{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.9990375360923965,
"eval_steps": 500,
"global_step": 519,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0019249278152069298,
"grad_norm": 166.78744506835938,
"learning_rate": 5.769230769230769e-06,
"loss": 32.2268,
"step": 1
},
{
"epoch": 0.009624639076034648,
"grad_norm": 132.89913940429688,
"learning_rate": 2.8846153846153845e-05,
"loss": 29.726,
"step": 5
},
{
"epoch": 0.019249278152069296,
"grad_norm": 64.4798812866211,
"learning_rate": 5.769230769230769e-05,
"loss": 25.1178,
"step": 10
},
{
"epoch": 0.028873917228103944,
"grad_norm": 12.633763313293457,
"learning_rate": 8.653846153846152e-05,
"loss": 19.6514,
"step": 15
},
{
"epoch": 0.03849855630413859,
"grad_norm": 9.107071876525879,
"learning_rate": 0.00011538461538461538,
"loss": 17.2219,
"step": 20
},
{
"epoch": 0.04812319538017324,
"grad_norm": 4.696881294250488,
"learning_rate": 0.00014423076923076922,
"loss": 15.4404,
"step": 25
},
{
"epoch": 0.05774783445620789,
"grad_norm": 3.118898868560791,
"learning_rate": 0.00017307692307692304,
"loss": 14.695,
"step": 30
},
{
"epoch": 0.06737247353224254,
"grad_norm": 4.403642654418945,
"learning_rate": 0.00020192307692307691,
"loss": 13.7845,
"step": 35
},
{
"epoch": 0.07699711260827719,
"grad_norm": 7.105109214782715,
"learning_rate": 0.00023076923076923076,
"loss": 12.8307,
"step": 40
},
{
"epoch": 0.08662175168431184,
"grad_norm": 16.107545852661133,
"learning_rate": 0.0002596153846153846,
"loss": 10.7864,
"step": 45
},
{
"epoch": 0.09624639076034648,
"grad_norm": 24.628265380859375,
"learning_rate": 0.00028846153846153843,
"loss": 7.478,
"step": 50
},
{
"epoch": 0.10587102983638114,
"grad_norm": 8.615938186645508,
"learning_rate": 0.00029996945395293625,
"loss": 3.5198,
"step": 55
},
{
"epoch": 0.11549566891241578,
"grad_norm": 5.187551021575928,
"learning_rate": 0.0002997828287165724,
"loss": 2.5417,
"step": 60
},
{
"epoch": 0.12512030798845045,
"grad_norm": 2.919438362121582,
"learning_rate": 0.00029942675913693153,
"loss": 2.0943,
"step": 65
},
{
"epoch": 0.1347449470644851,
"grad_norm": 1.5105671882629395,
"learning_rate": 0.0002989016480237121,
"loss": 1.8019,
"step": 70
},
{
"epoch": 0.14436958614051973,
"grad_norm": 0.6488280892372131,
"learning_rate": 0.0002982080894176895,
"loss": 1.6616,
"step": 75
},
{
"epoch": 0.15399422521655437,
"grad_norm": 1.0447803735733032,
"learning_rate": 0.0002973468679186978,
"loss": 1.5605,
"step": 80
},
{
"epoch": 0.16361886429258904,
"grad_norm": 0.5504654049873352,
"learning_rate": 0.0002963189577980372,
"loss": 1.5028,
"step": 85
},
{
"epoch": 0.17324350336862368,
"grad_norm": 0.7152052521705627,
"learning_rate": 0.0002951255218963139,
"loss": 1.4363,
"step": 90
},
{
"epoch": 0.18286814244465832,
"grad_norm": 1.1773169040679932,
"learning_rate": 0.0002937679103079571,
"loss": 1.4284,
"step": 95
},
{
"epoch": 0.19249278152069296,
"grad_norm": 0.908633828163147,
"learning_rate": 0.00029224765885390143,
