Gemma-2-9B-sft-ultrachat-safeRLHF / trainer_state.json
AmberYifan's picture
Model save
7d50dde verified
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 1.0,
"eval_steps": 500,
"global_step": 1563,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0006397952655150352,
"grad_norm": 9.818162324935793,
"learning_rate": 1.2738853503184715e-07,
"loss": 2.1195,
"step": 1
},
{
"epoch": 0.003198976327575176,
"grad_norm": 9.505459485507984,
"learning_rate": 6.369426751592357e-07,
"loss": 2.0459,
"step": 5
},
{
"epoch": 0.006397952655150352,
"grad_norm": 6.399275095836299,
"learning_rate": 1.2738853503184715e-06,
"loss": 2.0889,
"step": 10
},
{
"epoch": 0.009596928982725527,
"grad_norm": 4.401353406055998,
"learning_rate": 1.9108280254777074e-06,
"loss": 1.8888,
"step": 15
},
{
"epoch": 0.012795905310300703,
"grad_norm": 4.385991217899751,
"learning_rate": 2.547770700636943e-06,
"loss": 1.8725,
"step": 20
},
{
"epoch": 0.01599488163787588,
"grad_norm": 4.518574974361523,
"learning_rate": 3.1847133757961785e-06,
"loss": 1.7284,
"step": 25
},
{
"epoch": 0.019193857965451054,
"grad_norm": 4.834892269579109,
"learning_rate": 3.821656050955415e-06,
"loss": 1.7245,
"step": 30
},
{
"epoch": 0.022392834293026232,
"grad_norm": 3.756079300112788,
"learning_rate": 4.45859872611465e-06,
"loss": 1.6296,
"step": 35
},
{
"epoch": 0.025591810620601407,
"grad_norm": 3.7234946569353635,
"learning_rate": 5.095541401273886e-06,
"loss": 1.6382,
"step": 40
},
{
"epoch": 0.028790786948176585,
"grad_norm": 4.150736324278731,
"learning_rate": 5.732484076433121e-06,
"loss": 1.6168,
"step": 45
},
{
"epoch": 0.03198976327575176,
"grad_norm": 3.895173253057345,
"learning_rate": 6.369426751592357e-06,
"loss": 1.6558,
"step": 50
},
{
"epoch": 0.035188739603326934,
"grad_norm": 3.485478395791523,
"learning_rate": 7.006369426751593e-06,
"loss": 1.6513,
"step": 55
},
{
"epoch": 0.03838771593090211,
"grad_norm": 3.5195775660121638,
"learning_rate": 7.64331210191083e-06,
"loss": 1.6396,
"step": 60
},
{
"epoch": 0.04158669225847729,
"grad_norm": 3.595990481188834,
"learning_rate": 8.280254777070064e-06,
"loss": 1.6568,
"step": 65
},
{
"epoch": 0.044785668586052464,
"grad_norm": 3.5463472470575703,
"learning_rate": 8.9171974522293e-06,
"loss": 1.6289,
"step": 70
},
{
"epoch": 0.04798464491362764,
"grad_norm": 3.6001979942436586,
"learning_rate": 9.554140127388536e-06,
"loss": 1.6104,
"step": 75
},
{
"epoch": 0.05118362124120281,
"grad_norm": 3.6006925973698785,
"learning_rate": 1.0191082802547772e-05,
"loss": 1.6654,
"step": 80
},
{
"epoch": 0.05438259756877799,
"grad_norm": 3.716531800009205,
"learning_rate": 1.0828025477707008e-05,
"loss": 1.6388,
"step": 85
},
{
"epoch": 0.05758157389635317,
"grad_norm": 3.487280295033822,
"learning_rate": 1.1464968152866242e-05,
"loss": 1.628,
"step": 90
},
{
"epoch": 0.060780550223928344,
"grad_norm": 3.4105805165678964,
"learning_rate": 1.2101910828025478e-05,
"loss": 1.6259,
"step": 95
},
{
"epoch": 0.06397952655150352,
"grad_norm": 3.512439191363734,
"learning_rate": 1.2738853503184714e-05,
"loss": 1.6506,
"step": 100
},
{
"epoch": 0.0671785028790787,
"grad_norm": 3.7709019777624495,
"learning_rate": 1.337579617834395e-05,
"loss": 1.6537,
"step": 105
},
{
"epoch": 0.07037747920665387,
"grad_norm": 3.6562830068013574,
"learning_rate": 1.4012738853503186e-05,
"loss": 1.6325,
"step": 110
},
{
"epoch": 0.07357645553422905,
"grad_norm": 3.449617694349996,
"learning_rate": 1.464968152866242e-05,
"loss": 1.6366,
"step": 115
},
{
"epoch": 0.07677543186180422,
"grad_norm": 3.3992478775853208,
"learning_rate": 1.528662420382166e-05,
"loss": 1.6543,
"step": 120
},
{
"epoch": 0.0799744081893794,
"grad_norm": 3.620655324659728,
"learning_rate": 1.5923566878980894e-05,
"loss": 1.6708,
"step": 125
},
{
"epoch": 0.08317338451695458,
"grad_norm": 3.198472040582768,
"learning_rate": 1.6560509554140128e-05,
"loss": 1.6503,
"step": 130
},
{
"epoch": 0.08637236084452975,
"grad_norm": 3.4666596666281815,
"learning_rate": 1.7197452229299365e-05,
"loss": 1.6706,
"step": 135
},
{
"epoch": 0.08957133717210493,
"grad_norm": 3.288828271719929,
"learning_rate": 1.78343949044586e-05,
"loss": 1.7237,
"step": 140
},
{
"epoch": 0.0927703134996801,
"grad_norm": 3.304421530141081,
"learning_rate": 1.8471337579617837e-05,
"loss": 1.7127,
"step": 145
},
{
"epoch": 0.09596928982725528,
"grad_norm": 3.587569284428709,
"learning_rate": 1.910828025477707e-05,
"loss": 1.7149,
"step": 150
},
{
"epoch": 0.09916826615483046,
"grad_norm": 3.339928814927714,
"learning_rate": 1.9745222929936306e-05,
"loss": 1.7396,
"step": 155
},
{
"epoch": 0.10236724248240563,
"grad_norm": 3.0920538847531702,
"learning_rate": 1.9999775332635076e-05,
"loss": 1.6822,
"step": 160
},
{
"epoch": 0.10556621880998081,
"grad_norm": 3.1119179011877045,
"learning_rate": 1.999840240196313e-05,
"loss": 1.6455,
"step": 165
},
{
"epoch": 0.10876519513755598,
"grad_norm": 3.2607882610929178,
"learning_rate": 1.9995781526975738e-05,
"loss": 1.6995,
"step": 170
},
{
"epoch": 0.11196417146513116,
"grad_norm": 3.065897985133359,
"learning_rate": 1.9991913034795768e-05,
"loss": 1.7373,
"step": 175
},
{
"epoch": 0.11516314779270634,
"grad_norm": 3.2287016110869,
"learning_rate": 1.9986797408266636e-05,
"loss": 1.7028,
"step": 180
},
{
"epoch": 0.1183621241202815,
"grad_norm": 3.0788887296688023,
"learning_rate": 1.9980435285892056e-05,
"loss": 1.7842,
"step": 185
},
{
