ryanmarten's picture
End of training
ed7612a verified
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 12.30379746835443,
"eval_steps": 500,
"global_step": 117,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.10126582278481013,
"grad_norm": 6.728910117257435,
"learning_rate": 8.333333333333333e-07,
"loss": 0.8179,
"step": 1
},
{
"epoch": 0.20253164556962025,
"grad_norm": 6.473835658322127,
"learning_rate": 1.6666666666666667e-06,
"loss": 0.8168,
"step": 2
},
{
"epoch": 0.3037974683544304,
"grad_norm": 7.398821648581466,
"learning_rate": 2.5e-06,
"loss": 0.834,
"step": 3
},
{
"epoch": 0.4050632911392405,
"grad_norm": 6.584310954429219,
"learning_rate": 3.3333333333333333e-06,
"loss": 0.7798,
"step": 4
},
{
"epoch": 0.5063291139240507,
"grad_norm": 5.084032702587007,
"learning_rate": 4.166666666666667e-06,
"loss": 0.7524,
"step": 5
},
{
"epoch": 0.6075949367088608,
"grad_norm": 3.2198552942423944,
"learning_rate": 5e-06,
"loss": 0.7267,
"step": 6
},
{
"epoch": 0.7088607594936709,
"grad_norm": 2.55272306221328,
"learning_rate": 5.833333333333334e-06,
"loss": 0.6673,
"step": 7
},
{
"epoch": 0.810126582278481,
"grad_norm": 3.8837417264179663,
"learning_rate": 6.666666666666667e-06,
"loss": 0.6795,
"step": 8
},
{
"epoch": 0.9113924050632911,
"grad_norm": 3.6714765633824995,
"learning_rate": 7.500000000000001e-06,
"loss": 0.6527,
"step": 9
},
{
"epoch": 1.0506329113924051,
"grad_norm": 3.8825834260208496,
"learning_rate": 8.333333333333334e-06,
"loss": 0.6359,
"step": 10
},
{
"epoch": 1.1518987341772151,
"grad_norm": 3.9729236944508886,
"learning_rate": 9.166666666666666e-06,
"loss": 0.5631,
"step": 11
},
{
"epoch": 1.2531645569620253,
"grad_norm": 3.6138955588837867,
"learning_rate": 1e-05,
"loss": 0.5601,
"step": 12
},
{
"epoch": 1.3544303797468356,
"grad_norm": 3.265414836256373,
"learning_rate": 9.997762161417517e-06,
"loss": 0.5788,
"step": 13
},
{
"epoch": 1.4556962025316456,
"grad_norm": 1.9909572119770438,
"learning_rate": 9.991050648838676e-06,
"loss": 0.517,
"step": 14
},
{
"epoch": 1.5569620253164556,
"grad_norm": 1.8360673956993219,
"learning_rate": 9.979871469976197e-06,
"loss": 0.4874,
"step": 15
},
{
"epoch": 1.6582278481012658,
"grad_norm": 2.567833203838775,
"learning_rate": 9.964234631709188e-06,
"loss": 0.5641,
"step": 16
},
{
"epoch": 1.759493670886076,
"grad_norm": 1.9811665993980823,
"learning_rate": 9.944154131125643e-06,
"loss": 0.5136,
"step": 17
},
{
"epoch": 1.8607594936708862,
"grad_norm": 1.3837110420757606,
"learning_rate": 9.91964794299315e-06,
"loss": 0.5057,
"step": 18
},
{
"epoch": 1.9620253164556962,
"grad_norm": 1.112706689940529,
"learning_rate": 9.890738003669029e-06,
"loss": 0.4584,
"step": 19
},
{
"epoch": 2.1012658227848102,
"grad_norm": 1.4254973278869227,
"learning_rate": 9.857450191464337e-06,
"loss": 0.4599,
"step": 20
},
{
"epoch": 2.2025316455696204,
"grad_norm": 1.0081891560682015,
"learning_rate": 9.819814303479268e-06,
"loss": 0.4511,
"step": 21
},
{
"epoch": 2.3037974683544302,
"grad_norm": 0.9947534992001078,
"learning_rate": 9.777864028930705e-06,
"loss": 0.4246,
"step": 22
},
{
"epoch": 2.4050632911392404,
"grad_norm": 0.9206621584796949,
"learning_rate": 9.731636918995821e-06,
"loss": 0.4366,
"step": 23
},
{
"epoch": 2.5063291139240507,
