467,
0.5026195049285889,
0.4715512990951538,
0.5032576322555542,
0.5153825879096985,
0.5001385807991028,
0.6857117414474487,
0.6296961307525635,
0.58381187915802,
0.5933532118797302,
0.5141075253486633,
0.5213065147399902,
0.4824189841747284,
0.5641835331916809,
0.5198292136192322,
0.5848154425621033,
0.6065869927406311,
0.5906622409820557,
0.39757779240608215,
0.37057143449783325,
0.33965691924095154,
0.638043999671936,
0.75816410779953,
0.8119453191757202,
0.4706781804561615
],
"eval_code_hard_runtime": 110.7399,
"eval_code_hard_samples_per_second": 0.903,
"eval_code_hard_steps_per_second": 0.036,
"epoch": 0.44,
"step": 25
},
{
"loss": 0.1187,
"learning_rate": 0.0002,
"epoch": 0.46,
"step": 26
},
{
"loss": 0.1277,
"learning_rate": 0.0002,
"epoch": 0.47,
"step": 27
},
{
"loss": 0.0581,
"learning_rate": 0.0002,
"epoch": 0.49,
"step": 28
},
{
"loss": 0.1622,
"learning_rate": 0.0002,
"epoch": 0.51,
"step": 29
},
{
"loss": 0.0558,
"learning_rate": 0.0002,
"epoch": 0.53,
"step": 30
},
{
"loss": 0.0873,
"learning_rate": 0.0002,
"epoch": 0.54,
"step": 31
},
{
"loss": 0.1237,
"learning_rate": 0.0002,
"epoch": 0.56,
"step": 32
},
{
"loss": 0.1127,
"learning_rate": 0.0002,
"epoch": 0.58,
"step": 33
},
{
"loss": 0.0673,
"learning_rate": 0.0002,
"epoch": 0.6,
"step": 34
},
{
"loss": 0.0628,
"learning_rate": 0.0002,
"epoch": 0.61,
"step": 35
},
{
"loss": 0.1647,
"learning_rate": 0.0002,
"epoch": 0.63,
"step": 36
},
{
"loss": 0.0389,
"learning_rate": 0.0002,
"epoch": 0.65,
"step": 37
},
{
"loss": 0.0162,
"learning_rate": 0.0002,
"epoch": 0.67,
"step": 38
},
{
"loss": 0.1205,
"learning_rate": 0.0002,
"epoch": 0.68,
"step": 39
},
{
"loss": 0.2798,
"learning_rate": 0.0002,
"epoch": 0.7,
"step": 40
},
{
"loss": 0.0202,
"learning_rate": 0.0002,
"epoch": 0.72,
"step": 41
},
{
"loss": 0.0563,
"learning_rate": 0.0002,
"epoch": 0.74,
"step": 42
},
{
"loss": 0.0959,
"learning_rate": 0.0002,
"epoch": 0.75,
"step": 43
},
{
"loss": 0.033,
"learning_rate": 0.0002,
"epoch": 0.77,
"step": 44
},
{
"loss": 0.0377,
"learning_rate": 0.0002,
"epoch": 0.79,
"step": 45
},
{
"loss": 0.1312,
"learning_rate": 0.0002,
"epoch": 0.81,
"step": 46
},
{
"loss": 0.1171,
"learning_rate": 0.0002,
"epoch": 0.82,
"step": 47
},
{
"loss": 0.0586,
"learning_rate": 0.0002,
"epoch": 0.84,
"step": 48
},
{
"loss": 0.0373,
"learning_rate": 0.0002,
"epoch": 0.86,
"step": 49
},
{
"loss": 0.1418,
"learning_rate": 0.0002,
"epoch": 0.88,
"step": 50
},
{
"eval_code_easy_loss": 0.09489491581916809,
"eval_code_easy_score": -0.030822403728961945,
"eval_code_easy_brier_score": 0.030822403728961945,
"eval_code_easy_average_probability": 0.9424028992652893,
"eval_code_easy_accuracy": 0.95,
"eval_code_easy_probabilities": [
0.9999914169311523,
1.0,
1.0,
0.9999921321868896,
0.9999855756759644,
0.9999556541442871,
0.4764171540737152,
1.0,
1.0,
0.9999983310699463,
0.9999955892562866,
0.9999998807907104,
0.9997754693031311,
0.9999978542327881,
0.9999951124191284,
0.9944272041320801,
1.0,
1.0,
1.0,
0.9332470893859863,
0.686340868473053,
0.9999998807907104,
0.9999998807907104,
0.9999996423721313,
1.0,
0.9999998807907104,
0.9999986886978149,
0.9998589754104614,
0.9998730421066284,
0.9999005794525146,
0.7265980839729309,
0.8790691494941711,
0.8301546573638916,
0.9983548521995544,
0.9948050379753113,
0.9825286865234375,
0.962716817855835,
0.9991486072540283,
0.9315124750137329,
0.9939038157463074,
0.9942259192466736,
0.9889645576477051,
0.9977827668190002,
0.9999998807907104,
0.9890486001968384,
0.9997009038925171,
0.20336638391017914,
0.9882842302322388,
0.9745821952819824,
0.9941431879997253,
0.9512994289398193,
0.7942199110984802,
1.0,
1.0,
1.0,
1.0,
1.0,
0.999992847442627,
0.254838228225708,
0.9999922513961792,
1.0,
1.0,
0.9999960660934448,
1.0,
1.0,
1.0,
1.0,
1.0,
1.0,
0.9282342791557312,
0.25453829765319824,
0.1509198695421219,
0.9998898506164551,