"epoch": 0.12156110044785669,
"grad_norm": 3.2566132546479087,
"learning_rate": 1.9972827461756335e-05,
"loss": 1.711,
"step": 190
},
{
"epoch": 0.12476007677543186,
"grad_norm": 3.3047400174990926,
"learning_rate": 1.9963974885425267e-05,
"loss": 1.6575,
"step": 195
},
{
"epoch": 0.12795905310300704,
"grad_norm": 3.0777466153557986,
"learning_rate": 1.9953878661827603e-05,
"loss": 1.7255,
"step": 200
},
{
"epoch": 0.13115802943058222,
"grad_norm": 3.080857837349394,
"learning_rate": 1.994254005111715e-05,
"loss": 1.7133,
"step": 205
},
{
"epoch": 0.1343570057581574,
"grad_norm": 2.9854955668238916,
"learning_rate": 1.992996046851548e-05,
"loss": 1.694,
"step": 210
},
{
"epoch": 0.13755598208573255,
"grad_norm": 3.1079992917047106,
"learning_rate": 1.9916141484135297e-05,
"loss": 1.7346,
"step": 215
},
{
"epoch": 0.14075495841330773,
"grad_norm": 3.0113817008188613,
"learning_rate": 1.990108482278446e-05,
"loss": 1.7173,
"step": 220
},
{
"epoch": 0.14395393474088292,
"grad_norm": 2.925776698790043,
"learning_rate": 1.9884792363750684e-05,
"loss": 1.7477,
"step": 225
},
{
"epoch": 0.1471529110684581,
"grad_norm": 3.160949311743821,
"learning_rate": 1.9867266140567024e-05,
"loss": 1.7345,
"step": 230
},
{
"epoch": 0.15035188739603328,
"grad_norm": 2.9503878572877924,
"learning_rate": 1.9848508340758014e-05,
"loss": 1.7435,
"step": 235
},
{
"epoch": 0.15355086372360843,
"grad_norm": 2.941101822473189,
"learning_rate": 1.9828521305566647e-05,
"loss": 1.7259,
"step": 240
},
{
"epoch": 0.15674984005118361,
"grad_norm": 2.767610044521168,
"learning_rate": 1.9807307529662175e-05,
"loss": 1.805,
"step": 245
},
{
"epoch": 0.1599488163787588,
"grad_norm": 3.053392700653573,
"learning_rate": 1.9784869660828708e-05,
"loss": 1.7101,
"step": 250
},
{
"epoch": 0.16314779270633398,
"grad_norm": 2.989674959136692,
"learning_rate": 1.9761210499634754e-05,
"loss": 1.704,
"step": 255
},
{
"epoch": 0.16634676903390916,
"grad_norm": 3.0705122515165217,
"learning_rate": 1.973633299908365e-05,
"loss": 1.66,
"step": 260
},
{
"epoch": 0.1695457453614843,
"grad_norm": 3.024248714962046,
"learning_rate": 1.9710240264245005e-05,
"loss": 1.7734,
"step": 265
},
{
"epoch": 0.1727447216890595,
"grad_norm": 2.8383349511632763,
"learning_rate": 1.9682935551867132e-05,
"loss": 1.6998,
"step": 270
},
{
"epoch": 0.17594369801663468,
"grad_norm": 2.9015050169982017,
"learning_rate": 1.9654422269970545e-05,
"loss": 1.727,
"step": 275
},
{
"epoch": 0.17914267434420986,
"grad_norm": 3.0464927667956494,
"learning_rate": 1.9624703977422624e-05,
"loss": 1.7099,
"step": 280
},
{
"epoch": 0.18234165067178504,
"grad_norm": 2.9313739570308512,
"learning_rate": 1.959378438349338e-05,
"loss": 1.702,
"step": 285
},
{
"epoch": 0.1855406269993602,
"grad_norm": 2.907358102224952,
"learning_rate": 1.956166734739251e-05,
"loss": 1.7237,
"step": 290
},
{
"epoch": 0.18873960332693537,
"grad_norm": 2.811799487921981,
"learning_rate": 1.95283568777877e-05,
"loss": 1.711,
"step": 295
},
{
"epoch": 0.19193857965451055,
"grad_norm": 4.19572219544196,
"learning_rate": 1.9493857132304295e-05,
"loss": 1.6867,
"step": 300
},
{
"epoch": 0.19513755598208574,
"grad_norm": 2.8426947984401174,
"learning_rate": 1.9458172417006347e-05,
"loss": 1.8003,
"step": 305
},
{
"epoch": 0.19833653230966092,
"grad_norm": 2.883375733442829,
"learning_rate": 1.942130718585919e-05,
"loss": 1.679,
"step": 310
},
{
"epoch": 0.20153550863723607,
"grad_norm": 3.3624114664304003,
"learning_rate": 1.938326604017349e-05,
"loss": 1.7129,
"step": 315
},
{
"epoch": 0.20473448496481125,
"grad_norm": 2.725134459559926,
"learning_rate": 1.9344053728030952e-05,
"loss": 1.7181,
"step": 320
},
{
"epoch": 0.20793346129238643,
"grad_norm": 2.8242686598972946,
"learning_rate": 1.9303675143691683e-05,
"loss": 1.7315,
"step": 325
},
{
"epoch": 0.21113243761996162,
"grad_norm": 2.845090885429897,
"learning_rate": 1.9262135326983326e-05,
"loss": 1.7547,
"step": 330
},
{
"epoch": 0.2143314139475368,
"grad_norm": 3.206220986637601,
"learning_rate": 1.921943946267201e-05,
"loss": 1.6785,
"step": 335
},
{
"epoch": 0.21753039027511195,
"grad_norm": 3.0342577046806993,
"learning_rate": 1.9175592879815217e-05,
"loss": 1.7607,
"step": 340
},
{
"epoch": 0.22072936660268713,
"grad_norm": 3.032797295242115,
"learning_rate": 1.9130601051096655e-05,
"loss": 1.6788,
"step": 345
},
{
"epoch": 0.22392834293026231,
"grad_norm": 2.8960925106145177,
"learning_rate": 1.9084469592143154e-05,
"loss": 1.7025,
"step": 350
},
{
"epoch": 0.2271273192578375,
"grad_norm": 2.8762638217277012,
"learning_rate": 1.9037204260823788e-05,
"loss": 1.6549,
"step": 355
},
{
"epoch": 0.23032629558541268,
"grad_norm": 3.0122484094070585,
"learning_rate": 1.89888109565312e-05,
"loss": 1.7939,
"step": 360
},
{
"epoch": 0.23352527191298783,
"grad_norm": 2.9401025842808655,
"learning_rate": 1.893929571944527e-05,
"loss": 1.7231,
"step": 365
},
{
"epoch": 0.236724248240563,
"grad_norm": 3.036767799191246,
"learning_rate": 1.8888664729779205e-05,
"loss": 1.7363,
"step": 370
},
{
"epoch": 0.2399232245681382,
"grad_norm": 3.1496202191826312,
"learning_rate": 1.883692430700818e-05,
"loss": 1.6916,
"step": 375
},
{
"epoch": 0.24312220089571338,
"grad_norm": 3.093781443318627,
"learning_rate": 1.8784080909080568e-05,
"loss": 1.6775,
"step": 380
},
{
"epoch": 0.24632117722328856,
"grad_norm": 2.763430573340402,
"learning_rate": 1.8730141131611882e-05,
"loss": 1.7038,
