terry69's picture
Model save
449b1e1 verified
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.9994683678894205,
"eval_steps": 500,
"global_step": 470,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.002126528442317916,
"grad_norm": 1.1132217598089287,
"learning_rate": 4.255319148936171e-06,
"loss": 1.475,
"step": 1
},
{
"epoch": 0.01063264221158958,
"grad_norm": 1.0674803445954786,
"learning_rate": 2.1276595744680852e-05,
"loss": 1.3986,
"step": 5
},
{
"epoch": 0.02126528442317916,
"grad_norm": 0.27714288812132803,
"learning_rate": 4.2553191489361704e-05,
"loss": 1.3198,
"step": 10
},
{
"epoch": 0.03189792663476874,
"grad_norm": 0.24323884897635753,
"learning_rate": 6.382978723404256e-05,
"loss": 1.3201,
"step": 15
},
{
"epoch": 0.04253056884635832,
"grad_norm": 0.17616144014740132,
"learning_rate": 8.510638297872341e-05,
"loss": 1.2681,
"step": 20
},
{
"epoch": 0.0531632110579479,
"grad_norm": 0.14425961839274853,
"learning_rate": 0.00010638297872340425,
"loss": 1.2294,
"step": 25
},
{
"epoch": 0.06379585326953748,
"grad_norm": 0.09993563340523522,
"learning_rate": 0.00012765957446808513,
"loss": 1.2203,
"step": 30
},
{
"epoch": 0.07442849548112707,
"grad_norm": 0.09436627728374515,
"learning_rate": 0.00014893617021276596,
"loss": 1.1787,
"step": 35
},
{
"epoch": 0.08506113769271664,
"grad_norm": 0.08152779581665957,
"learning_rate": 0.00017021276595744682,
"loss": 1.213,
"step": 40
},
{
"epoch": 0.09569377990430622,
"grad_norm": 0.08994938392428893,
"learning_rate": 0.00019148936170212768,
"loss": 1.1771,
"step": 45
},
{
"epoch": 0.1063264221158958,
"grad_norm": 0.08889534543964475,
"learning_rate": 0.0001999751793267259,
"loss": 1.1937,
"step": 50
},
{
"epoch": 0.11695906432748537,
"grad_norm": 0.0818192481558469,
"learning_rate": 0.00019982354205260347,
"loss": 1.1549,
"step": 55
},
{
"epoch": 0.12759170653907495,
"grad_norm": 0.08592131915598437,
"learning_rate": 0.0001995342655949951,
"loss": 1.1452,
"step": 60
},
{
"epoch": 0.13822434875066453,
"grad_norm": 0.08847958460762677,
"learning_rate": 0.000199107748815478,
"loss": 1.1381,
"step": 65
},
{
"epoch": 0.14885699096225413,
"grad_norm": 0.09175483942286262,
"learning_rate": 0.000198544579806,
"loss": 1.1512,
"step": 70
},
{
"epoch": 0.1594896331738437,
"grad_norm": 0.07308896698803534,
"learning_rate": 0.00019784553507800349,
"loss": 1.1245,
"step": 75
},
{
"epoch": 0.17012227538543329,
"grad_norm": 0.07352573225963126,
"learning_rate": 0.00019701157849175228,
"loss": 1.1456,
"step": 80
},
{
"epoch": 0.18075491759702286,
"grad_norm": 0.08124735026425489,
"learning_rate": 0.00019604385992733715,
"loss": 1.138,
"step": 85
},
{
"epoch": 0.19138755980861244,
"grad_norm": 0.07150321514192684,
"learning_rate": 0.0001949437136991925,
"loss": 1.1375,
"step": 90
},
{
"epoch": 0.20202020202020202,
"grad_norm": 0.07663110018396366,
"learning_rate": 0.00019371265671631037,
"loss": 1.123,
"step": 95
},
{
"epoch": 0.2126528442317916,
"grad_norm": 0.08486863588170293,
"learning_rate": 0.00019235238639068856,
"loss": 1.1357,
"step": 100
},
{
"epoch": 0.22328548644338117,
"grad_norm": 0.0762558400001174,
"learning_rate": 0.00019086477829689685,
"loss": 1.131,
"step": 105
},
{
"epoch": 0.23391812865497075,
"grad_norm": 0.07452030437492205,
"learning_rate": 0.00018925188358598813,
"loss": 1.1442,
"step": 110
},
{
"epoch": 0.24455077086656035,
"grad_norm": 0.07570723134200702,
"learning_rate": 0.00018751592615732005,