"grad_norm": 0.905341675595253,
"learning_rate": 9.681174353198687e-06,
"loss": 0.4095,
"step": 24
},
{
"epoch": 2.607594936708861,
"grad_norm": 0.8475454129162597,
"learning_rate": 9.626521502369984e-06,
"loss": 0.4003,
"step": 25
},
{
"epoch": 2.708860759493671,
"grad_norm": 0.9286044335225481,
"learning_rate": 9.567727288213005e-06,
"loss": 0.3898,
"step": 26
},
{
"epoch": 2.810126582278481,
"grad_norm": 1.0108216538432937,
"learning_rate": 9.504844339512096e-06,
"loss": 0.3755,
"step": 27
},
{
"epoch": 2.911392405063291,
"grad_norm": 0.7359666712313218,
"learning_rate": 9.437928945022772e-06,
"loss": 0.3531,
"step": 28
},
{
"epoch": 3.050632911392405,
"grad_norm": 1.1003060487256515,
"learning_rate": 9.36704100308565e-06,
"loss": 0.3543,
"step": 29
},
{
"epoch": 3.151898734177215,
"grad_norm": 1.022158232189453,
"learning_rate": 9.292243968009332e-06,
"loss": 0.3238,
"step": 30
},
{
"epoch": 3.2531645569620253,
"grad_norm": 0.9632808975543754,
"learning_rate": 9.213604793270196e-06,
"loss": 0.2875,
"step": 31
},
{
"epoch": 3.3544303797468356,
"grad_norm": 0.8757977440972499,
"learning_rate": 9.131193871579975e-06,
"loss": 0.2798,
"step": 32
},
{
"epoch": 3.4556962025316453,
"grad_norm": 1.0464451495734772,
"learning_rate": 9.045084971874738e-06,
"loss": 0.307,
"step": 33
},
{
"epoch": 3.5569620253164556,
"grad_norm": 0.7870485523892643,
"learning_rate": 8.955355173281709e-06,
"loss": 0.3339,
"step": 34
},
{
"epoch": 3.6582278481012658,
"grad_norm": 0.8749939817689344,
"learning_rate": 8.862084796122998e-06,
"loss": 0.2654,
"step": 35
},
{
"epoch": 3.759493670886076,
"grad_norm": 0.9511433986476833,
"learning_rate": 8.765357330018056e-06,
"loss": 0.2948,
"step": 36
},
{
"epoch": 3.8607594936708862,
"grad_norm": 0.9162284510959156,
"learning_rate": 8.665259359149132e-06,
"loss": 0.2821,
"step": 37
},
{
"epoch": 3.962025316455696,
"grad_norm": 0.864631769850015,
"learning_rate": 8.561880484756726e-06,
"loss": 0.3166,
"step": 38
},
{
"epoch": 4.10126582278481,
"grad_norm": 0.6896951075260729,
"learning_rate": 8.455313244934324e-06,
"loss": 0.2576,
"step": 39
},
{
"epoch": 4.2025316455696204,
"grad_norm": 0.8101365328392494,
"learning_rate": 8.345653031794292e-06,
"loss": 0.1959,
"step": 40
},
{
"epoch": 4.30379746835443,
"grad_norm": 0.7463914560938447,
"learning_rate": 8.232998006078998e-06,
"loss": 0.1915,
"step": 41
},
{
"epoch": 4.405063291139241,
"grad_norm": 0.988480478229143,
"learning_rate": 8.117449009293668e-06,
"loss": 0.2314,
"step": 42
},
{
"epoch": 4.506329113924051,
"grad_norm": 0.951141131289634,
"learning_rate": 7.99910947343957e-06,
"loss": 0.1972,
"step": 43
},
{
"epoch": 4.6075949367088604,
"grad_norm": 0.7553914171595718,
"learning_rate": 7.87808532842837e-06,
"loss": 0.2049,
"step": 44
},
{
"epoch": 4.708860759493671,
"grad_norm": 0.8454222299909153,
"learning_rate": 7.754484907260513e-06,
"loss": 0.1914,
"step": 45
},
{
"epoch": 4.810126582278481,
"grad_norm": 0.8862976393099018,
"learning_rate": 7.628418849052523e-06,
"loss": 0.1886,
"step": 46
},
{
"epoch": 4.911392405063291,
"grad_norm": 0.7541590683184227,
"learning_rate": 7.500000000000001e-06,
"loss": 0.2172,
"step": 47
},
{
"epoch": 5.050632911392405,
"grad_norm": 0.7195197835742276,