0.9999701976776123,
0.999139666557312,
0.8278753161430359,
0.8814498782157898,
0.9445298910140991,
1.0,
0.9999996423721313,
1.0,
0.9991635084152222,
1.0,
0.9999971389770508,
0.955471932888031,
0.9967992305755615,
0.9889751672744751,
0.9923612475395203,
0.9318004250526428,
0.9118591547012329,
1.0,
0.95877605676651,
0.9997400641441345,
1.0,
1.0,
1.0,
1.0,
1.0,
1.0,
0.9998133778572083
],
"eval_code_easy_runtime": 31.1754,
"eval_code_easy_samples_per_second": 3.208,
"eval_code_easy_steps_per_second": 0.128,
"epoch": 0.88,
"step": 50
},
{
"eval_code_hard_loss": 0.48955029249191284,
"eval_code_hard_score": -0.16734105348587036,
"eval_code_hard_brier_score": 0.16734105348587036,
"eval_code_hard_average_probability": 0.6507197022438049,
"eval_code_hard_accuracy": 0.74,
"eval_code_hard_probabilities": [
0.998719334602356,
0.4495692551136017,
0.560586154460907,
0.9998689889907837,
0.9994572997093201,
0.999840259552002,
0.40500903129577637,
0.4450298845767975,
0.39399078488349915,
0.48165562748908997,
0.4001981019973755,
0.6641137599945068,
0.999977707862854,
0.9998791217803955,
0.9998584985733032,
0.8799082636833191,
0.766524612903595,
0.775595486164093,
0.9999852180480957,
0.9999836683273315,
0.9999798536300659,
0.8477662205696106,
0.7681900262832642,
0.7254011631011963,
0.4458688199520111,
0.4710305333137512,
0.3777993619441986,
0.09769346565008163,
0.4915977418422699,
0.5548089146614075,
0.6629347801208496,
0.7435128092765808,
0.5462533831596375,
0.6541544198989868,
0.4705035388469696,
0.5168669819831848,
0.42897677421569824,
0.3096539080142975,
0.4461301267147064,
0.461626797914505,
0.46588248014450073,
0.5764793157577515,
0.5880526304244995,
0.5188709497451782,
0.5498936772346497,
0.8593714237213135,
0.5482467412948608,
0.5750262141227722,
0.67604660987854,
0.8179431557655334,
0.8246544003486633,
0.4119476079940796,
0.7206862568855286,
0.6743555068969727,
0.9999415874481201,
0.9999454021453857,
0.9999594688415527,
0.470983624458313,
0.5161882042884827,
0.5438310503959656,
0.9999815225601196,
0.9999749660491943,
0.9999841451644897,
0.9674224257469177,
0.9396224617958069,
0.9942786693572998,
0.6115734577178955,
0.9555063247680664,
0.6700289249420166,
0.8057522177696228,
0.6384589672088623,
0.5619309544563293,
0.6012542843818665,
0.6818827986717224,
0.709420382976532,
0.6579272747039795,
0.42321380972862244,
0.44099465012550354,
0.5047284364700317,
0.506719172000885,
0.5062504410743713,
0.6765443682670593,
0.7465528249740601,
0.6119047999382019,
0.5429056882858276,
0.5111532807350159,
0.5241301655769348,
0.4921115040779114,
0.6113234162330627,
0.5521866679191589,
0.5868935585021973,
0.6085503697395325,
0.5654861927032471,
0.3610905706882477,
0.3744204640388489,
0.31809136271476746,
0.5574522018432617,
0.641338586807251,
0.48631271719932556,
0.5477950572967529
],
"eval_code_hard_runtime": 110.6958,
"eval_code_hard_samples_per_second": 0.903,
"eval_code_hard_steps_per_second": 0.036,
"epoch": 0.88,
"step": 50
},
{
"loss": 0.0819,
"learning_rate": 0.0002,
"epoch": 0.89,
"step": 51
},
{
"loss": 0.0037,
"learning_rate": 0.0002,
"epoch": 0.91,
"step": 52
},
{
"loss": 0.1525,
"learning_rate": 0.0002,
"epoch": 0.93,
"step": 53
},
{
"loss": 0.2635,
"learning_rate": 0.0002,
"epoch": 0.95,
"step": 54
},
{
"loss": 0.1162,
"learning_rate": 0.0002,
"epoch": 0.96,
"step": 55
},
{
"loss": 0.0352,
"learning_rate": 0.0002,
"epoch": 0.98,
"step": 56
},
{
"loss": 0.1891,
"learning_rate": 0.0002,
"epoch": 1.0,
"step": 57
},
{
"loss": 0.0453,
"learning_rate": 0.0002,
"epoch": 1.02,
"step": 58
},
{
"loss": 0.0562,
"learning_rate": 0.0002,
"epoch": 1.04,
"step": 59
},
{
"loss": 0.0006,
"learning_rate": 0.0002,
"epoch": 1.05,
"step": 60
},
{
"loss": 0.0445,
"learning_rate": 0.0002,
"epoch": 1.07,
"step": 61
},
{
"loss": 0.0125,
"learning_rate": 0.0002,
"epoch": 1.09,
"step": 62
},
{
"loss": 0.0378,
"learning_rate": 0.0002,