"step": 385
},
{
"epoch": 0.2495201535508637,
"grad_norm": 2.7323495634260153,
"learning_rate": 1.867511170706157e-05,
"loss": 1.6371,
"step": 390
},
{
"epoch": 0.2527191298784389,
"grad_norm": 2.835885929233083,
"learning_rate": 1.861899950389269e-05,
"loss": 1.7327,
"step": 395
},
{
"epoch": 0.2559181062060141,
"grad_norm": 2.872513210302631,
"learning_rate": 1.856181152571463e-05,
"loss": 1.6571,
"step": 400
},
{
"epoch": 0.2591170825335892,
"grad_norm": 2.7458919501060746,
"learning_rate": 1.850355491040897e-05,
"loss": 1.6627,
"step": 405
},
{
"epoch": 0.26231605886116444,
"grad_norm": 2.642756700263611,
"learning_rate": 1.8444236929238556e-05,
"loss": 1.7072,
"step": 410
},
{
"epoch": 0.2655150351887396,
"grad_norm": 2.761453621063587,
"learning_rate": 1.8383864985939944e-05,
"loss": 1.6628,
"step": 415
},
{
"epoch": 0.2687140115163148,
"grad_norm": 2.8517438570867237,
"learning_rate": 1.8322446615799317e-05,
"loss": 1.6606,
"step": 420
},
{
"epoch": 0.27191298784388995,
"grad_norm": 2.791749625737528,
"learning_rate": 1.825998948471197e-05,
"loss": 1.6425,
"step": 425
},
{
"epoch": 0.2751119641714651,
"grad_norm": 2.7275042264122247,
"learning_rate": 1.819650138822548e-05,
"loss": 1.6929,
"step": 430
},
{
"epoch": 0.2783109404990403,
"grad_norm": 2.6242059640774955,
"learning_rate": 1.8131990250566733e-05,
"loss": 1.5758,
"step": 435
},
{
"epoch": 0.28150991682661547,
"grad_norm": 2.93301077735713,
"learning_rate": 1.8066464123652857e-05,
"loss": 1.7194,
"step": 440
},
{
"epoch": 0.2847088931541907,
"grad_norm": 2.5298341337807004,
"learning_rate": 1.7999931186086225e-05,
"loss": 1.6585,
"step": 445
},
{
"epoch": 0.28790786948176583,
"grad_norm": 2.7901766387801135,
"learning_rate": 1.793239974213364e-05,
"loss": 1.6331,
"step": 450
},
{
"epoch": 0.291106845809341,
"grad_norm": 2.707410220926997,
"learning_rate": 1.786387822068987e-05,
"loss": 1.6948,
"step": 455
},
{
"epoch": 0.2943058221369162,
"grad_norm": 2.8462563263590437,
"learning_rate": 1.7794375174225565e-05,
"loss": 1.7126,
"step": 460
},
{
"epoch": 0.29750479846449135,
"grad_norm": 2.7058643635028785,
"learning_rate": 1.7723899277719815e-05,
"loss": 1.6655,
"step": 465
},
{
"epoch": 0.30070377479206656,
"grad_norm": 2.8492698678800967,
"learning_rate": 1.7652459327577377e-05,
"loss": 1.6712,
"step": 470
},
{
"epoch": 0.3039027511196417,
"grad_norm": 2.737103283616345,
"learning_rate": 1.7580064240530746e-05,
"loss": 1.6477,
"step": 475
},
{
"epoch": 0.30710172744721687,
"grad_norm": 2.851777367830193,
"learning_rate": 1.7506723052527243e-05,
"loss": 1.6659,
"step": 480
},
{
"epoch": 0.3103007037747921,
"grad_norm": 2.81278535074852,
"learning_rate": 1.7432444917601183e-05,
"loss": 1.6698,
"step": 485
},
{
"epoch": 0.31349968010236723,
"grad_norm": 2.703148164333518,
"learning_rate": 1.735723910673132e-05,
"loss": 1.6434,
"step": 490
},
{
"epoch": 0.31669865642994244,
"grad_norm": 2.7572458644536404,
"learning_rate": 1.7281115006683687e-05,
"loss": 1.6401,
"step": 495
},
{
"epoch": 0.3198976327575176,
"grad_norm": 2.580280933814772,
"learning_rate": 1.7204082118840035e-05,
"loss": 1.6773,
"step": 500
},
{
"epoch": 0.32309660908509275,
"grad_norm": 2.809100743738505,
"learning_rate": 1.712615005801185e-05,
"loss": 1.6801,
"step": 505
},
{
"epoch": 0.32629558541266795,
"grad_norm": 2.718786555298531,
"learning_rate": 1.704732855124037e-05,
"loss": 1.7159,
"step": 510
},
{
"epoch": 0.3294945617402431,
"grad_norm": 2.7400741747111,
"learning_rate": 1.6967627436582445e-05,
"loss": 1.6919,
"step": 515
},
{
"epoch": 0.3326935380678183,
"grad_norm": 2.743563722428457,
"learning_rate": 1.6887056661882644e-05,
"loss": 1.7074,
"step": 520
},
{
"epoch": 0.33589251439539347,
"grad_norm": 2.882564956787008,
"learning_rate": 1.6805626283531592e-05,
"loss": 1.6343,
"step": 525
},
{
"epoch": 0.3390914907229686,
"grad_norm": 2.6521574148568066,
"learning_rate": 1.6723346465210815e-05,
"loss": 1.6902,
"step": 530
},
{
"epoch": 0.34229046705054383,
"grad_norm": 2.5884328719721044,
"learning_rate": 1.6640227476624163e-05,
"loss": 1.6956,
"step": 535
},
{
"epoch": 0.345489443378119,
"grad_norm": 2.87864958257569,
"learning_rate": 1.655627969221598e-05,
"loss": 1.6107,
"step": 540
},
{
"epoch": 0.3486884197056942,
"grad_norm": 3.0980520926972095,
"learning_rate": 1.6471513589876247e-05,
"loss": 1.661,
"step": 545
},
{
"epoch": 0.35188739603326935,
"grad_norm": 2.620565708665741,
"learning_rate": 1.638593974963278e-05,
"loss": 1.6772,
"step": 550
},
{
"epoch": 0.3550863723608445,
"grad_norm": 2.6689699527397917,
"learning_rate": 1.6299568852330703e-05,
"loss": 1.6939,
"step": 555
},
{
"epoch": 0.3582853486884197,
"grad_norm": 2.8262202588686547,
"learning_rate": 1.6212411678299306e-05,
"loss": 1.6404,
"step": 560
},
{
"epoch": 0.36148432501599487,
"grad_norm": 2.4712441299514207,
"learning_rate": 1.612447910600652e-05,
"loss": 1.6178,
"step": 565
},
{
"epoch": 0.3646833013435701,
"grad_norm": 2.5707414875075436,
"learning_rate": 1.6035782110701125e-05,
"loss": 1.6527,
"step": 570
},
{
"epoch": 0.36788227767114523,
"grad_norm": 2.6041967986181542,
"learning_rate": 1.594633176304287e-05,
"loss": 1.6641,
"step": 575
},
{
"epoch": 0.3710812539987204,
"grad_norm": 2.848409750236944,
"learning_rate": 1.5856139227720714e-05,
"loss": 1.668,
"step": 580
},
{
"epoch": 0.3742802303262956,
"grad_norm": 2.644618328434638,
"learning_rate": 1.5765215762059304e-05,