"loss": 1.3772,
"step": 100
},
{
"epoch": 0.20211742059672763,
"grad_norm": 0.9546915292739868,
"learning_rate": 0.0002905664873441643,
"loss": 1.3789,
"step": 105
},
{
"epoch": 0.21174205967276227,
"grad_norm": 0.6474579572677612,
"learning_rate": 0.00028872629763228145,
"loss": 1.3446,
"step": 110
},
{
"epoch": 0.22136669874879691,
"grad_norm": 0.6460303068161011,
"learning_rate": 0.0002867291714638035,
"loss": 1.3102,
"step": 115
},
{
"epoch": 0.23099133782483156,
"grad_norm": 0.8272395730018616,
"learning_rate": 0.0002845773681212862,
"loss": 1.3315,
"step": 120
},
{
"epoch": 0.24061597690086622,
"grad_norm": 1.1858478784561157,
"learning_rate": 0.00028227332186843884,
"loss": 1.3232,
"step": 125
},
{
"epoch": 0.2502406159769009,
"grad_norm": 1.020283579826355,
"learning_rate": 0.0002798196391963229,
"loss": 1.298,
"step": 130
},
{
"epoch": 0.2598652550529355,
"grad_norm": 1.3490053415298462,
"learning_rate": 0.0002772190958747147,
"loss": 1.3026,
"step": 135
},
{
"epoch": 0.2694898941289702,
"grad_norm": 1.1797901391983032,
"learning_rate": 0.00027447463381196973,
"loss": 1.2889,
"step": 140
},
{
"epoch": 0.2791145332050048,
"grad_norm": 1.916670799255371,
"learning_rate": 0.0002715893577269389,
"loss": 1.2773,
"step": 145
},
{
"epoch": 0.28873917228103946,
"grad_norm": 0.5742417573928833,
"learning_rate": 0.0002685665316367035,
"loss": 1.2937,
"step": 150
},
{
"epoch": 0.2983638113570741,
"grad_norm": 0.6991013884544373,
"learning_rate": 0.0002654095751641007,
"loss": 1.2667,
"step": 155
},
{
"epoch": 0.30798845043310874,
"grad_norm": 1.23212468624115,
"learning_rate": 0.00026212205966921786,
"loss": 1.2571,
"step": 160
},
{
"epoch": 0.3176130895091434,
"grad_norm": 0.8666950464248657,
"learning_rate": 0.0002587077042092314,
"loss": 1.2649,
"step": 165
},
{
"epoch": 0.3272377285851781,
"grad_norm": 0.945165753364563,
"learning_rate": 0.00025517037133116085,
"loss": 1.2262,
"step": 170
},
{
"epoch": 0.3368623676612127,
"grad_norm": 0.9452968835830688,
"learning_rate": 0.0002515140627022976,
"loss": 1.2332,
"step": 175
},
{
"epoch": 0.34648700673724736,
"grad_norm": 0.7168753743171692,
"learning_rate": 0.00024774291458325127,
"loss": 1.2303,
"step": 180
},
{
"epoch": 0.35611164581328203,
"grad_norm": 1.55524742603302,
"learning_rate": 0.00024386119314873578,
"loss": 1.2258,
"step": 185
},
{
"epoch": 0.36573628488931664,
"grad_norm": 0.549445629119873,
"learning_rate": 0.00023987328966138704,
"loss": 1.2098,
"step": 190
},
{
"epoch": 0.3753609239653513,
"grad_norm": 0.6080393195152283,
"learning_rate": 0.00023578371550407354,
"loss": 1.2385,
"step": 195
},
{
"epoch": 0.3849855630413859,
"grad_norm": 2.411741256713867,
"learning_rate": 0.0002315970970763186,
"loss": 1.2144,
"step": 200
},
{
"epoch": 0.3946102021174206,
"grad_norm": 1.1420116424560547,
"learning_rate": 0.00022731817056060802,
"loss": 1.1943,
"step": 205
},
{
"epoch": 0.40423484119345526,
"grad_norm": 1.1481698751449585,
"learning_rate": 0.00022295177656450404,