"epoch": 1.11,
"step": 63
},
{
"loss": 0.0367,
"learning_rate": 0.0002,
"epoch": 1.12,
"step": 64
},
{
"loss": 0.0261,
"learning_rate": 0.0002,
"epoch": 1.14,
"step": 65
},
{
"loss": 0.0022,
"learning_rate": 0.0002,
"epoch": 1.16,
"step": 66
},
{
"loss": 0.0151,
"learning_rate": 0.0002,
"epoch": 1.18,
"step": 67
},
{
"loss": 0.0262,
"learning_rate": 0.0002,
"epoch": 1.19,
"step": 68
},
{
"loss": 0.0025,
"learning_rate": 0.0002,
"epoch": 1.21,
"step": 69
},
{
"loss": 0.009,
"learning_rate": 0.0002,
"epoch": 1.23,
"step": 70
},
{
"loss": 0.0233,
"learning_rate": 0.0002,
"epoch": 1.25,
"step": 71
},
{
"loss": 0.0167,
"learning_rate": 0.0002,
"epoch": 1.26,
"step": 72
},
{
"loss": 0.0088,
"learning_rate": 0.0002,
"epoch": 1.28,
"step": 73
},
{
"loss": 0.0003,
"learning_rate": 0.0002,
"epoch": 1.3,
"step": 74
},
{
"loss": 0.0111,
"learning_rate": 0.0002,
"epoch": 1.32,
"step": 75
},
{
"eval_code_easy_loss": 0.08589933067560196,
"eval_code_easy_score": -0.029182305559515953,
"eval_code_easy_brier_score": 0.029182305559515953,
"eval_code_easy_average_probability": 0.9475240111351013,
"eval_code_easy_accuracy": 0.95,
"eval_code_easy_probabilities": [
1.0,
1.0,
1.0,
1.0,
1.0,
0.9999991655349731,
0.44722074270248413,
1.0,
1.0,
0.9999874830245972,
0.9999817609786987,
0.9999996423721313,
0.9983553290367126,
1.0,
0.9999998807907104,
0.969874918460846,
1.0,
1.0,
1.0,
0.9999077320098877,
0.7078168392181396,
1.0,
1.0,
1.0,
0.9998465776443481,
0.9997827410697937,
0.9978830218315125,
0.9988357424736023,
0.9989570379257202,
0.9992606043815613,
0.7118145227432251,
0.8777998089790344,
0.7065884470939636,
0.9980650544166565,
0.9990893602371216,
0.9781619310379028,
0.9999847412109375,
1.0,
0.9993658661842346,
0.9999990463256836,
0.9999980926513672,
0.9999912977218628,
0.9999923706054688,
1.0,
0.9999966621398926,
0.9973452687263489,
0.13087554275989532,
0.8122653365135193,
0.9999538660049438,
0.9999990463256836,
0.9998875856399536,
0.9902234673500061,
1.0,
1.0,
0.9999982118606567,
0.9999996423721313,
0.9999994039535522,
0.9999998807907104,
0.396033376455307,
0.9999998807907104,
0.9999998807907104,
1.0,
0.9975220561027527,
1.0,
1.0,
1.0,
0.9999996423721313,
1.0,
0.9999997615814209,
0.9743689298629761,
0.3422013521194458,
0.25521236658096313,
0.9973738193511963,
0.9997031092643738,
0.9819728136062622,
0.9939416646957397,
0.997735857963562,
0.9992177486419678,
1.0,
0.9999991655349731,
0.9999998807907104,
0.9803919196128845,
1.0,
0.9999996423721313,
0.9999098777770996,
0.9999998807907104,
0.9999986886978149,
0.9992156028747559,
0.9822015166282654,
0.9914659261703491,
0.9999808073043823,
0.5561801791191101,
0.9888289570808411,
1.0,
1.0,
1.0,
1.0,
1.0,
1.0,
0.9998490810394287
],
"eval_code_easy_runtime": 31.1417,
"eval_code_easy_samples_per_second": 3.211,
"eval_code_easy_steps_per_second": 0.128,
"epoch": 1.32,
"step": 75
},
{
"eval_code_hard_loss": 0.4541158378124237,
"eval_code_hard_score": -0.15187427401542664,
"eval_code_hard_brier_score": 0.15187427401542664,
"eval_code_hard_average_probability": 0.7136721014976501,
"eval_code_hard_accuracy": 0.79,
"eval_code_hard_probabilities": [
0.9999998807907104,
0.42837128043174744,
0.6079556345939636,
0.9998014569282532,
0.9960059523582458,
0.9952560067176819,
0.25668519735336304,
0.37132731080055237,
0.18430009484291077,
0.5929830074310303,
0.3093147277832031,
0.7880420088768005,
0.999980092048645,
0.9994723200798035,
0.9992823004722595,
0.9913004040718079,
0.9134578704833984,
0.9550933837890625,
0.9999967813491821,
0.9999939203262329,
0.9999887943267822,
0.9903974533081055,
0.9929130673408508,
0.9700443744659424,
0.5675995945930481,
0.4452306032180786,
0.39090991020202637,
0.05753504857420921,
0.5650666952133179,
0.6610203981399536,
0.8633843660354614,
0.9552722573280334,
0.6663053035736084,
0.7623808979988098,
0.4418239891529083,
0.5061418414115906,
0.2837110459804535,