"loss": 1.6565,
"step": 585
},
{
"epoch": 0.37747920665387075,
"grad_norm": 2.7579483259716673,
"learning_rate": 1.5673572714613886e-05,
"loss": 1.6924,
"step": 590
},
{
"epoch": 0.38067818298144596,
"grad_norm": 2.7922858159984516,
"learning_rate": 1.558122152375387e-05,
"loss": 1.6436,
"step": 595
},
{
"epoch": 0.3838771593090211,
"grad_norm": 2.6930353850952353,
"learning_rate": 1.548817371623513e-05,
"loss": 1.6009,
"step": 600
},
{
"epoch": 0.38707613563659626,
"grad_norm": 2.36486151768098,
"learning_rate": 1.539444090576132e-05,
"loss": 1.6542,
"step": 605
},
{
"epoch": 0.3902751119641715,
"grad_norm": 2.6788082031369544,
"learning_rate": 1.5300034791534297e-05,
"loss": 1.6232,
"step": 610
},
{
"epoch": 0.3934740882917466,
"grad_norm": 2.4948030786170325,
"learning_rate": 1.520496715679391e-05,
"loss": 1.6156,
"step": 615
},
{
"epoch": 0.39667306461932184,
"grad_norm": 2.6411190301581353,
"learning_rate": 1.5109249867347276e-05,
"loss": 1.6339,
"step": 620
},
{
"epoch": 0.399872040946897,
"grad_norm": 2.4210113436531815,
"learning_rate": 1.5012894870087751e-05,
"loss": 1.6615,
"step": 625
},
{
"epoch": 0.40307101727447214,
"grad_norm": 2.5335344489734575,
"learning_rate": 1.4915914191503792e-05,
"loss": 1.6646,
"step": 630
},
{
"epoch": 0.40626999360204735,
"grad_norm": 2.807065076481401,
"learning_rate": 1.4818319936177885e-05,
"loss": 1.6435,
"step": 635
},
{
"epoch": 0.4094689699296225,
"grad_norm": 2.6650352639498487,
"learning_rate": 1.4720124285275703e-05,
"loss": 1.6288,
"step": 640
},
{
"epoch": 0.4126679462571977,
"grad_norm": 2.564827986402318,
"learning_rate": 1.4621339495025731e-05,
"loss": 1.6064,
"step": 645
},
{
"epoch": 0.41586692258477287,
"grad_norm": 2.7424460417510836,
"learning_rate": 1.4521977895189518e-05,
"loss": 1.594,
"step": 650
},
{
"epoch": 0.419065898912348,
"grad_norm": 2.868571986668108,
"learning_rate": 1.4422051887522735e-05,
"loss": 1.6503,
"step": 655
},
{
"epoch": 0.42226487523992323,
"grad_norm": 2.465998133123683,
"learning_rate": 1.4321573944227254e-05,
"loss": 1.6385,
"step": 660
},
{
"epoch": 0.4254638515674984,
"grad_norm": 2.5570954092252256,
"learning_rate": 1.4220556606394465e-05,
"loss": 1.642,
"step": 665
},
{
"epoch": 0.4286628278950736,
"grad_norm": 2.7008899441296172,
"learning_rate": 1.4119012482439929e-05,
"loss": 1.6255,
"step": 670
},
{
"epoch": 0.43186180422264875,
"grad_norm": 2.813582387432177,
"learning_rate": 1.4016954246529697e-05,
"loss": 1.5675,
"step": 675
},
{
"epoch": 0.4350607805502239,
"grad_norm": 2.540329992393098,
"learning_rate": 1.3914394636998374e-05,
"loss": 1.5953,
"step": 680
},
{
"epoch": 0.4382597568777991,
"grad_norm": 2.7487026331400366,
"learning_rate": 1.3811346454759211e-05,
"loss": 1.6582,
"step": 685
},
{
"epoch": 0.44145873320537427,
"grad_norm": 2.5702771778443996,
"learning_rate": 1.3707822561706336e-05,
"loss": 1.653,
"step": 690
},
{
"epoch": 0.4446577095329495,
"grad_norm": 3.2593493965983478,
"learning_rate": 1.3603835879109442e-05,
"loss": 1.5555,
"step": 695
},
{
"epoch": 0.44785668586052463,
"grad_norm": 2.4878223293276682,
"learning_rate": 1.349939938600099e-05,
"loss": 1.6246,
"step": 700
},
{
"epoch": 0.4510556621880998,
"grad_norm": 2.677733559213416,
"learning_rate": 1.3394526117556277e-05,
"loss": 1.6108,
"step": 705
},
{
"epoch": 0.454254638515675,
"grad_norm": 2.6110465429578067,
"learning_rate": 1.3289229163466421e-05,
"loss": 1.6129,
"step": 710
},
{
"epoch": 0.45745361484325014,
"grad_norm": 2.8063523069177485,
"learning_rate": 1.3183521666304611e-05,
"loss": 1.5705,
"step": 715
},
{
"epoch": 0.46065259117082535,
"grad_norm": 2.7217688329683205,
"learning_rate": 1.3077416819885707e-05,
"loss": 1.6206,
"step": 720
},
{
"epoch": 0.4638515674984005,
"grad_norm": 2.6080086588256286,
"learning_rate": 1.297092786761946e-05,
"loss": 1.6527,
"step": 725
},
{
"epoch": 0.46705054382597566,
"grad_norm": 2.6321954724937435,
"learning_rate": 1.2864068100857565e-05,
"loss": 1.5964,
"step": 730
},
{
"epoch": 0.47024952015355087,
"grad_norm": 2.3989213442692496,
"learning_rate": 1.2756850857234686e-05,
"loss": 1.5436,
"step": 735
},
{
"epoch": 0.473448496481126,
"grad_norm": 2.5105453091923953,
"learning_rate": 1.2649289519003739e-05,
"loss": 1.6223,
"step": 740
},
{
"epoch": 0.47664747280870123,
"grad_norm": 2.7099701621767722,
"learning_rate": 1.2541397511365584e-05,
"loss": 1.6423,
"step": 745
},
{
"epoch": 0.4798464491362764,
"grad_norm": 2.681641071296778,
"learning_rate": 1.2433188300793399e-05,
"loss": 1.6358,
"step": 750
},
{
"epoch": 0.48304542546385154,
"grad_norm": 2.594163580036891,
"learning_rate": 1.2324675393351818e-05,
"loss": 1.6422,
"step": 755
},
{
"epoch": 0.48624440179142675,
"grad_norm": 2.5737251899733486,
"learning_rate": 1.221587233301123e-05,
"loss": 1.651,
"step": 760
},
{
"epoch": 0.4894433781190019,
"grad_norm": 2.8820386294007685,
"learning_rate": 1.2106792699957264e-05,
"loss": 1.6009,
"step": 765
},
{
"epoch": 0.4926423544465771,
"grad_norm": 2.552280511541425,
"learning_rate": 1.1997450108895807e-05,
"loss": 1.6205,
"step": 770
},
{
"epoch": 0.49584133077415227,
"grad_norm": 2.641562455451703,
"learning_rate": 1.1887858207353678e-05,
"loss": 1.6006,
"step": 775
},
{
"epoch": 0.4990403071017274,
"grad_norm": 2.668113567749088,
"learning_rate": 1.1778030673975227e-05,
"loss": 1.6431,
"step": 780
},
{
"epoch": 0.5022392834293026,
"grad_norm": 2.3817328266515188,