"loss": 1.2004,
"step": 210
},
{
"epoch": 0.4138594802694899,
"grad_norm": 3.898259401321411,
"learning_rate": 0.00021850285464462677,
"loss": 1.2195,
"step": 215
},
{
"epoch": 0.42348411934552455,
"grad_norm": 0.6957814693450928,
"learning_rate": 0.0002139764377186976,
"loss": 1.2159,
"step": 220
},
{
"epoch": 0.4331087584215592,
"grad_norm": 0.9523270726203918,
"learning_rate": 0.00020937764637196638,
"loss": 1.186,
"step": 225
},
{
"epoch": 0.44273339749759383,
"grad_norm": 1.081147313117981,
"learning_rate": 0.00020471168306446336,
"loss": 1.2156,
"step": 230
},
{
"epoch": 0.4523580365736285,
"grad_norm": 2.2060320377349854,
"learning_rate": 0.0001999838262456287,
"loss": 1.1998,
"step": 235
},
{
"epoch": 0.4619826756496631,
"grad_norm": 1.1269886493682861,
"learning_rate": 0.0001951994243829781,
"loss": 1.2017,
"step": 240
},
{
"epoch": 0.4716073147256978,
"grad_norm": 1.608312964439392,
"learning_rate": 0.00019036388991155846,
"loss": 1.1873,
"step": 245
},
{
"epoch": 0.48123195380173245,
"grad_norm": 1.0753167867660522,
"learning_rate": 0.0001854826931110403,
"loss": 1.1984,
"step": 250
},
{
"epoch": 0.49085659287776706,
"grad_norm": 1.3986575603485107,
"learning_rate": 0.0001805613559173714,
"loss": 1.181,
"step": 255
},
{
"epoch": 0.5004812319538018,
"grad_norm": 0.47483909130096436,
"learning_rate": 0.0001756054456759944,
"loss": 1.173,
"step": 260
},
{
"epoch": 0.5101058710298364,
"grad_norm": 0.8258511424064636,
"learning_rate": 0.00017062056884369325,
"loss": 1.1736,
"step": 265
},
{
"epoch": 0.519730510105871,
"grad_norm": 0.7987772822380066,
"learning_rate": 0.0001656123646461951,
"loss": 1.1616,
"step": 270
},
{
"epoch": 0.5293551491819056,
"grad_norm": 0.7245559692382812,
"learning_rate": 0.00016058649869870098,
"loss": 1.1846,
"step": 275
},
{
"epoch": 0.5389797882579404,
"grad_norm": 1.228034496307373,
"learning_rate": 0.00015554865659656367,
"loss": 1.165,
"step": 280
},
{
"epoch": 0.548604427333975,
"grad_norm": 0.8788666725158691,
"learning_rate": 0.00015050453748336224,
"loss": 1.1707,
"step": 285
},
{
"epoch": 0.5582290664100096,
"grad_norm": 0.7828085422515869,
"learning_rate": 0.00014545984760365,
"loss": 1.1609,
"step": 290
},
{
"epoch": 0.5678537054860443,
"grad_norm": 0.7824317216873169,
"learning_rate": 0.00014042029384766938,
"loss": 1.1656,
"step": 295
},
{
"epoch": 0.5774783445620789,
"grad_norm": 0.9950030446052551,
"learning_rate": 0.00013539157729533678,
"loss": 1.1626,
"step": 300
},
{
"epoch": 0.5871029836381135,
"grad_norm": 1.1628440618515015,
"learning_rate": 0.00013037938676679957,
"loss": 1.1633,
"step": 305
},
{
"epoch": 0.5967276227141483,
"grad_norm": 0.7875819802284241,
"learning_rate": 0.00012538939238686286,
"loss": 1.1721,
"step": 310
},
{
"epoch": 0.6063522617901829,
"grad_norm": 1.3848762512207031,
"learning_rate": 0.0001204272391705654,
"loss": 1.1667,
"step": 315
},
{
"epoch": 0.6159769008662175,
"grad_norm": 0.8932669162750244,
"learning_rate": 0.00011549854063716169,