0.19440007209777832,
0.4149368107318878,
0.6865063905715942,
0.653810441493988,
0.8122990131378174,
0.7234838604927063,
0.5346352458000183,
0.6917294263839722,
0.9355289340019226,
0.5552300810813904,
0.6327811479568481,
0.8747856020927429,
0.962817907333374,
0.9655711650848389,
0.18094369769096375,
0.8358926177024841,
0.876140296459198,
0.9999440908432007,
0.9999480247497559,
0.9999721050262451,
0.2512325644493103,
0.5562063455581665,
0.623430609703064,
0.9999799728393555,
0.9999500513076782,
0.9999836683273315,
0.9980340600013733,
0.9961856007575989,
0.9999278783798218,
0.9568101167678833,
0.9999703168869019,
0.9811659455299377,
0.9707834720611572,
0.8236635327339172,
0.4347587823867798,
0.8745930790901184,
0.9302994608879089,
0.9327366352081299,
0.766298234462738,
0.3869512379169464,
0.4064312279224396,
0.5269957184791565,
0.5092197060585022,
0.5739120841026306,
0.9161906838417053,
0.9152584075927734,
0.8292853832244873,
0.6077045798301697,
0.4789629280567169,
0.55214923620224,
0.5199791193008423,
0.834286093711853,
0.7272135615348816,
0.5919612050056458,
0.628190279006958,
0.5904282331466675,
0.20717588067054749,
0.23157484829425812,
0.17664135992527008,
0.7453339099884033,
0.9322812557220459,
0.7623251676559448,
0.5810563564300537
],
"eval_code_hard_runtime": 110.6654,
"eval_code_hard_samples_per_second": 0.904,
"eval_code_hard_steps_per_second": 0.036,
"epoch": 1.32,
"step": 75
},
{
"loss": 0.0328,
"learning_rate": 0.0002,
"epoch": 1.33,
"step": 76
},
{
"loss": 0.0814,
"learning_rate": 0.0002,
"epoch": 1.35,
"step": 77
},
{
"loss": 0.0022,
"learning_rate": 0.0002,
"epoch": 1.37,
"step": 78
},
{
"loss": 0.0048,
"learning_rate": 0.0002,
"epoch": 1.39,
"step": 79
},
{
"loss": 0.0439,
"learning_rate": 0.0002,
"epoch": 1.4,
"step": 80
},
{
"loss": 0.0316,
"learning_rate": 0.0002,
"epoch": 1.42,
"step": 81
},
{
"loss": 0.0956,
"learning_rate": 0.0002,
"epoch": 1.44,
"step": 82
},
{
"loss": 0.0138,
"learning_rate": 0.0002,
"epoch": 1.46,
"step": 83
},
{
"loss": 0.0014,
"learning_rate": 0.0002,
"epoch": 1.47,
"step": 84
},
{
"loss": 0.0353,
"learning_rate": 0.0002,
"epoch": 1.49,
"step": 85
},
{
"loss": 0.0024,
"learning_rate": 0.0002,
"epoch": 1.51,
"step": 86
},
{
"loss": 0.0206,
"learning_rate": 0.0002,
"epoch": 1.53,
"step": 87
},
{
"loss": 0.0007,
"learning_rate": 0.0002,
"epoch": 1.54,
"step": 88
},
{
"loss": 0.0472,
"learning_rate": 0.0002,
"epoch": 1.56,
"step": 89
},
{
"loss": 0.0521,
"learning_rate": 0.0002,
"epoch": 1.58,
"step": 90
},
{
"loss": 0.0073,
"learning_rate": 0.0002,
"epoch": 1.6,
"step": 91
},
{
"loss": 0.1226,
"learning_rate": 0.0002,
"epoch": 1.61,
"step": 92
},
{
"loss": 0.0452,
"learning_rate": 0.0002,
"epoch": 1.63,
"step": 93
},
{
"loss": 0.0143,
"learning_rate": 0.0002,
"epoch": 1.65,
"step": 94
},
{
"loss": 0.0593,
"learning_rate": 0.0002,
"epoch": 1.67,
"step": 95
},
{
"loss": 0.0165,
"learning_rate": 0.0002,
"epoch": 1.68,
"step": 96
},
{
"loss": 0.0016,
"learning_rate": 0.0002,
"epoch": 1.7,
"step": 97
},
{
"loss": 0.0025,
"learning_rate": 0.0002,
"epoch": 1.72,
"step": 98
},
{
"loss": 0.0032,
"learning_rate": 0.0002,
"epoch": 1.74,
"step": 99
},
{
"loss": 0.0065,
"learning_rate": 0.0002,
"epoch": 1.75,
"step": 100
},
{
"eval_code_easy_loss": 0.1302385777235031,
"eval_code_easy_score": -0.038255855441093445,
"eval_code_easy_brier_score": 0.038255855441093445,
"eval_code_easy_average_probability": 0.9474854469299316,
"eval_code_easy_accuracy": 0.96,
"eval_code_easy_probabilities": [
0.9999997615814209,
1.0,
0.9999998807907104,
1.0,
0.9999998807907104,
0.9999867677688599,
0.14685745537281036,
1.0,
1.0,
0.9999862909317017,
0.9999769926071167,
0.9999992847442627,
0.9967072606086731,
1.0,
0.9999990463256836,
0.9817181825637817,
1.0,
1.0,
1.0,
0.9999679327011108,
0.989346444606781,
1.0,
1.0,
0.9999996423721313,
1.0,
1.0,
0.9997485280036926,
0.9999996423721313,