"loss": 1.1342,
"step": 115
},
{
"epoch": 0.2551834130781499,
"grad_norm": 0.15468624584475818,
"learning_rate": 0.00018565929959218758,
"loss": 1.1158,
"step": 120
},
{
"epoch": 0.2658160552897395,
"grad_norm": 0.08459758525414239,
"learning_rate": 0.00018368456385349334,
"loss": 1.1103,
"step": 125
},
{
"epoch": 0.27644869750132905,
"grad_norm": 0.0699573756198723,
"learning_rate": 0.00018159444175600703,
"loss": 1.1375,
"step": 130
},
{
"epoch": 0.28708133971291866,
"grad_norm": 0.09015868621107682,
"learning_rate": 0.000179391815212081,
"loss": 1.1351,
"step": 135
},
{
"epoch": 0.29771398192450826,
"grad_norm": 0.07786434215665021,
"learning_rate": 0.00017707972125799735,
"loss": 1.1094,
"step": 140
},
{
"epoch": 0.3083466241360978,
"grad_norm": 0.0722359719751335,
"learning_rate": 0.0001746613478664271,
"loss": 1.109,
"step": 145
},
{
"epoch": 0.3189792663476874,
"grad_norm": 0.07007415812680974,
"learning_rate": 0.00017214002955077393,
"loss": 1.118,
"step": 150
},
{
"epoch": 0.32961190855927697,
"grad_norm": 0.07271324414680513,
"learning_rate": 0.00016951924276746425,
"loss": 1.1093,
"step": 155
},
{
"epoch": 0.34024455077086657,
"grad_norm": 0.07579503380547685,
"learning_rate": 0.0001668026011225225,
"loss": 1.132,
"step": 160
},
{
"epoch": 0.3508771929824561,
"grad_norm": 0.0755893077221133,
"learning_rate": 0.00016399385038904138,
"loss": 1.1041,
"step": 165
},
{
"epoch": 0.3615098351940457,
"grad_norm": 0.07521138011494473,
"learning_rate": 0.00016109686334241655,
"loss": 1.1106,
"step": 170
},
{
"epoch": 0.3721424774056353,
"grad_norm": 0.07475367435754098,
"learning_rate": 0.00015811563442046767,
"loss": 1.0981,
"step": 175
},
{
"epoch": 0.3827751196172249,
"grad_norm": 0.07569742075334658,
"learning_rate": 0.00015505427421580808,
"loss": 1.1027,
"step": 180
},
{
"epoch": 0.3934077618288145,
"grad_norm": 0.06970712543989657,
"learning_rate": 0.00015191700380805752,
"loss": 1.1148,
"step": 185
},
{
"epoch": 0.40404040404040403,
"grad_norm": 0.07988874051314146,
"learning_rate": 0.00014870814894371245,
"loss": 1.1113,
"step": 190
},
{
"epoch": 0.41467304625199364,
"grad_norm": 0.07160684049436854,
"learning_rate": 0.0001454321340716992,
"loss": 1.0861,
"step": 195
},
{
"epoch": 0.4253056884635832,
"grad_norm": 0.07005079598481417,
"learning_rate": 0.0001420934762428335,
"loss": 1.0971,
"step": 200
},
{
"epoch": 0.4359383306751728,
"grad_norm": 0.07394470646142667,
"learning_rate": 0.00013869677888159887,
"loss": 1.1136,
"step": 205
},
{
"epoch": 0.44657097288676234,
"grad_norm": 0.07523040228715787,
"learning_rate": 0.00013524672543882996,
"loss": 1.1196,
"step": 210
},
{
"epoch": 0.45720361509835195,
"grad_norm": 0.0740459816152623,
"learning_rate": 0.00013174807293405428,
"loss": 1.122,
"step": 215
},
{
"epoch": 0.4678362573099415,
"grad_norm": 0.07407785053371259,
"learning_rate": 0.00012820564539639512,
"loss": 1.1101,
"step": 220
},
{
"epoch": 0.4784688995215311,
"grad_norm": 0.07407385022434756,
"learning_rate": 0.0001246243272130804,
"loss": 1.1093,
"step": 225
},
{
"epoch": 0.4891015417331207,
"grad_norm": 0.0691999860793088,
"learning_rate": 0.00012100905639472779,
"loss": 1.0947,
"step": 230
},
{
"epoch": 0.49973418394471025,
"grad_norm": 0.07023344891744904,
"learning_rate": 0.00011736481776669306,
"loss": 1.1322,
"step": 235
},
{
"epoch": 0.5103668261562998,
"grad_norm": 0.07064724436775818,
