terry69's picture
Model save
4a1239d verified
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 1.0,
"eval_steps": 500,
"global_step": 454,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0022026431718061676,
"grad_norm": 1.0637864900736664,
"learning_rate": 4.347826086956522e-06,
"loss": 1.4366,
"step": 1
},
{
"epoch": 0.011013215859030838,
"grad_norm": 1.153707883328667,
"learning_rate": 2.173913043478261e-05,
"loss": 1.4048,
"step": 5
},
{
"epoch": 0.022026431718061675,
"grad_norm": 0.24033826612761172,
"learning_rate": 4.347826086956522e-05,
"loss": 1.3638,
"step": 10
},
{
"epoch": 0.03303964757709251,
"grad_norm": 0.23098755386890235,
"learning_rate": 6.521739130434783e-05,
"loss": 1.3035,
"step": 15
},
{
"epoch": 0.04405286343612335,
"grad_norm": 0.18769938367180908,
"learning_rate": 8.695652173913044e-05,
"loss": 1.2767,
"step": 20
},
{
"epoch": 0.05506607929515418,
"grad_norm": 0.14239545432122608,
"learning_rate": 0.00010869565217391305,
"loss": 1.2196,
"step": 25
},
{
"epoch": 0.06607929515418502,
"grad_norm": 0.10372099126300886,
"learning_rate": 0.00013043478260869567,
"loss": 1.21,
"step": 30
},
{
"epoch": 0.07709251101321586,
"grad_norm": 0.10109724419729917,
"learning_rate": 0.00015217391304347827,
"loss": 1.2161,
"step": 35
},
{
"epoch": 0.0881057268722467,
"grad_norm": 0.08677348076375273,
"learning_rate": 0.00017391304347826088,
"loss": 1.2002,
"step": 40
},
{
"epoch": 0.09911894273127753,
"grad_norm": 0.07988238174290498,
"learning_rate": 0.0001956521739130435,
"loss": 1.1907,
"step": 45
},
{
"epoch": 0.11013215859030837,
"grad_norm": 0.08057978560906214,
"learning_rate": 0.0001999525719713366,
"loss": 1.1572,
"step": 50
},
{
"epoch": 0.1211453744493392,
"grad_norm": 0.08125586602301926,
"learning_rate": 0.0001997599727063717,
"loss": 1.1575,
"step": 55
},
{
"epoch": 0.13215859030837004,
"grad_norm": 0.08174265113795184,
"learning_rate": 0.00019941952317728147,
"loss": 1.1662,
"step": 60
},
{
"epoch": 0.14317180616740088,
"grad_norm": 0.09240936715995939,
"learning_rate": 0.00019893172795069144,
"loss": 1.1561,
"step": 65
},
{
"epoch": 0.15418502202643172,
"grad_norm": 0.07863084458052047,
"learning_rate": 0.0001982973099683902,
"loss": 1.137,
"step": 70
},
{
"epoch": 0.16519823788546256,
"grad_norm": 0.08035585420548778,
"learning_rate": 0.00019751720947588602,
"loss": 1.1417,
"step": 75
},
{
"epoch": 0.1762114537444934,
"grad_norm": 0.07156903107546937,
"learning_rate": 0.00019659258262890683,
"loss": 1.156,
"step": 80
},
{
"epoch": 0.18722466960352424,
"grad_norm": 0.07056162758348125,
"learning_rate": 0.000195524799779908,
"loss": 1.135,
"step": 85
},
{
"epoch": 0.19823788546255505,
"grad_norm": 0.07468007657011916,
"learning_rate": 0.00019431544344712776,
"loss": 1.1285,
"step": 90
},
{
"epoch": 0.2092511013215859,
"grad_norm": 0.07250524328449914,
"learning_rate": 0.00019296630596920023,
"loss": 1.1335,
"step": 95
},
{
"epoch": 0.22026431718061673,
"grad_norm": 0.06955764714773345,
"learning_rate": 0.0001914793868488021,
"loss": 1.1512,
"step": 100
},
{
"epoch": 0.23127753303964757,
"grad_norm": 0.07221286385490396,
"learning_rate": 0.0001898568897892697,
"loss": 1.1306,
"step": 105
},
{
"epoch": 0.2422907488986784,
"grad_norm": 0.07486140156540808,
"learning_rate": 0.00018810121942857845,
"loss": 1.1342,
"step": 110
},
{
"epoch": 0.2533039647577093,
"grad_norm": 0.07131722083122453,
"learning_rate": 0.00018621497777552507,
"loss": 1.1307,
"step": 115
},
{
"epoch": 0.2643171806167401,
"grad_norm": 0.07348336925535924,
"learning_rate": 0.00018420096035339452,
"loss": 1.1222,
"step": 120