0.9999998807907104,
0.9999997615814209,
0.7592384815216064,
0.9866325259208679,
0.7520292401313782,
0.9871376752853394,
0.9395173788070679,
0.9484317898750305,
0.99998939037323,
1.0,
0.987211287021637,
0.9999986886978149,
0.9999985694885254,
0.999853253364563,
0.9999595880508423,
0.9999998807907104,
0.9999158382415771,
0.9994738698005676,
0.07437732070684433,
0.8850747346878052,
0.9996064305305481,
0.9999996423721313,
0.9999347925186157,
0.9994233846664429,
1.0,
0.9999998807907104,
1.0,
1.0,
1.0,
1.0,
0.754472017288208,
0.9999998807907104,
0.9999997615814209,
1.0,
0.9901873469352722,
1.0,
1.0,
1.0,
1.0,
1.0,
1.0,
0.9974935054779053,
0.0717310905456543,
0.029978960752487183,
1.0,
1.0,
0.999995231628418,
0.9984604120254517,
0.9997397065162659,
0.9999804496765137,
1.0,
1.0,
1.0,
0.9670922160148621,
0.9999996423721313,
0.9999791383743286,
0.998187243938446,
0.9999996423721313,
0.9999892711639404,
0.9999653100967407,
0.9966863989830017,
0.999233603477478,
1.0,
0.513279378414154,
0.9999966621398926,
1.0,
1.0,
1.0,
1.0,
1.0,
1.0,
0.9999997615814209
],
"eval_code_easy_runtime": 31.1546,
"eval_code_easy_samples_per_second": 3.21,
"eval_code_easy_steps_per_second": 0.128,
"epoch": 1.75,
"step": 100
},
{
"eval_code_hard_loss": 0.4152863323688507,
"eval_code_hard_score": -0.13397492468357086,
"eval_code_hard_brier_score": 0.13397492468357086,
"eval_code_hard_average_probability": 0.7467895746231079,
"eval_code_hard_accuracy": 0.82,
"eval_code_hard_probabilities": [
0.999998927116394,
0.3584747314453125,
0.5524832010269165,
0.9999052286148071,
0.9991750121116638,
0.9992142915725708,
0.17623497545719147,
0.41959860920906067,
0.08018570393323898,
0.6593999862670898,
0.26273950934410095,
0.7871806621551514,
0.999994158744812,
0.9992138147354126,
0.9983713030815125,
0.9971001744270325,
0.9297299981117249,
0.9683568477630615,
0.9999955892562866,
0.9999574422836304,
0.9999104738235474,
0.9948668479919434,
0.9949619770050049,
0.9704132080078125,
0.6137145757675171,
0.5000441074371338,
0.39442917704582214,
0.11175673454999924,
0.797296941280365,
0.8252298831939697,
0.9333888292312622,
0.9808279871940613,
0.7203049659729004,
0.8538256287574768,
0.4518895745277405,
0.613561749458313,
0.5261998772621155,
0.38816678524017334,
0.7502338886260986,
0.6585968136787415,
0.6722922325134277,
0.8558183908462524,
0.8481931090354919,
0.5945469737052917,
0.8612274527549744,
0.9837346076965332,
0.5298483371734619,
0.7162337303161621,
0.9290578365325928,
0.9855068922042847,
0.9883608818054199,
0.20835253596305847,
0.9197254776954651,
0.8866751194000244,
0.9997254014015198,
0.9997286200523376,
0.9998807907104492,
0.4348519742488861,
0.5857678651809692,
0.5733372569084167,
0.9998024106025696,
0.9995396137237549,
0.9998968839645386,
0.9989103078842163,
0.9946140646934509,
0.9999970197677612,
0.8636531829833984,
0.9994328618049622,
0.9395280480384827,
0.9852699637413025,
0.8448395729064941,
0.33028489351272583,
0.9747959971427917,
0.996696949005127,
0.9959298968315125,
0.8669978976249695,
0.32030242681503296,
0.336773544549942,
0.5315243005752563,
0.4963895082473755,
0.5617468357086182,
0.9487797021865845,
0.9507840275764465,
0.8413812518119812,
0.8480523228645325,
0.5873783230781555,
0.5967154502868652,
0.5515139698982239,
0.7987790107727051,
0.7132774591445923,
0.7350783348083496,
0.7808082699775696,
0.746001124382019,
0.17566508054733276,
0.21035020053386688,
0.10222747921943665,
0.6762077212333679,
0.9854094386100769,
0.8317046761512756,
0.7221127152442932
],
"eval_code_hard_runtime": 110.7468,
"eval_code_hard_samples_per_second": 0.903,
"eval_code_hard_steps_per_second": 0.036,
"epoch": 1.75,
"step": 100
},
{
"loss": 0.0416,
"learning_rate": 0.0002,
"epoch": 1.77,
"step": 101
},
{
"loss": 0.0034,
"learning_rate": 0.0002,
"epoch": 1.79,
"step": 102
},
{
"loss": 0.1193,
"learning_rate": 0.0002,
"epoch": 1.81,