"loss": 1.1761,
"step": 320
},
{
"epoch": 0.6256015399422522,
"grad_norm": 0.7711289525032043,
"learning_rate": 0.00011060887245973355,
"loss": 1.1694,
"step": 325
},
{
"epoch": 0.6352261790182868,
"grad_norm": 2.6940572261810303,
"learning_rate": 0.00010576376615761647,
"loss": 1.1535,
"step": 330
},
{
"epoch": 0.6448508180943214,
"grad_norm": 0.9238442778587341,
"learning_rate": 0.00010096870283877523,
"loss": 1.1737,
"step": 335
},
{
"epoch": 0.6544754571703562,
"grad_norm": 1.135506510734558,
"learning_rate": 9.62291069992085e-05,
"loss": 1.1513,
"step": 340
},
{
"epoch": 0.6641000962463908,
"grad_norm": 0.5180495381355286,
"learning_rate": 9.155034038639637e-05,
"loss": 1.1538,
"step": 345
},
{
"epoch": 0.6737247353224254,
"grad_norm": 0.5448501706123352,
"learning_rate": 8.693769593373337e-05,
"loss": 1.1342,
"step": 350
},
{
"epoch": 0.6833493743984601,
"grad_norm": 0.47162362933158875,
"learning_rate": 8.239639177280888e-05,
"loss": 1.1429,
"step": 355
},
{
"epoch": 0.6929740134744947,
"grad_norm": 0.599886417388916,
"learning_rate": 7.793156533030761e-05,
"loss": 1.1604,
"step": 360
},
{
"epoch": 0.7025986525505293,
"grad_norm": 0.5403456687927246,
"learning_rate": 7.354826751620954e-05,
"loss": 1.1378,
"step": 365
},
{
"epoch": 0.7122232916265641,
"grad_norm": 0.8314358592033386,
"learning_rate": 6.925145700986301e-05,
"loss": 1.1433,
"step": 370
},
{
"epoch": 0.7218479307025987,
"grad_norm": 1.6931346654891968,
"learning_rate": 6.504599465039542e-05,
"loss": 1.1502,
"step": 375
},
{
"epoch": 0.7314725697786333,
"grad_norm": 0.534213125705719,
"learning_rate": 6.093663793780725e-05,
"loss": 1.1438,
"step": 380
},
{
"epoch": 0.7410972088546679,
"grad_norm": 0.9398422837257385,
"learning_rate": 5.692803565096988e-05,
"loss": 1.1542,
"step": 385
},
{
"epoch": 0.7507218479307026,
"grad_norm": 1.0625075101852417,
"learning_rate": 5.302472258861687e-05,
"loss": 1.1346,
"step": 390
},
{
"epoch": 0.7603464870067372,
"grad_norm": 0.48856019973754883,
"learning_rate": 4.923111443927615e-05,
"loss": 1.1385,
"step": 395
},
{
"epoch": 0.7699711260827719,
"grad_norm": 0.4903761148452759,
"learning_rate": 4.5551502785948405e-05,
"loss": 1.1556,
"step": 400
},
{
"epoch": 0.7795957651588066,
"grad_norm": 0.6831708550453186,
"learning_rate": 4.199005025118158e-05,
"loss": 1.1529,
"step": 405
},
{
"epoch": 0.7892204042348412,
"grad_norm": 0.4595310688018799,
"learning_rate": 3.855078578803424e-05,
"loss": 1.1267,
"step": 410
},
{
"epoch": 0.7988450433108758,
"grad_norm": 0.652646005153656,
"learning_rate": 3.5237600122254437e-05,
"loss": 1.1412,
"step": 415
},
{
"epoch": 0.8084696823869105,
"grad_norm": 0.6624627113342285,
"learning_rate": 3.2054241350831046e-05,
"loss": 1.1323,
"step": 420
},
{
"epoch": 0.8180943214629451,
"grad_norm": 0.4390796720981598,
"learning_rate": 2.9004310701895837e-05,
"loss": 1.13,
"step": 425
},
{
"epoch": 0.8277189605389798,
"grad_norm": 0.9988071918487549,
"learning_rate": 2.6091258460773862e-05,