"learning_rate": 0.00011369663609586854,
"loss": 1.1188,
"step": 240
},
{
"epoch": 0.5209994683678895,
"grad_norm": 0.07583291773979836,
"learning_rate": 0.00011000956916240985,
"loss": 1.0995,
"step": 245
},
{
"epoch": 0.531632110579479,
"grad_norm": 0.07152892272662684,
"learning_rate": 0.00010630870078594249,
"loss": 1.1236,
"step": 250
},
{
"epoch": 0.5422647527910686,
"grad_norm": 0.07101072737185006,
"learning_rate": 0.0001025991338158651,
"loss": 1.1311,
"step": 255
},
{
"epoch": 0.5528973950026581,
"grad_norm": 0.0702113295280909,
"learning_rate": 9.888598309541347e-05,
"loss": 1.1021,
"step": 260
},
{
"epoch": 0.5635300372142478,
"grad_norm": 0.07560382456183587,
"learning_rate": 9.517436840918766e-05,
"loss": 1.1193,
"step": 265
},
{
"epoch": 0.5741626794258373,
"grad_norm": 0.07181091122638743,
"learning_rate": 9.146940742386553e-05,
"loss": 1.1171,
"step": 270
},
{
"epoch": 0.5847953216374269,
"grad_norm": 0.0719306314196102,
"learning_rate": 8.777620863183657e-05,
"loss": 1.0973,
"step": 275
},
{
"epoch": 0.5954279638490165,
"grad_norm": 0.06768671936995115,
"learning_rate": 8.409986430748545e-05,
"loss": 1.0993,
"step": 280
},
{
"epoch": 0.6060606060606061,
"grad_norm": 0.07166544718736949,
"learning_rate": 8.044544348583755e-05,
"loss": 1.1116,
"step": 285
},
{
"epoch": 0.6166932482721956,
"grad_norm": 0.07026460560964139,
"learning_rate": 7.681798497324716e-05,
"loss": 1.1215,
"step": 290
},
{
"epoch": 0.6273258904837852,
"grad_norm": 0.07493637213313761,
"learning_rate": 7.322249039976608e-05,
"loss": 1.0951,
"step": 295
},
{
"epoch": 0.6379585326953748,
"grad_norm": 0.0735588585321075,
"learning_rate": 6.966391732277143e-05,
"loss": 1.1051,
"step": 300
},
{
"epoch": 0.6485911749069644,
"grad_norm": 0.07103076233316474,
"learning_rate": 6.614717239136246e-05,
"loss": 1.1099,
"step": 305
},
{
"epoch": 0.6592238171185539,
"grad_norm": 0.06929087442452384,
"learning_rate": 6.267710458095053e-05,
"loss": 1.0891,
"step": 310
},
{
"epoch": 0.6698564593301436,
"grad_norm": 0.07048509973214657,
"learning_rate": 5.9258498507371194e-05,
"loss": 1.1308,
"step": 315
},
{
"epoch": 0.6804891015417331,
"grad_norm": 0.06987687341789249,
"learning_rate": 5.589606782973683e-05,
"loss": 1.1288,
"step": 320
},
{
"epoch": 0.6911217437533227,
"grad_norm": 0.0726818503990417,
"learning_rate": 5.259444875112624e-05,
"loss": 1.1166,
"step": 325
},
{
"epoch": 0.7017543859649122,
"grad_norm": 0.07076508826669288,
"learning_rate": 4.93581936260724e-05,
"loss": 1.1094,
"step": 330
},
{
"epoch": 0.7123870281765019,
"grad_norm": 0.06894707145543243,
"learning_rate": 4.6191764683662744e-05,
"loss": 1.097,
"step": 335
},
{
"epoch": 0.7230196703880915,
"grad_norm": 0.07530097462452943,
"learning_rate": 4.309952787490689e-05,
"loss": 1.1017,
"step": 340
},
{
"epoch": 0.733652312599681,
"grad_norm": 0.06877541461151461,
"learning_rate": 4.008574685285442e-05,
"loss": 1.1034,
"step": 345
},
{
"epoch": 0.7442849548112705,
"grad_norm": 0.07024805662579879,
"learning_rate": 3.7154577093764334e-05,
"loss": 1.0996,
"step": 350
},
{
"epoch": 0.7549175970228602,
"grad_norm": 0.06979054766691796,
"learning_rate": 3.4310060167430725e-05,
"loss": 1.0993,
"step": 355
},
{
"epoch": 0.7655502392344498,
"grad_norm": 0.07127452889238414,
"learning_rate": 3.155611816456586e-05,
"loss": 1.1108,
"step": 360
},
{
"epoch": 0.7761828814460393,
"grad_norm": 0.06929411714148895,
"learning_rate": 2.889654828892393e-05,