"learning_rate": 7.369343312364994e-06,
"loss": 0.1794,
"step": 48
},
{
"epoch": 5.151898734177215,
"grad_norm": 0.6024166545036832,
"learning_rate": 7.236565741578163e-06,
"loss": 0.1595,
"step": 49
},
{
"epoch": 5.253164556962025,
"grad_norm": 0.6554237065642754,
"learning_rate": 7.101786141547829e-06,
"loss": 0.1304,
"step": 50
},
{
"epoch": 5.3544303797468356,
"grad_norm": 0.6509051204915938,
"learning_rate": 6.965125158269619e-06,
"loss": 0.1497,
"step": 51
},
{
"epoch": 5.455696202531645,
"grad_norm": 0.6952076058819326,
"learning_rate": 6.8267051218319766e-06,
"loss": 0.1158,
"step": 52
},
{
"epoch": 5.556962025316456,
"grad_norm": 0.7042238449961647,
"learning_rate": 6.686649936914151e-06,
"loss": 0.1724,
"step": 53
},
{
"epoch": 5.658227848101266,
"grad_norm": 0.6337129163776152,
"learning_rate": 6.545084971874738e-06,
"loss": 0.1457,
"step": 54
},
{
"epoch": 5.759493670886076,
"grad_norm": 0.6078950596216507,
"learning_rate": 6.402136946530014e-06,
"loss": 0.1611,
"step": 55
},
{
"epoch": 5.860759493670886,
"grad_norm": 0.6364592856208567,
"learning_rate": 6.257933818722544e-06,
"loss": 0.1592,
"step": 56
},
{
"epoch": 5.962025316455696,
"grad_norm": 0.672548179809244,
"learning_rate": 6.112604669781572e-06,
"loss": 0.2142,
"step": 57
},
{
"epoch": 6.10126582278481,
"grad_norm": 0.5560263904880448,
"learning_rate": 5.9662795889777666e-06,
"loss": 0.1173,
"step": 58
},
{
"epoch": 6.2025316455696204,
"grad_norm": 0.5095979932767248,
"learning_rate": 5.819089557075689e-06,
"loss": 0.0833,
"step": 59
},
{
"epoch": 6.30379746835443,
"grad_norm": 0.5290858382090516,
"learning_rate": 5.671166329088278e-06,
"loss": 0.1027,
"step": 60
},
{
"epoch": 6.405063291139241,
"grad_norm": 0.6398843105260311,
"learning_rate": 5.522642316338268e-06,
"loss": 0.16,
"step": 61
},
{
"epoch": 6.506329113924051,
"grad_norm": 0.6550217956331817,
"learning_rate": 5.373650467932122e-06,
"loss": 0.0839,
"step": 62
},
{
"epoch": 6.6075949367088604,
"grad_norm": 0.6183671514154334,
"learning_rate": 5.224324151752575e-06,
"loss": 0.1668,
"step": 63
},
{
"epoch": 6.708860759493671,
"grad_norm": 0.6083583530709221,
"learning_rate": 5.074797035076319e-06,
"loss": 0.1145,
"step": 64
},
{
"epoch": 6.810126582278481,
"grad_norm": 0.5606800200783669,
"learning_rate": 4.9252029649236835e-06,
"loss": 0.1313,
"step": 65
},
{
"epoch": 6.911392405063291,
"grad_norm": 0.5475586507338089,
"learning_rate": 4.775675848247427e-06,
"loss": 0.1189,
"step": 66
},
{
"epoch": 7.050632911392405,
"grad_norm": 0.46921696096132526,
"learning_rate": 4.626349532067879e-06,
"loss": 0.0563,
"step": 67
},
{
"epoch": 7.151898734177215,
"grad_norm": 0.4536792946030882,
"learning_rate": 4.477357683661734e-06,
"loss": 0.0746,
"step": 68
},
{
"epoch": 7.253164556962025,
"grad_norm": 0.4405628749997321,
"learning_rate": 4.3288336709117246e-06,
"loss": 0.0738,
"step": 69
},
{
"epoch": 7.3544303797468356,
"grad_norm": 0.41243523179311004,
"learning_rate": 4.180910442924312e-06,
"loss": 0.0713,
"step": 70
},
{
"epoch": 7.455696202531645,
"grad_norm": 0.436190503718249,
"learning_rate": 4.033720411022235e-06,
"loss": 0.0942,
"step": 71
},
{
"epoch": 7.556962025316456,
"grad_norm": 0.6862103479550395,
"learning_rate": 3.887395330218429e-06,
"loss": 0.1044,
"step": 72