"step": 103
},
{
"loss": 0.0152,
"learning_rate": 0.0002,
"epoch": 1.82,
"step": 104
},
{
"loss": 0.0074,
"learning_rate": 0.0002,
"epoch": 1.84,
"step": 105
},
{
"loss": 0.0236,
"learning_rate": 0.0002,
"epoch": 1.86,
"step": 106
},
{
"loss": 0.0077,
"learning_rate": 0.0002,
"epoch": 1.88,
"step": 107
},
{
"loss": 0.1086,
"learning_rate": 0.0002,
"epoch": 1.89,
"step": 108
},
{
"loss": 0.0718,
"learning_rate": 0.0002,
"epoch": 1.91,
"step": 109
},
{
"loss": 0.0042,
"learning_rate": 0.0002,
"epoch": 1.93,
"step": 110
},
{
"loss": 0.0068,
"learning_rate": 0.0002,
"epoch": 1.95,
"step": 111
},
{
"loss": 0.0774,
"learning_rate": 0.0002,
"epoch": 1.96,
"step": 112
},
{
"loss": 0.0374,
"learning_rate": 0.0002,
"epoch": 1.98,
"step": 113
},
{
"loss": 0.0035,
"learning_rate": 0.0002,
"epoch": 2.0,
"step": 114
},
{
"loss": 0.0083,
"learning_rate": 0.0002,
"epoch": 2.02,
"step": 115
},
{
"loss": 0.0147,
"learning_rate": 0.0002,
"epoch": 2.04,
"step": 116
},
{
"loss": 0.0079,
"learning_rate": 0.0002,
"epoch": 2.05,
"step": 117
},
{
"loss": 0.0012,
"learning_rate": 0.0002,
"epoch": 2.07,
"step": 118
},
{
"loss": 0.0137,
"learning_rate": 0.0002,
"epoch": 2.09,
"step": 119
},
{
"loss": 0.029,
"learning_rate": 0.0002,
"epoch": 2.11,
"step": 120
},
{
"loss": 0.0125,
"learning_rate": 0.0002,
"epoch": 2.12,
"step": 121
},
{
"loss": 0.0102,
"learning_rate": 0.0002,
"epoch": 2.14,
"step": 122
},
{
"loss": 0.0057,
"learning_rate": 0.0002,
"epoch": 2.16,
"step": 123
},
{
"loss": 0.0001,
"learning_rate": 0.0002,
"epoch": 2.18,
"step": 124
},
{
"loss": 0.1814,
"learning_rate": 0.0002,
"epoch": 2.19,
"step": 125
},
{
"eval_code_easy_loss": 0.10904992371797562,
"eval_code_easy_score": -0.03530557453632355,
"eval_code_easy_brier_score": 0.03530557453632355,
"eval_code_easy_average_probability": 0.9442441463470459,
"eval_code_easy_accuracy": 0.95,
"eval_code_easy_probabilities": [
0.9999860525131226,
0.9999575614929199,
0.9999876022338867,
1.0,
1.0,
0.9999998807907104,
0.3663080930709839,
1.0,
1.0,
0.9986967444419861,
0.9995228052139282,
0.9994437098503113,
0.9989981055259705,
1.0,
1.0,
0.9903301000595093,
1.0,
1.0,
1.0,
1.0,
0.9906626343727112,
1.0,
1.0,
1.0,
0.9999998807907104,
1.0,
0.9999880790710449,
0.9878363013267517,
0.9968984127044678,
0.9852358102798462,
0.7452288866043091,
0.8594924807548523,
0.625163733959198,
0.9999972581863403,
0.9998786449432373,
0.9998183846473694,
0.9999641180038452,
0.9999997615814209,
0.986954391002655,
0.9998384714126587,
0.9996883869171143,
0.9997907280921936,
1.0,
1.0,
1.0,
0.6630300283432007,
0.3178897202014923,
0.23424135148525238,
0.9830185174942017,
0.9976402521133423,
0.9478833079338074,
1.0,
1.0,
1.0,
1.0,
1.0,
1.0,
0.9999997615814209,
0.9950571060180664,
0.9999996423721313,
0.9999940395355225,
0.9999960660934448,
0.998701810836792,
1.0,
1.0,
1.0,
1.0,
1.0,
1.0,
0.9900442361831665,
0.11046496033668518,
0.05695008859038353,
0.9999970197677612,
0.9999996423721313,
0.999971866607666,
0.9758032560348511,
0.987618088722229,
0.9728375673294067,
1.0,
0.9999998807907104,
1.0,
0.9999992847442627,
1.0,
1.0,
0.9805415868759155,
0.9995923638343811,
0.9991359114646912,
0.9988665580749512,
0.9318742156028748,
0.9750478863716125,
0.9999997615814209,
0.7831814289093018,
0.9958741068840027,
1.0,
1.0,
1.0,
1.0,
1.0,
1.0,
0.9994946718215942
],
"eval_code_easy_runtime": 31.118,
"eval_code_easy_samples_per_second": 3.214,
"eval_code_easy_steps_per_second": 0.129,
"epoch": 2.19,
"step": 125
},
{
"eval_code_hard_loss": 0.4987069070339203,
"eval_code_hard_score": -0.15576696395874023,
"eval_code_hard_brier_score": 0.15576696395874023,
"eval_code_hard_average_probability": 0.6946533918380737,
"eval_code_hard_accuracy": 0.79,
"eval_code_hard_probabilities": [
0.9995853304862976,
0.20991425216197968,
0.549527108669281,
1.0,
0.9999998807907104,
1.0,
0.22219569981098175,
0.37956300377845764,