},
{
"epoch": 0.2753303964757709,
"grad_norm": 0.07501908796003587,
"learning_rate": 0.00018206215205682683,
"loss": 1.1116,
"step": 125
},
{
"epoch": 0.28634361233480177,
"grad_norm": 0.07330385148735,
"learning_rate": 0.000179801722728024,
"loss": 1.1373,
"step": 130
},
{
"epoch": 0.2973568281938326,
"grad_norm": 0.073420013451975,
"learning_rate": 0.00017742302245885383,
"loss": 1.1053,
"step": 135
},
{
"epoch": 0.30837004405286345,
"grad_norm": 0.07954193429333527,
"learning_rate": 0.00017492957662581295,
"loss": 1.1232,
"step": 140
},
{
"epoch": 0.31938325991189426,
"grad_norm": 0.07217417992157714,
"learning_rate": 0.00017232508066520702,
"loss": 1.1237,
"step": 145
},
{
"epoch": 0.3303964757709251,
"grad_norm": 0.07802555489353481,
"learning_rate": 0.0001696133945962927,
"loss": 1.1208,
"step": 150
},
{
"epoch": 0.34140969162995594,
"grad_norm": 0.08970080549205696,
"learning_rate": 0.00016679853730049743,
"loss": 1.1404,
"step": 155
},
{
"epoch": 0.3524229074889868,
"grad_norm": 0.07726401697227049,
"learning_rate": 0.00016388468056519612,
"loss": 1.0981,
"step": 160
},
{
"epoch": 0.3634361233480176,
"grad_norm": 0.0738469257596569,
"learning_rate": 0.00016087614290087208,
"loss": 1.1245,
"step": 165
},
{
"epoch": 0.3744493392070485,
"grad_norm": 0.0760961629898493,
"learning_rate": 0.00015777738314082514,
"loss": 1.1282,
"step": 170
},
{
"epoch": 0.3854625550660793,
"grad_norm": 0.07461575712201869,
"learning_rate": 0.00015459299383291345,
"loss": 1.1206,
"step": 175
},
{
"epoch": 0.3964757709251101,
"grad_norm": 0.07597642898156688,
"learning_rate": 0.00015132769443312207,
"loss": 1.1151,
"step": 180
},
{
"epoch": 0.40748898678414097,
"grad_norm": 0.07220979125832422,
"learning_rate": 0.00014798632431104592,
"loss": 1.1313,
"step": 185
},
{
"epoch": 0.4185022026431718,
"grad_norm": 0.07235062387347811,
"learning_rate": 0.00014457383557765386,
"loss": 1.126,
"step": 190
},
{
"epoch": 0.42951541850220265,
"grad_norm": 0.07249541910324554,
"learning_rate": 0.00014109528574596301,
"loss": 1.1223,
"step": 195
},
{
"epoch": 0.44052863436123346,
"grad_norm": 0.07276295542602365,
"learning_rate": 0.00013755583023550126,
"loss": 1.0954,
"step": 200
},
{
"epoch": 0.45154185022026433,
"grad_norm": 0.08222792797320272,
"learning_rate": 0.00013396071473166613,
"loss": 1.1109,
"step": 205
},
{
"epoch": 0.46255506607929514,
"grad_norm": 0.07000329815419548,
"learning_rate": 0.00013031526741130435,
"loss": 1.1122,
"step": 210
},
{
"epoch": 0.473568281938326,
"grad_norm": 0.07273029465748866,
"learning_rate": 0.0001266248910460341,
"loss": 1.1098,
"step": 215
},
{
"epoch": 0.4845814977973568,
"grad_norm": 0.07891968539218898,
"learning_rate": 0.0001228950549950134,
"loss": 1.1235,
"step": 220
},
{
"epoch": 0.4955947136563877,
"grad_norm": 0.07948656131261005,
"learning_rate": 0.00011913128709902181,
"loss": 1.119,
"step": 225
},
{
"epoch": 0.5066079295154186,
"grad_norm": 0.07426812917091022,
"learning_rate": 0.00011533916548786857,
"loss": 1.1153,
"step": 230
},
{
"epoch": 0.5176211453744494,
"grad_norm": 0.07320813310432901,
"learning_rate": 0.00011152431031326978,
"loss": 1.1189,
"step": 235
},
{
"epoch": 0.5286343612334802,
"grad_norm": 0.07027215565752497,
"learning_rate": 0.0001076923754194464,
"loss": 1.0921,
"step": 240
},
{
"epoch": 0.539647577092511,
"grad_norm": 0.07197803964893518,
"learning_rate": 0.00010384903996378783,
"loss": 1.139,
"step": 245
},
{
"epoch": 0.5506607929515418,
"grad_norm": 0.07476406384806204,
"learning_rate": 0.0001,
"loss": 1.113,
"step": 250
},
{
"epoch": 0.5616740088105727,