"learning_rate": 1.166798121681505e-05,
"loss": 1.5542,
"step": 785
},
{
"epoch": 0.5054382597568778,
"grad_norm": 2.818008392997897,
"learning_rate": 1.1557723571627016e-05,
"loss": 1.5985,
"step": 790
},
{
"epoch": 0.508637236084453,
"grad_norm": 2.5108401300141217,
"learning_rate": 1.1447271500149849e-05,
"loss": 1.6126,
"step": 795
},
{
"epoch": 0.5118362124120281,
"grad_norm": 2.6986253307519843,
"learning_rate": 1.1336638788389473e-05,
"loss": 1.6063,
"step": 800
},
{
"epoch": 0.5150351887396033,
"grad_norm": 2.5657124724923515,
"learning_rate": 1.122583924489832e-05,
"loss": 1.5637,
"step": 805
},
{
"epoch": 0.5182341650671785,
"grad_norm": 2.4728800977468266,
"learning_rate": 1.1114886699051803e-05,
"loss": 1.6121,
"step": 810
},
{
"epoch": 0.5214331413947537,
"grad_norm": 2.543002609864959,
"learning_rate": 1.1003794999322246e-05,
"loss": 1.6128,
"step": 815
},
{
"epoch": 0.5246321177223289,
"grad_norm": 2.511457139025529,
"learning_rate": 1.089257801155037e-05,
"loss": 1.6028,
"step": 820
},
{
"epoch": 0.527831094049904,
"grad_norm": 2.5162086062315496,
"learning_rate": 1.0781249617214642e-05,
"loss": 1.6015,
"step": 825
},
{
"epoch": 0.5310300703774792,
"grad_norm": 2.55046430905809,
"learning_rate": 1.0669823711698668e-05,
"loss": 1.6142,
"step": 830
},
{
"epoch": 0.5342290467050543,
"grad_norm": 2.518860317454054,
"learning_rate": 1.0558314202556866e-05,
"loss": 1.594,
"step": 835
},
{
"epoch": 0.5374280230326296,
"grad_norm": 2.589751443712825,
"learning_rate": 1.0446735007778563e-05,
"loss": 1.623,
"step": 840
},
{
"epoch": 0.5406269993602048,
"grad_norm": 2.5007253774124454,
"learning_rate": 1.0335100054050877e-05,
"loss": 1.5575,
"step": 845
},
{
"epoch": 0.5438259756877799,
"grad_norm": 2.5936001548887933,
"learning_rate": 1.0223423275020431e-05,
"loss": 1.5692,
"step": 850
},
{
"epoch": 0.5470249520153551,
"grad_norm": 2.481866088130802,
"learning_rate": 1.0111718609554254e-05,
"loss": 1.6002,
"step": 855
},
{
"epoch": 0.5502239283429302,
"grad_norm": 2.821487824788201,
"learning_rate": 1e-05,
"loss": 1.5642,
"step": 860
},
{
"epoch": 0.5534229046705055,
"grad_norm": 2.5771744373052945,
"learning_rate": 9.888281390445747e-06,
"loss": 1.5128,
"step": 865
},
{
"epoch": 0.5566218809980806,
"grad_norm": 2.5278106745609614,
"learning_rate": 9.776576724979572e-06,
"loss": 1.4949,
"step": 870
},
{
"epoch": 0.5598208573256558,
"grad_norm": 2.583594185224458,
"learning_rate": 9.664899945949128e-06,
"loss": 1.6606,
"step": 875
},
{
"epoch": 0.5630198336532309,
"grad_norm": 2.5550033072779224,
"learning_rate": 9.553264992221442e-06,
"loss": 1.5574,
"step": 880
},
{
"epoch": 0.5662188099808061,
"grad_norm": 2.712024254697942,
"learning_rate": 9.441685797443138e-06,
"loss": 1.6173,
"step": 885
},
{
"epoch": 0.5694177863083814,
"grad_norm": 2.478440548372789,
"learning_rate": 9.330176288301332e-06,
"loss": 1.5935,
"step": 890
},
{
"epoch": 0.5726167626359565,
"grad_norm": 2.4956205980217865,
"learning_rate": 9.21875038278536e-06,
"loss": 1.6817,
"step": 895
},
{
"epoch": 0.5758157389635317,
"grad_norm": 2.4089458198757994,
"learning_rate": 9.107421988449632e-06,
"loss": 1.5366,
"step": 900
},
{
"epoch": 0.5790147152911068,
"grad_norm": 2.6696674248702483,
"learning_rate": 8.996205000677758e-06,
"loss": 1.5626,
"step": 905
},
{
"epoch": 0.582213691618682,
"grad_norm": 2.6682279733761263,
"learning_rate": 8.885113300948199e-06,
"loss": 1.5511,
"step": 910
},
{
"epoch": 0.5854126679462572,
"grad_norm": 2.857427668381031,
"learning_rate": 8.774160755101685e-06,
"loss": 1.5307,
"step": 915
},
{
"epoch": 0.5886116442738324,
"grad_norm": 2.5605534221291184,
"learning_rate": 8.663361211610529e-06,
"loss": 1.5169,
"step": 920
},
{
"epoch": 0.5918106206014075,
"grad_norm": 2.6595574752499327,
"learning_rate": 8.552728499850153e-06,
"loss": 1.5537,
"step": 925
},
{
"epoch": 0.5950095969289827,
"grad_norm": 2.441996186360761,
"learning_rate": 8.442276428372986e-06,
"loss": 1.6084,
"step": 930
},
{
"epoch": 0.5982085732565579,
"grad_norm": 2.6986999386462758,
"learning_rate": 8.332018783184952e-06,
"loss": 1.6059,
"step": 935
},
{
"epoch": 0.6014075495841331,
"grad_norm": 2.4337265509835384,
"learning_rate": 8.221969326024776e-06,
"loss": 1.5153,
"step": 940
},
{
"epoch": 0.6046065259117083,
"grad_norm": 2.449633886402664,
"learning_rate": 8.112141792646326e-06,
"loss": 1.6195,
"step": 945
},
{
"epoch": 0.6078055022392834,
"grad_norm": 2.6811696820797635,
"learning_rate": 8.002549891104196e-06,
"loss": 1.5391,
"step": 950
},
{
"epoch": 0.6110044785668586,
"grad_norm": 2.777581392446066,
"learning_rate": 7.89320730004274e-06,
"loss": 1.5534,
"step": 955
},
{
"epoch": 0.6142034548944337,
"grad_norm": 2.5037502626670487,
"learning_rate": 7.784127666988774e-06,
"loss": 1.6004,
"step": 960
},
{
"epoch": 0.617402431222009,
"grad_norm": 2.467142433296175,
"learning_rate": 7.675324606648187e-06,
"loss": 1.4984,
"step": 965
},
{
"epoch": 0.6206014075495841,
"grad_norm": 2.7326282047339685,
"learning_rate": 7.566811699206604e-06,
"loss": 1.5654,
"step": 970
},
{
"epoch": 0.6238003838771593,
"grad_norm": 2.585347799703874,
"learning_rate": 7.458602488634416e-06,
"loss": 1.5479,
"step": 975
},
{
"epoch": 0.6269993602047345,
"grad_norm": 2.6646634487527305,
"learning_rate": 7.350710480996266e-06,
"loss": 1.5655,
"step": 980
},
{
"epoch": 0.6301983365323096,
"grad_norm": 2.7391721025093614,
"learning_rate": 7.243149142765317e-06,