"loss": 1.1014,
"step": 365
},
{
"epoch": 0.786815523657629,
"grad_norm": 0.06986765320557396,
"learning_rate": 2.6335017621622116e-05,
"loss": 1.1178,
"step": 370
},
{
"epoch": 0.7974481658692185,
"grad_norm": 0.14171152314730037,
"learning_rate": 2.3875058064877807e-05,
"loss": 1.1056,
"step": 375
},
{
"epoch": 0.8080808080808081,
"grad_norm": 0.07523630706911914,
"learning_rate": 2.1520061472133902e-05,
"loss": 1.1196,
"step": 380
},
{
"epoch": 0.8187134502923976,
"grad_norm": 0.06927631258661882,
"learning_rate": 1.927327497128706e-05,
"loss": 1.1165,
"step": 385
},
{
"epoch": 0.8293460925039873,
"grad_norm": 0.07022924736143368,
"learning_rate": 1.7137796487466797e-05,
"loss": 1.102,
"step": 390
},
{
"epoch": 0.8399787347155768,
"grad_norm": 0.06841887685146186,
"learning_rate": 1.5116570471539293e-05,
"loss": 1.1,
"step": 395
},
{
"epoch": 0.8506113769271664,
"grad_norm": 0.0680433251960951,
"learning_rate": 1.3212383840225329e-05,
"loss": 1.1007,
"step": 400
},
{
"epoch": 0.861244019138756,
"grad_norm": 0.06849436910046028,
"learning_rate": 1.1427862133430156e-05,
"loss": 1.0919,
"step": 405
},
{
"epoch": 0.8718766613503456,
"grad_norm": 0.06705533678821948,
"learning_rate": 9.765465894083636e-06,
"loss": 1.1018,
"step": 410
},
{
"epoch": 0.8825093035619351,
"grad_norm": 0.06997290719332996,
"learning_rate": 8.227487275482592e-06,
"loss": 1.1141,
"step": 415
},
{
"epoch": 0.8931419457735247,
"grad_norm": 0.06955661856691482,
"learning_rate": 6.81604688081271e-06,
"loss": 1.1179,
"step": 420
},
{
"epoch": 0.9037745879851143,
"grad_norm": 0.06939654515531442,
"learning_rate": 5.533090839208133e-06,
"loss": 1.0859,
"step": 425
},
{
"epoch": 0.9144072301967039,
"grad_norm": 0.06890766733706018,
"learning_rate": 4.380388122380141e-06,
"loss": 1.1041,
"step": 430
},
{
"epoch": 0.9250398724082934,
"grad_norm": 0.06774543961855677,
"learning_rate": 3.359528105515064e-06,
"loss": 1.1054,
"step": 435
},
{
"epoch": 0.935672514619883,
"grad_norm": 0.06656651460274651,
"learning_rate": 2.471918375804105e-06,
"loss": 1.1018,
"step": 440
},
{
"epoch": 0.9463051568314726,
"grad_norm": 0.06852233232210171,
"learning_rate": 1.7187827916271382e-06,
"loss": 1.1129,
"step": 445
},
{
"epoch": 0.9569377990430622,
"grad_norm": 0.06708991081678066,
"learning_rate": 1.1011597950663865e-06,
"loss": 1.1046,
"step": 450
},
{
"epoch": 0.9675704412546517,
"grad_norm": 0.06813589171568658,
"learning_rate": 6.199009800765265e-07,
"loss": 1.0986,
"step": 455
},
{
"epoch": 0.9782030834662414,
"grad_norm": 0.0700679858970667,
"learning_rate": 2.756699182858369e-07,
"loss": 1.111,
"step": 460
},
{
"epoch": 0.988835725677831,
"grad_norm": 0.07404495795184537,
"learning_rate": 6.894124404711599e-08,
"loss": 1.1039,
"step": 465
},
{
"epoch": 0.9994683678894205,
"grad_norm": 0.07042216599583477,
"learning_rate": 0.0,
"loss": 1.1192,
"step": 470
},
{
"epoch": 0.9994683678894205,
"eval_runtime": 5.9798,
"eval_samples_per_second": 1.672,
"eval_steps_per_second": 0.502,
"step": 470
},
{
"epoch": 0.9994683678894205,
"step": 470,
"total_flos": 3.0618166709610086e+17,
"train_loss": 1.1288761854171754,
"train_runtime": 18761.7867,
"train_samples_per_second": 3.207,
"train_steps_per_second": 0.025
}
],
"logging_steps": 5,
"max_steps": 470,
"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": 3.0618166709610086e+17,
"train_batch_size": 8,
"trial_name": null,
"trial_params": null
}