},
{
"epoch": 7.658227848101266,
"grad_norm": 0.5299771123589089,
"learning_rate": 3.7420661812774577e-06,
"loss": 0.0767,
"step": 73
},
{
"epoch": 7.759493670886076,
"grad_norm": 0.489112686998527,
"learning_rate": 3.5978630534699873e-06,
"loss": 0.1233,
"step": 74
},
{
"epoch": 7.860759493670886,
"grad_norm": 0.467512841460672,
"learning_rate": 3.4549150281252635e-06,
"loss": 0.0734,
"step": 75
},
{
"epoch": 7.962025316455696,
"grad_norm": 0.5235724074035072,
"learning_rate": 3.3133500630858507e-06,
"loss": 0.0953,
"step": 76
},
{
"epoch": 8.10126582278481,
"grad_norm": 0.4501277083179513,
"learning_rate": 3.173294878168025e-06,
"loss": 0.0776,
"step": 77
},
{
"epoch": 8.20253164556962,
"grad_norm": 0.4061682395030546,
"learning_rate": 3.0348748417303826e-06,
"loss": 0.0876,
"step": 78
},
{
"epoch": 8.30379746835443,
"grad_norm": 0.563583814150666,
"learning_rate": 2.8982138584521734e-06,
"loss": 0.1033,
"step": 79
},
{
"epoch": 8.405063291139241,
"grad_norm": 0.3611301836829534,
"learning_rate": 2.7634342584218364e-06,
"loss": 0.0711,
"step": 80
},
{
"epoch": 8.50632911392405,
"grad_norm": 0.29468565290173476,
"learning_rate": 2.6306566876350072e-06,
"loss": 0.0721,
"step": 81
},
{
"epoch": 8.60759493670886,
"grad_norm": 0.3599075827244934,
"learning_rate": 2.5000000000000015e-06,
"loss": 0.0639,
"step": 82
},
{
"epoch": 8.708860759493671,
"grad_norm": 0.6094194349745108,
"learning_rate": 2.371581150947476e-06,
"loss": 0.077,
"step": 83
},
{
"epoch": 8.810126582278482,
"grad_norm": 0.4126927223027438,
"learning_rate": 2.245515092739488e-06,
"loss": 0.03,
"step": 84
},
{
"epoch": 8.91139240506329,
"grad_norm": 0.5169353485223714,
"learning_rate": 2.1219146715716332e-06,
"loss": 0.092,
"step": 85
},
{
"epoch": 9.050632911392405,
"grad_norm": 0.32556490687670286,
"learning_rate": 2.0008905265604316e-06,
"loss": 0.0633,
"step": 86
},
{
"epoch": 9.151898734177216,
"grad_norm": 0.30150834723927206,
"learning_rate": 1.8825509907063328e-06,
"loss": 0.0483,
"step": 87
},
{
"epoch": 9.253164556962025,
"grad_norm": 0.2975815205222138,
"learning_rate": 1.7670019939210025e-06,
"loss": 0.0618,
"step": 88
},
{
"epoch": 9.354430379746836,
"grad_norm": 0.3006894295436638,
"learning_rate": 1.6543469682057105e-06,
"loss": 0.0514,
"step": 89
},
{
"epoch": 9.455696202531646,
"grad_norm": 0.34996593746798194,
"learning_rate": 1.544686755065677e-06,
"loss": 0.0883,
"step": 90
},
{
"epoch": 9.556962025316455,
"grad_norm": 0.24859979242344823,
"learning_rate": 1.438119515243277e-06,
"loss": 0.0414,
"step": 91
},
{
"epoch": 9.658227848101266,
"grad_norm": 0.3165392376817483,
"learning_rate": 1.3347406408508695e-06,
"loss": 0.0603,
"step": 92
},
{
"epoch": 9.759493670886076,
"grad_norm": 0.2926995890709802,
"learning_rate": 1.234642669981946e-06,
"loss": 0.0581,
"step": 93
},
{
"epoch": 9.860759493670885,
"grad_norm": 0.29442873934011776,
"learning_rate": 1.137915203877003e-06,
"loss": 0.0665,
"step": 94
},
{
"epoch": 9.962025316455696,
"grad_norm": 0.3315346643694066,
"learning_rate": 1.044644826718295e-06,
"loss": 0.0539,
"step": 95
},
{
"epoch": 10.10126582278481,
"grad_norm": 0.27098073813593876,
"learning_rate": 9.549150281252633e-07,
"loss": 0.0784,
"step": 96
},
{
"epoch": 10.20253164556962,