"loss": 1.5602,
"step": 985
},
{
"epoch": 0.6333973128598849,
"grad_norm": 2.7654275208324273,
"learning_rate": 7.135931899142438e-06,
"loss": 1.5485,
"step": 990
},
{
"epoch": 0.63659628918746,
"grad_norm": 2.5820177212146604,
"learning_rate": 7.029072132380543e-06,
"loss": 1.5623,
"step": 995
},
{
"epoch": 0.6397952655150352,
"grad_norm": 2.4886301143293226,
"learning_rate": 6.922583180114299e-06,
"loss": 1.5748,
"step": 1000
},
{
"epoch": 0.6429942418426103,
"grad_norm": 2.671487485285395,
"learning_rate": 6.816478333695392e-06,
"loss": 1.507,
"step": 1005
},
{
"epoch": 0.6461932181701855,
"grad_norm": 2.5145997353426597,
"learning_rate": 6.710770836533584e-06,
"loss": 1.5717,
"step": 1010
},
{
"epoch": 0.6493921944977608,
"grad_norm": 2.4504595904164534,
"learning_rate": 6.605473882443725e-06,
"loss": 1.4887,
"step": 1015
},
{
"epoch": 0.6525911708253359,
"grad_norm": 2.5730482872625786,
"learning_rate": 6.5006006139990115e-06,
"loss": 1.5854,
"step": 1020
},
{
"epoch": 0.6557901471529111,
"grad_norm": 2.639233751706901,
"learning_rate": 6.396164120890562e-06,
"loss": 1.5601,
"step": 1025
},
{
"epoch": 0.6589891234804862,
"grad_norm": 2.467716305790526,
"learning_rate": 6.292177438293665e-06,
"loss": 1.5988,
"step": 1030
},
{
"epoch": 0.6621880998080614,
"grad_norm": 2.668132725528914,
"learning_rate": 6.188653545240793e-06,
"loss": 1.5106,
"step": 1035
},
{
"epoch": 0.6653870761356366,
"grad_norm": 2.4811419339967054,
"learning_rate": 6.085605363001628e-06,
"loss": 1.5038,
"step": 1040
},
{
"epoch": 0.6685860524632118,
"grad_norm": 2.7088704790000717,
"learning_rate": 5.983045753470308e-06,
"loss": 1.5676,
"step": 1045
},
{
"epoch": 0.6717850287907869,
"grad_norm": 2.5218510067802864,
"learning_rate": 5.880987517560075e-06,
"loss": 1.4863,
"step": 1050
},
{
"epoch": 0.6749840051183621,
"grad_norm": 2.5082686107813608,
"learning_rate": 5.779443393605536e-06,
"loss": 1.5104,
"step": 1055
},
{
"epoch": 0.6781829814459372,
"grad_norm": 2.7673109834002414,
"learning_rate": 5.678426055772746e-06,
"loss": 1.5495,
"step": 1060
},
{
"epoch": 0.6813819577735125,
"grad_norm": 2.56459855705922,
"learning_rate": 5.577948112477271e-06,
"loss": 1.5748,
"step": 1065
},
{
"epoch": 0.6845809341010877,
"grad_norm": 2.5362785161188905,
"learning_rate": 5.478022104810483e-06,
"loss": 1.5208,
"step": 1070
},
{
"epoch": 0.6877799104286628,
"grad_norm": 2.5358478018436705,
"learning_rate": 5.378660504974271e-06,
"loss": 1.4861,
"step": 1075
},
{
"epoch": 0.690978886756238,
"grad_norm": 2.6305951406603523,
"learning_rate": 5.2798757147242986e-06,
"loss": 1.5838,
"step": 1080
},
{
"epoch": 0.6941778630838131,
"grad_norm": 2.9098866036161057,
"learning_rate": 5.1816800638221176e-06,
"loss": 1.5352,
"step": 1085
},
{
"epoch": 0.6973768394113884,
"grad_norm": 2.668160321681298,
"learning_rate": 5.084085808496211e-06,
"loss": 1.5426,
"step": 1090
},
{
"epoch": 0.7005758157389635,
"grad_norm": 2.5658504273904628,
"learning_rate": 4.987105129912252e-06,
"loss": 1.5071,
"step": 1095
},
{
"epoch": 0.7037747920665387,
"grad_norm": 2.922481421797977,
"learning_rate": 4.890750132652724e-06,
"loss": 1.5437,
"step": 1100
},
{
"epoch": 0.7069737683941139,
"grad_norm": 2.613942442364072,
"learning_rate": 4.795032843206091e-06,
"loss": 1.505,
"step": 1105
},
{
"epoch": 0.710172744721689,
"grad_norm": 2.928010010015714,
"learning_rate": 4.699965208465702e-06,
"loss": 1.5768,
"step": 1110
},
{
"epoch": 0.7133717210492643,
"grad_norm": 2.6380472109566746,
"learning_rate": 4.605559094238681e-06,
"loss": 1.6016,
"step": 1115
},
{
"epoch": 0.7165706973768394,
"grad_norm": 2.3906070266983828,
"learning_rate": 4.511826283764872e-06,
"loss": 1.5583,
"step": 1120
},
{
"epoch": 0.7197696737044146,
"grad_norm": 2.311463185295313,
"learning_rate": 4.418778476246132e-06,
"loss": 1.4451,
"step": 1125
},
{
"epoch": 0.7229686500319897,
"grad_norm": 2.732571114266201,
"learning_rate": 4.326427285386118e-06,
"loss": 1.5291,
"step": 1130
},
{
"epoch": 0.7261676263595649,
"grad_norm": 2.47052905361262,
"learning_rate": 4.234784237940705e-06,
"loss": 1.4847,
"step": 1135
},
{
"epoch": 0.7293666026871402,
"grad_norm": 2.6063156623195893,
"learning_rate": 4.143860772279287e-06,
"loss": 1.5131,
"step": 1140
},
{
"epoch": 0.7325655790147153,
"grad_norm": 2.7370913449212444,
"learning_rate": 4.053668236957135e-06,
"loss": 1.5475,
"step": 1145
},
{
"epoch": 0.7357645553422905,
"grad_norm": 2.5669442133096925,
"learning_rate": 3.964217889298882e-06,
"loss": 1.5196,
"step": 1150
},
{
"epoch": 0.7389635316698656,
"grad_norm": 2.507377481660436,
"learning_rate": 3.875520893993482e-06,
"loss": 1.5489,
"step": 1155
},
{
"epoch": 0.7421625079974408,
"grad_norm": 2.528603671848079,
"learning_rate": 3.787588321700697e-06,
"loss": 1.5712,
"step": 1160
},
{
"epoch": 0.745361484325016,
"grad_norm": 2.509655655176294,
"learning_rate": 3.7004311476692977e-06,
"loss": 1.5489,
"step": 1165
},
{
"epoch": 0.7485604606525912,
"grad_norm": 2.5144058197876147,
"learning_rate": 3.6140602503672217e-06,
"loss": 1.5288,
"step": 1170
},
{
"epoch": 0.7517594369801663,
"grad_norm": 2.607098479339003,
"learning_rate": 3.528486410123758e-06,
"loss": 1.4888,
"step": 1175
},
{
"epoch": 0.7549584133077415,
"grad_norm": 2.5874268429840988,
"learning_rate": 3.443720307784022e-06,
"loss": 1.5054,
"step": 1180
},
{
"epoch": 0.7581573896353166,
"grad_norm": 2.7008730474734772,
"learning_rate": 3.359772523375837e-06,