"loss": 1.1343,
"step": 430
},
{
"epoch": 0.8373435996150145,
"grad_norm": 0.481611430644989,
"learning_rate": 2.3318380066789787e-05,
"loss": 1.1217,
"step": 435
},
{
"epoch": 0.8469682386910491,
"grad_norm": 0.49289077520370483,
"learning_rate": 2.0688812385247176e-05,
"loss": 1.1299,
"step": 440
},
{
"epoch": 0.8565928777670837,
"grad_norm": 0.5225491523742676,
"learning_rate": 1.8205530158796505e-05,
"loss": 1.119,
"step": 445
},
{
"epoch": 0.8662175168431184,
"grad_norm": 0.5030360817909241,
"learning_rate": 1.587134264220778e-05,
"loss": 1.1305,
"step": 450
},
{
"epoch": 0.875842155919153,
"grad_norm": 0.4824385643005371,
"learning_rate": 1.3688890424353726e-05,
"loss": 1.1192,
"step": 455
},
{
"epoch": 0.8854667949951877,
"grad_norm": 0.5466002821922302,
"learning_rate": 1.1660642440999196e-05,
"loss": 1.1396,
"step": 460
},
{
"epoch": 0.8950914340712224,
"grad_norm": 0.5023711323738098,
"learning_rate": 9.788893181776297e-06,
"loss": 1.1369,
"step": 465
},
{
"epoch": 0.904716073147257,
"grad_norm": 0.4702525734901428,
"learning_rate": 8.07576009450408e-06,
"loss": 1.1296,
"step": 470
},
{
"epoch": 0.9143407122232916,
"grad_norm": 0.47570234537124634,
"learning_rate": 6.5231811897903714e-06,
"loss": 1.1365,
"step": 475
},
{
"epoch": 0.9239653512993262,
"grad_norm": 0.4453733265399933,
"learning_rate": 5.13291284862452e-06,
"loss": 1.1219,
"step": 480
},
{
"epoch": 0.933589990375361,
"grad_norm": 0.4282439053058624,
"learning_rate": 3.906527835442064e-06,
"loss": 1.1193,
"step": 485
},
{
"epoch": 0.9432146294513956,
"grad_norm": 0.4609365165233612,
"learning_rate": 2.8454135189082684e-06,
"loss": 1.145,
"step": 490
},
{
"epoch": 0.9528392685274302,
"grad_norm": 0.46295586228370667,
"learning_rate": 1.950770302434157e-06,
"loss": 1.1418,
"step": 495
},
{
"epoch": 0.9624639076034649,
"grad_norm": 0.4474254250526428,
"learning_rate": 1.223610266200009e-06,
"loss": 1.1285,
"step": 500
},
{
"epoch": 0.9720885466794995,
"grad_norm": 0.5148175358772278,
"learning_rate": 6.647560222224957e-07,
"loss": 1.1177,
"step": 505
},
{
"epoch": 0.9817131857555341,
"grad_norm": 0.4380834400653839,
"learning_rate": 2.748397837611105e-07,
"loss": 1.14,
"step": 510
},
{
"epoch": 0.9913378248315688,
"grad_norm": 0.5519540309906006,
"learning_rate": 5.430265011625579e-08,
"loss": 1.1207,
"step": 515
},
{
"epoch": 0.9990375360923965,
"eval_loss": 2.2841131687164307,
"eval_runtime": 1.3512,
"eval_samples_per_second": 4.441,
"eval_steps_per_second": 0.74,
"step": 519
},
{
"epoch": 0.9990375360923965,
"step": 519,
"total_flos": 7.912619964723364e+17,
"train_loss": 2.7495121248425316,
"train_runtime": 6850.0708,
"train_samples_per_second": 2.426,
"train_steps_per_second": 0.076
}
],
"logging_steps": 5,
"max_steps": 519,
"num_input_tokens_seen": 0,
"num_train_epochs": 1,
"save_steps": 100,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 7.912619964723364e+17,
"train_batch_size": 4,
"trial_name": null,
"trial_params": null
}