"grad_norm": 0.2647348750888826,
"learning_rate": 8.688061284200266e-07,
"loss": 0.0714,
"step": 97
},
{
"epoch": 10.30379746835443,
"grad_norm": 0.2784953663272517,
"learning_rate": 7.863952067298042e-07,
"loss": 0.0594,
"step": 98
},
{
"epoch": 10.405063291139241,
"grad_norm": 0.19778748823093734,
"learning_rate": 7.077560319906696e-07,
"loss": 0.0317,
"step": 99
},
{
"epoch": 10.50632911392405,
"grad_norm": 0.29517988678939616,
"learning_rate": 6.329589969143518e-07,
"loss": 0.0708,
"step": 100
},
{
"epoch": 10.60759493670886,
"grad_norm": 0.20754297353657086,
"learning_rate": 5.620710549772295e-07,
"loss": 0.031,
"step": 101
},
{
"epoch": 10.708860759493671,
"grad_norm": 0.2940858791604344,
"learning_rate": 4.951556604879049e-07,
"loss": 0.0731,
"step": 102
},
{
"epoch": 10.810126582278482,
"grad_norm": 0.261713833222536,
"learning_rate": 4.322727117869951e-07,
"loss": 0.0358,
"step": 103
},
{
"epoch": 10.91139240506329,
"grad_norm": 0.23803340026965883,
"learning_rate": 3.734784976300165e-07,
"loss": 0.0431,
"step": 104
},
{
"epoch": 11.050632911392405,
"grad_norm": 0.2824612271203908,
"learning_rate": 3.18825646801314e-07,
"loss": 0.0783,
"step": 105
},
{
"epoch": 11.151898734177216,
"grad_norm": 0.25265121565812304,
"learning_rate": 2.6836308100417874e-07,
"loss": 0.0657,
"step": 106
},
{
"epoch": 11.253164556962025,
"grad_norm": 0.25620323133440565,
"learning_rate": 2.2213597106929608e-07,
"loss": 0.0597,
"step": 107
},
{
"epoch": 11.354430379746836,
"grad_norm": 0.2151041931514239,
"learning_rate": 1.801856965207338e-07,
"loss": 0.0431,
"step": 108
},
{
"epoch": 11.455696202531646,
"grad_norm": 0.19922324339269704,
"learning_rate": 1.4254980853566248e-07,
"loss": 0.0274,
"step": 109
},
{
"epoch": 11.556962025316455,
"grad_norm": 0.2524072327724453,
"learning_rate": 1.0926199633097156e-07,
"loss": 0.0675,
"step": 110
},
{
"epoch": 11.658227848101266,
"grad_norm": 0.2477956602271147,
"learning_rate": 8.035205700685167e-08,
"loss": 0.0625,
"step": 111
},
{
"epoch": 11.759493670886076,
"grad_norm": 0.2015856230534981,
"learning_rate": 5.584586887435739e-08,
"loss": 0.0451,
"step": 112
},
{
"epoch": 11.860759493670885,
"grad_norm": 0.21600615142224264,
"learning_rate": 3.576536829081323e-08,
"loss": 0.0573,
"step": 113
},
{
"epoch": 11.962025316455696,
"grad_norm": 0.22005227561723967,
"learning_rate": 2.012853002380466e-08,
"loss": 0.0349,
"step": 114
},
{
"epoch": 12.10126582278481,
"grad_norm": 0.2155276475929545,
"learning_rate": 8.949351161324227e-09,
"loss": 0.0451,
"step": 115
},
{
"epoch": 12.20253164556962,
"grad_norm": 0.21265275258133334,
"learning_rate": 2.237838582483387e-09,
"loss": 0.0456,
"step": 116
},
{
"epoch": 12.30379746835443,
"grad_norm": 0.2129554703078614,
"learning_rate": 0.0,
"loss": 0.0369,
"step": 117
},
{
"epoch": 12.30379746835443,
"step": 117,
"total_flos": 4.943496163937485e+16,
"train_loss": 0.2279800043369715,
"train_runtime": 2783.2711,
"train_samples_per_second": 1.476,
"train_steps_per_second": 0.042
}
],
"logging_steps": 1,
"max_steps": 117,
"num_input_tokens_seen": 0,
"num_train_epochs": 13,
"save_steps": 500,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 4.943496163937485e+16,
"train_batch_size": 1,
"trial_name": null,
"trial_params": null
}