"loss": 1.5543,
"step": 1185
},
{
"epoch": 0.7613563659628919,
"grad_norm": 2.7184986676140968,
"learning_rate": 3.276653534789185e-06,
"loss": 1.5286,
"step": 1190
},
{
"epoch": 0.7645553422904671,
"grad_norm": 2.5056748056086335,
"learning_rate": 3.1943737164684094e-06,
"loss": 1.5233,
"step": 1195
},
{
"epoch": 0.7677543186180422,
"grad_norm": 2.556577305103137,
"learning_rate": 3.11294333811736e-06,
"loss": 1.457,
"step": 1200
},
{
"epoch": 0.7709532949456174,
"grad_norm": 2.483199441635862,
"learning_rate": 3.032372563417556e-06,
"loss": 1.5272,
"step": 1205
},
{
"epoch": 0.7741522712731925,
"grad_norm": 2.8562065290992216,
"learning_rate": 2.952671448759631e-06,
"loss": 1.522,
"step": 1210
},
{
"epoch": 0.7773512476007678,
"grad_norm": 2.551521601442759,
"learning_rate": 2.8738499419881517e-06,
"loss": 1.5306,
"step": 1215
},
{
"epoch": 0.780550223928343,
"grad_norm": 2.8147572671341754,
"learning_rate": 2.795917881159973e-06,
"loss": 1.4583,
"step": 1220
},
{
"epoch": 0.7837492002559181,
"grad_norm": 2.614247349699639,
"learning_rate": 2.718884993316311e-06,
"loss": 1.5314,
"step": 1225
},
{
"epoch": 0.7869481765834933,
"grad_norm": 2.6225396269815153,
"learning_rate": 2.642760893268684e-06,
"loss": 1.5035,
"step": 1230
},
{
"epoch": 0.7901471529110684,
"grad_norm": 2.5615607917045478,
"learning_rate": 2.5675550823988184e-06,
"loss": 1.513,
"step": 1235
},
{
"epoch": 0.7933461292386437,
"grad_norm": 2.7212788943352466,
"learning_rate": 2.493276947472756e-06,
"loss": 1.5181,
"step": 1240
},
{
"epoch": 0.7965451055662188,
"grad_norm": 2.6432604071155352,
"learning_rate": 2.4199357594692564e-06,
"loss": 1.4582,
"step": 1245
},
{
"epoch": 0.799744081893794,
"grad_norm": 2.720800985424245,
"learning_rate": 2.347540672422627e-06,
"loss": 1.4561,
"step": 1250
},
{
"epoch": 0.8029430582213691,
"grad_norm": 2.5092893195016805,
"learning_rate": 2.2761007222801866e-06,
"loss": 1.4987,
"step": 1255
},
{
"epoch": 0.8061420345489443,
"grad_norm": 2.676468357399732,
"learning_rate": 2.2056248257744383e-06,
"loss": 1.496,
"step": 1260
},
{
"epoch": 0.8093410108765196,
"grad_norm": 2.489080002899454,
"learning_rate": 2.136121779310132e-06,
"loss": 1.5072,
"step": 1265
},
{
"epoch": 0.8125399872040947,
"grad_norm": 2.697882739125852,
"learning_rate": 2.067600257866358e-06,
"loss": 1.5101,
"step": 1270
},
{
"epoch": 0.8157389635316699,
"grad_norm": 2.5898654023108625,
"learning_rate": 2.000068813913777e-06,
"loss": 1.5069,
"step": 1275
},
{
"epoch": 0.818937939859245,
"grad_norm": 2.517410602087706,
"learning_rate": 1.933535876347141e-06,
"loss": 1.5271,
"step": 1280
},
{
"epoch": 0.8221369161868202,
"grad_norm": 2.6904620614604426,
"learning_rate": 1.8680097494332682e-06,
"loss": 1.5001,
"step": 1285
},
{
"epoch": 0.8253358925143954,
"grad_norm": 2.404028412582033,
"learning_rate": 1.8034986117745245e-06,
"loss": 1.4672,
"step": 1290
},
{
"epoch": 0.8285348688419706,
"grad_norm": 2.7399471915387403,
"learning_rate": 1.7400105152880331e-06,
"loss": 1.484,
"step": 1295
},
{
"epoch": 0.8317338451695457,
"grad_norm": 2.4143809245788383,
"learning_rate": 1.6775533842006853e-06,
"loss": 1.593,
"step": 1300
},
{
"epoch": 0.8349328214971209,
"grad_norm": 2.4193698761219955,
"learning_rate": 1.6161350140600606e-06,
"loss": 1.4811,
"step": 1305
},
{
"epoch": 0.838131797824696,
"grad_norm": 2.597516158887157,
"learning_rate": 1.555763070761448e-06,
"loss": 1.546,
"step": 1310
},
{
"epoch": 0.8413307741522713,
"grad_norm": 2.462136896599469,
"learning_rate": 1.496445089591031e-06,
"loss": 1.5216,
"step": 1315
},
{
"epoch": 0.8445297504798465,
"grad_norm": 2.924679107559782,
"learning_rate": 1.4381884742853723e-06,
"loss": 1.5172,
"step": 1320
},
{
"epoch": 0.8477287268074216,
"grad_norm": 2.752800707578496,
"learning_rate": 1.381000496107313e-06,
"loss": 1.4762,
"step": 1325
},
{
"epoch": 0.8509277031349968,
"grad_norm": 2.7130328623372146,
"learning_rate": 1.3248882929384321e-06,
"loss": 1.4436,
"step": 1330
},
{
"epoch": 0.8541266794625719,
"grad_norm": 2.6217187777983315,
"learning_rate": 1.2698588683881185e-06,
"loss": 1.5213,
"step": 1335
},
{
"epoch": 0.8573256557901472,
"grad_norm": 2.7790023166898634,
"learning_rate": 1.215919090919434e-06,
"loss": 1.5587,
"step": 1340
},
{
"epoch": 0.8605246321177223,
"grad_norm": 2.750612332111246,
"learning_rate": 1.1630756929918218e-06,
"loss": 1.3674,
"step": 1345
},
{
"epoch": 0.8637236084452975,
"grad_norm": 2.7026209229203384,
"learning_rate": 1.111335270220798e-06,
"loss": 1.4509,
"step": 1350
},
{
"epoch": 0.8669225847728727,
"grad_norm": 2.464852660357827,
"learning_rate": 1.060704280554733e-06,
"loss": 1.4899,
"step": 1355
},
{
"epoch": 0.8701215611004478,
"grad_norm": 2.459276773810485,
"learning_rate": 1.0111890434688009e-06,
"loss": 1.5276,
"step": 1360
},
{
"epoch": 0.8733205374280231,
"grad_norm": 2.8059192316793387,
"learning_rate": 9.62795739176212e-07,
"loss": 1.5918,
"step": 1365
},
{
"epoch": 0.8765195137555982,
"grad_norm": 2.6222241793711953,
"learning_rate": 9.155304078568495e-07,
"loss": 1.4414,
"step": 1370
},
{
"epoch": 0.8797184900831734,
"grad_norm": 2.576035857811823,
"learning_rate": 8.693989489033494e-07,
"loss": 1.4072,
"step": 1375
},
{
"epoch": 0.8829174664107485,
"grad_norm": 2.46854082888577,
"learning_rate": 8.244071201847826e-07,
"loss": 1.4557,
"step": 1380
},
{
"epoch": 0.8861164427383237,
"grad_norm": 2.455376594334234,