0.20069900155067444,
0.6470227837562561,
0.31780990958213806,
0.6012275218963623,
1.0,
0.9999998807907104,
0.9999998807907104,
0.9296144247055054,
0.8540101051330566,
0.8961200714111328,
1.0,
1.0,
1.0,
0.9873550534248352,
0.9724990129470825,
0.9760501980781555,
0.38157281279563904,
0.42047223448753357,
0.3564426898956299,
0.0027432094793766737,
0.2510468661785126,
0.3818874955177307,
0.8423392176628113,
0.9236337542533875,
0.5938029289245605,
0.8878844380378723,
0.7184504866600037,
0.7954170107841492,
0.4380953311920166,
0.42004096508026123,
0.6841737627983093,
0.5444896817207336,
0.40916869044303894,
0.5416669249534607,
0.7608702182769775,
0.5757149457931519,
0.8424829840660095,
0.9364664554595947,
0.548158586025238,
0.6652321219444275,
0.8666991591453552,
0.876342236995697,
0.8881990313529968,
0.3995114266872406,
0.8002880215644836,
0.6442937254905701,
1.0,
1.0,
1.0,
0.202246755361557,
0.5754519104957581,
0.5530020594596863,
1.0,
1.0,
1.0,
0.9787330031394958,
0.9658524394035339,
0.9971476197242737,
0.7014413475990295,
0.9961981177330017,
0.8666495680809021,
0.8775137066841125,
0.5972762107849121,
0.5123103857040405,
0.6785091757774353,
0.7748872637748718,
0.8229066133499146,
0.6105999946594238,
0.33387431502342224,
0.2667674720287323,
0.520160436630249,
0.5227006673812866,
0.5882536172866821,
0.7376717329025269,
0.8348681926727295,
0.6894320249557495,
0.39057934284210205,
0.3577791750431061,
0.5273920893669128,
0.47946465015411377,
0.7557993531227112,
0.7129892110824585,
0.5154388546943665,
0.55854332447052,
0.5196318030357361,
0.7760440111160278,
0.8043990731239319,
0.7464650869369507,
0.760195791721344,
0.9593195915222168,
0.7079413533210754,
0.5481259822845459
],
"eval_code_hard_runtime": 110.6367,
"eval_code_hard_samples_per_second": 0.904,
"eval_code_hard_steps_per_second": 0.036,
"epoch": 2.19,
"step": 125
},
{
"loss": 0.0089,
"learning_rate": 0.0002,
"epoch": 2.21,
"step": 126
},
{
"loss": 0.008,
"learning_rate": 0.0002,
"epoch": 2.23,
"step": 127
},
{
"loss": 0.0025,
"learning_rate": 0.0002,
"epoch": 2.25,
"step": 128
},
{
"loss": 0.003,
"learning_rate": 0.0002,
"epoch": 2.26,
"step": 129
},
{
"loss": 0.0029,
"learning_rate": 0.0002,
"epoch": 2.28,
"step": 130
},
{
"loss": 0.012,
"learning_rate": 0.0002,
"epoch": 2.3,
"step": 131
},
{
"loss": 0.0097,
"learning_rate": 0.0002,
"epoch": 2.32,
"step": 132
},
{
"loss": 0.0024,
"learning_rate": 0.0002,
"epoch": 2.33,
"step": 133
},
{
"loss": 0.0047,
"learning_rate": 0.0002,
"epoch": 2.35,
"step": 134
},
{
"loss": 0.0007,
"learning_rate": 0.0002,
"epoch": 2.37,
"step": 135
},
{
"loss": 0.0027,
"learning_rate": 0.0002,
"epoch": 2.39,
"step": 136
},
{
"loss": 0.0,
"learning_rate": 0.0002,
"epoch": 2.4,
"step": 137
},
{
"loss": 0.0001,
"learning_rate": 0.0002,
"epoch": 2.42,
"step": 138
},
{
"loss": 0.0,
"learning_rate": 0.0002,
"epoch": 2.44,
"step": 139
},
{
"loss": 0.0016,
"learning_rate": 0.0002,
"epoch": 2.46,
"step": 140
},
{
"loss": 0.0002,
"learning_rate": 0.0002,
"epoch": 2.47,
"step": 141
},
{
"loss": 0.0696,
"learning_rate": 0.0002,
"epoch": 2.49,
"step": 142
},
{
"loss": 0.0059,
"learning_rate": 0.0002,
"epoch": 2.51,
"step": 143
},
{
"loss": 0.0005,
"learning_rate": 0.0002,
"epoch": 2.53,
"step": 144
},
{
"loss": 0.0,
"learning_rate": 0.0002,
"epoch": 2.54,
"step": 145
},
{
"loss": 0.0204,
"learning_rate": 0.0002,
"epoch": 2.56,
"step": 146
},
{
"loss": 0.0043,
"learning_rate": 0.0002,
"epoch": 2.58,
"step": 147
},
{
"loss": 0.008,
"learning_rate": 0.0002,
"epoch": 2.6,
"step": 148
},
{
"loss": 0.0462,
"learning_rate": 0.0002,
"epoch": 2.61,
"step": 149
},
{
"loss": 0.0002,
"learning_rate": 0.0002,
"epoch": 2.63,
"step": 150
},
{
"eval_code_easy_loss": 0.07808719575405121,
"eval_code_easy_score": -0.024909913539886475,
"eval_code_easy_brier_score": 0.024909913539886475,
"eval_code_easy_average_probability": 0.9644252061843872,
"eval_code_easy_accuracy": 0.96,
"eval_code_easy_probabilities": [
1.0,
1.0,
1.0,
1.0,
1.0,
1.0,
0.25757572054862976,
1.0,
1.0,
1.0,
1.0,
1.0,
0.9849016070365906,
1.0,
1.0,
0.9876202940940857,
1.0,
1.0,
1.0,
0.9999861717224121,
0.9844418168067932,
1.0,
1.0,
1.0,
1.0,
1.0,
0.9999815225601196,
0.9999998807907104,
1.0,
0.9999998807907104,
0.9330133199691772,
0.9998921155929565,
0.9747464060783386,
0.9998412132263184,
0.9905625581741333,
0.974348783493042,
0.9999979734420776,
1.0,
0.9953662157058716,
1.0,
1.0,
0.9999998807907104,
0.9999034404754639,
1.0,
0.999864935874939,
0.9999926090240479,
0.4541763961315155,
0.9853015542030334,
0.9999998807907104,
1.0,
1.0,
0.9991912245750427,
1.0,
1.0,
1.0,
1.0,
1.0,
1.0,
0.8326194882392883,
1.0,
1.0,
1.0,
0.9997536540031433,
1.0,
1.0,
1.0,
1.0,
1.0,
1.0,
0.9999920129776001,
0.16892309486865997,
0.04870113357901573,
1.0,
1.0,
1.0,
0.9999966621398926,
0.9999991655349731,
1.0,
1.0,
1.0,
1.0,
0.9763332009315491,
1.0,
1.0,
0.9999935626983643,
1.0,
1.0,
1.0,
0.9999570846557617,
0.9999858140945435,
1.0,
0.8955685496330261,
0.9999994039535522,
1.0,
1.0,
1.0,
1.0,
1.0,
1.0,
1.0
],
"eval_code_easy_runtime": 31.1286,
"eval_code_easy_samples_per_second": 3.212,
"eval_code_easy_steps_per_second": 0.128,
"epoch": 2.63,
"step": 150
},
{
"eval_code_hard_loss": 0.5192895531654358,
"eval_code_hard_score": -0.15126831829547882,
"eval_code_hard_brier_score": 0.15126831829547882,
"eval_code_hard_average_probability": 0.747479259967804,
"eval_code_hard_accuracy": 0.8,
"eval_code_hard_probabilities": [
1.0,
0.23972108960151672,
0.34699881076812744,
0.9999767541885376,
0.9995144605636597,
0.9983108043670654,
0.08374932408332825,
0.43437740206718445,
0.01867709495127201,
0.6382063031196594,
0.15230077505111694,
0.8396382331848145,
0.9999982118606567,
0.9995055198669434,
0.9989895224571228,
0.999188244342804,
0.9740042686462402,
0.9889111518859863,
0.9999994039535522,
0.9998956918716431,
0.9998466968536377,
0.9999018907546997,
0.999890923500061,
0.9991710186004639,
0.5053916573524475,
0.39974549412727356,
0.42459815740585327,
0.0008407415589317679,
0.42807716131210327,
0.5588884949684143,
0.9891363382339478,
0.9961511492729187,
0.7021331191062927,
0.9854065179824829,
0.8423302173614502,
0.8964764475822449,
0.1724509447813034,
0.12598082423210144,
0.7715829610824585,
0.7074795365333557,
0.5872290134429932,
0.8601648211479187,
0.9107590317726135,
0.6049399375915527,
0.9440699219703674,
0.9954397082328796,
0.5415297150611877,
0.787742555141449,
0.9602062702178955,
0.9986072182655334,
0.9982185959815979,
0.1667134314775467,
0.9801592826843262,
0.9148990511894226,
0.9998517036437988,
0.9998817443847656,
0.9999765157699585,
0.3098694384098053,
0.6319935917854309,
0.5650399923324585,
0.999991774559021,
0.9999759197235107,
0.999995231628418,
0.9999890327453613,
0.9999120235443115,
1.0,
0.9366697072982788,
0.9999974966049194,
0.9823035597801208,
0.9867276549339294,
0.8337392807006836,
0.13807430863380432,
0.9943860769271851,
0.9992918968200684,
0.9986520409584045,
0.8924828767776489,
0.1779462993144989,
0.14102093875408173,
0.5240833163261414,
0.4705985486507416,
0.5385470390319824,
0.9867573976516724,
0.9934804439544678,
0.9309672713279724,
0.7547566890716553,
0.46812355518341064,
0.6644623279571533,
0.5478729605674744,
0.9080039262771606,
0.9042761325836182,
0.564723789691925,
0.7475408911705017,
0.5996127128601074,
0.6277127861976624,
0.6707883477210999,
0.3496539294719696,
0.9058393836021423,
0.9995261430740356,
0.902912437915802,
0.6317912936210632
],
"eval_code_hard_runtime": 110.7028,
"eval_code_hard_samples_per_second": 0.903,
"eval_code_hard_steps_per_second": 0.036,
"epoch": 2.63,
"step": 150
},
{
"train_runtime": 4923.5411,
"train_samples_per_second": 0.975,
"train_steps_per_second": 0.03,
"total_flos": 0.0,
"train_loss": 0.09649283700817553,
"epoch": 2.63,
"step": 150
}
]