"learning_rate": 7.805605373279934e-07,
"loss": 1.478,
"step": 1385
},
{
"epoch": 0.889315419065899,
"grad_norm": 2.4975995164593505,
"learning_rate": 7.378646730166783e-07,
"loss": 1.4811,
"step": 1390
},
{
"epoch": 0.8925143953934741,
"grad_norm": 2.8006479473707113,
"learning_rate": 6.963248563083203e-07,
"loss": 1.4991,
"step": 1395
},
{
"epoch": 0.8957133717210493,
"grad_norm": 2.4482440589533563,
"learning_rate": 6.559462719690501e-07,
"loss": 1.4783,
"step": 1400
},
{
"epoch": 0.8989123480486244,
"grad_norm": 2.631924407331454,
"learning_rate": 6.167339598265109e-07,
"loss": 1.4827,
"step": 1405
},
{
"epoch": 0.9021113243761996,
"grad_norm": 3.0238656032655697,
"learning_rate": 5.78692814140811e-07,
"loss": 1.4484,
"step": 1410
},
{
"epoch": 0.9053103007037748,
"grad_norm": 2.4776963350793313,
"learning_rate": 5.418275829936537e-07,
"loss": 1.4801,
"step": 1415
},
{
"epoch": 0.90850927703135,
"grad_norm": 2.8527681506248013,
"learning_rate": 5.06142867695708e-07,
"loss": 1.468,
"step": 1420
},
{
"epoch": 0.9117082533589251,
"grad_norm": 2.612152611424273,
"learning_rate": 4.716431222122997e-07,
"loss": 1.4943,
"step": 1425
},
{
"epoch": 0.9149072296865003,
"grad_norm": 2.626107563673812,
"learning_rate": 4.3833265260749157e-07,
"loss": 1.4963,
"step": 1430
},
{
"epoch": 0.9181062060140754,
"grad_norm": 2.7357100972135204,
"learning_rate": 4.062156165066211e-07,
"loss": 1.5118,
"step": 1435
},
{
"epoch": 0.9213051823416507,
"grad_norm": 2.6579207889627567,
"learning_rate": 3.752960225773772e-07,
"loss": 1.4694,
"step": 1440
},
{
"epoch": 0.9245041586692259,
"grad_norm": 2.504449626402816,
"learning_rate": 3.4557773002945607e-07,
"loss": 1.4499,
"step": 1445
},
{
"epoch": 0.927703134996801,
"grad_norm": 2.7042323207994503,
"learning_rate": 3.170644481328711e-07,
"loss": 1.4805,
"step": 1450
},
{
"epoch": 0.9309021113243762,
"grad_norm": 2.83226432783713,
"learning_rate": 2.8975973575499526e-07,
"loss": 1.4986,
"step": 1455
},
{
"epoch": 0.9341010876519513,
"grad_norm": 2.5155980665060214,
"learning_rate": 2.636670009163522e-07,
"loss": 1.513,
"step": 1460
},
{
"epoch": 0.9373000639795266,
"grad_norm": 2.51434841790346,
"learning_rate": 2.3878950036524963e-07,
"loss": 1.3954,
"step": 1465
},
{
"epoch": 0.9404990403071017,
"grad_norm": 2.7043407811605205,
"learning_rate": 2.1513033917129334e-07,
"loss": 1.4829,
"step": 1470
},
{
"epoch": 0.9436980166346769,
"grad_norm": 2.7563171838390126,
"learning_rate": 1.9269247033782744e-07,
"loss": 1.4819,
"step": 1475
},
{
"epoch": 0.946896992962252,
"grad_norm": 2.8155182037170388,
"learning_rate": 1.7147869443335463e-07,
"loss": 1.4867,
"step": 1480
},
{
"epoch": 0.9500959692898272,
"grad_norm": 2.595936373884088,
"learning_rate": 1.5149165924199016e-07,
"loss": 1.4706,
"step": 1485
},
{
"epoch": 0.9532949456174025,
"grad_norm": 2.6688211843514273,
"learning_rate": 1.3273385943297746e-07,
"loss": 1.4533,
"step": 1490
},
{
"epoch": 0.9564939219449776,
"grad_norm": 2.7101964309089936,
"learning_rate": 1.1520763624931597e-07,
"loss": 1.6145,
"step": 1495
},
{
"epoch": 0.9596928982725528,
"grad_norm": 2.8635708781856883,
"learning_rate": 9.891517721554499e-08,
"loss": 1.4371,
"step": 1500
},
{
"epoch": 0.9628918746001279,
"grad_norm": 2.6919049595193933,
"learning_rate": 8.385851586470318e-08,
"loss": 1.4319,
"step": 1505
},
{
"epoch": 0.9660908509277031,
"grad_norm": 2.7599573530698596,
"learning_rate": 7.003953148452036e-08,
"loss": 1.5301,
"step": 1510
},
{
"epoch": 0.9692898272552783,
"grad_norm": 2.7397148500707815,
"learning_rate": 5.745994888285311e-08,
"loss": 1.4982,
"step": 1515
},
{
"epoch": 0.9724888035828535,
"grad_norm": 2.6249436716593704,
"learning_rate": 4.612133817239905e-08,
"loss": 1.4854,
"step": 1520
},
{
"epoch": 0.9756877799104287,
"grad_norm": 2.6820453755780997,
"learning_rate": 3.602511457473479e-08,
"loss": 1.4968,
"step": 1525
},
{
"epoch": 0.9788867562380038,
"grad_norm": 2.47834251866859,
"learning_rate": 2.7172538243666057e-08,
"loss": 1.4325,
"step": 1530
},
{
"epoch": 0.982085732565579,
"grad_norm": 2.4886629292169444,
"learning_rate": 1.9564714107945804e-08,
"loss": 1.5541,
"step": 1535
},
{
"epoch": 0.9852847088931542,
"grad_norm": 2.506082115077142,
"learning_rate": 1.3202591733365577e-08,
"loss": 1.4393,
"step": 1540
},
{
"epoch": 0.9884836852207294,
"grad_norm": 2.6014464272205866,
"learning_rate": 8.086965204233688e-09,
"loss": 1.5181,
"step": 1545
},
{
"epoch": 0.9916826615483045,
"grad_norm": 2.6809918810591213,
"learning_rate": 4.218473024261149e-09,
"loss": 1.5258,
"step": 1550
},
{
"epoch": 0.9948816378758797,
"grad_norm": 2.673588199710328,
"learning_rate": 1.5975980368709843e-09,
"loss": 1.4306,
"step": 1555
},
{
"epoch": 0.9980806142034548,
"grad_norm": 2.619551584330537,
"learning_rate": 2.2466736492643416e-10,
"loss": 1.4171,
"step": 1560
},
{
"epoch": 1.0,
"eval_loss": 1.525302767753601,
"eval_runtime": 27.2554,
"eval_samples_per_second": 18.345,
"eval_steps_per_second": 0.587,
"step": 1563
},
{
"epoch": 1.0,
"step": 1563,
"total_flos": 40228787421184.0,
"train_loss": 1.600957996328138,
"train_runtime": 9121.6421,
"train_samples_per_second": 5.481,
"train_steps_per_second": 0.171
}
],
"logging_steps": 5,
"max_steps": 1563,
"num_input_tokens_seen": 0,
"num_train_epochs": 1,
"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": 40228787421184.0,
"train_batch_size": 8,
"trial_name": null,
"trial_params": null
}