"grad_norm": 0.07773914011299175,
"learning_rate": 9.615096003621221e-05,
"loss": 1.1112,
"step": 255
},
{
"epoch": 0.5726872246696035,
"grad_norm": 0.07599058071843097,
"learning_rate": 9.230762458055363e-05,
"loss": 1.0823,
"step": 260
},
{
"epoch": 0.5837004405286343,
"grad_norm": 0.0791178168791927,
"learning_rate": 8.847568968673026e-05,
"loss": 1.1322,
"step": 265
},
{
"epoch": 0.5947136563876652,
"grad_norm": 0.08321320900809528,
"learning_rate": 8.466083451213144e-05,
"loss": 1.1129,
"step": 270
},
{
"epoch": 0.6057268722466961,
"grad_norm": 0.07548913525051125,
"learning_rate": 8.086871290097821e-05,
"loss": 1.1062,
"step": 275
},
{
"epoch": 0.6167400881057269,
"grad_norm": 0.07309802934315239,
"learning_rate": 7.710494500498662e-05,
"loss": 1.1129,
"step": 280
},
{
"epoch": 0.6277533039647577,
"grad_norm": 0.069929912224867,
"learning_rate": 7.337510895396591e-05,
"loss": 1.1142,
"step": 285
},
{
"epoch": 0.6387665198237885,
"grad_norm": 0.06916254050976436,
"learning_rate": 6.968473258869566e-05,
"loss": 1.1115,
"step": 290
},
{
"epoch": 0.6497797356828194,
"grad_norm": 0.07179141392925117,
"learning_rate": 6.603928526833387e-05,
"loss": 1.1114,
"step": 295
},
{
"epoch": 0.6607929515418502,
"grad_norm": 0.07393948815408477,
"learning_rate": 6.244416976449875e-05,
"loss": 1.0976,
"step": 300
},
{
"epoch": 0.6718061674008811,
"grad_norm": 0.06872619691818142,
"learning_rate": 5.890471425403703e-05,
"loss": 1.1146,
"step": 305
},
{
"epoch": 0.6828193832599119,
"grad_norm": 0.07576414521745392,
"learning_rate": 5.542616442234618e-05,
"loss": 1.1147,
"step": 310
},
{
"epoch": 0.6938325991189427,
"grad_norm": 0.07330914446070216,
"learning_rate": 5.201367568895408e-05,
"loss": 1.0951,
"step": 315
},
{
"epoch": 0.7048458149779736,
"grad_norm": 0.07224487325459926,
"learning_rate": 4.8672305566877964e-05,
"loss": 1.1086,
"step": 320
},
{
"epoch": 0.7158590308370044,
"grad_norm": 0.075306080937695,
"learning_rate": 4.540700616708658e-05,
"loss": 1.1092,
"step": 325
},
{
"epoch": 0.7268722466960352,
"grad_norm": 0.07437067674720446,
"learning_rate": 4.222261685917489e-05,
"loss": 1.0921,
"step": 330
},
{
"epoch": 0.737885462555066,
"grad_norm": 0.0711166096176443,
"learning_rate": 3.9123857099127936e-05,
"loss": 1.1074,
"step": 335
},
{
"epoch": 0.748898678414097,
"grad_norm": 0.07140111450874426,
"learning_rate": 3.6115319434803894e-05,
"loss": 1.1162,
"step": 340
},
{
"epoch": 0.7599118942731278,
"grad_norm": 0.07495611302185953,
"learning_rate": 3.32014626995026e-05,
"loss": 1.1003,
"step": 345
},
{
"epoch": 0.7709251101321586,
"grad_norm": 0.07717586378795258,
"learning_rate": 3.0386605403707346e-05,
"loss": 1.0927,
"step": 350
},
{
"epoch": 0.7819383259911894,
"grad_norm": 0.07229554748902306,
"learning_rate": 2.7674919334793035e-05,
"loss": 1.1108,
"step": 355
},
{
"epoch": 0.7929515418502202,
"grad_norm": 0.0768434750345143,
"learning_rate": 2.507042337418707e-05,
"loss": 1.1058,
"step": 360
},
{
"epoch": 0.8039647577092511,
"grad_norm": 0.07347159962460627,
"learning_rate": 2.2576977541146193e-05,
"loss": 1.1115,
"step": 365
},
{
"epoch": 0.8149779735682819,
"grad_norm": 0.0690702684713988,
"learning_rate": 2.0198277271976052e-05,
"loss": 1.1091,
"step": 370
},
{
"epoch": 0.8259911894273128,
"grad_norm": 0.06989352121524041,
"learning_rate": 1.793784794317319e-05,
"loss": 1.0988,
"step": 375
},
{
"epoch": 0.8370044052863436,
"grad_norm": 0.06951408734875156,
"learning_rate": 1.5799039646605486e-05,
"loss": 1.1,
"step": 380
},
{
"epoch": 0.8480176211453745,
"grad_norm": 0.06836481993836284,
"learning_rate": 1.3785022224474942e-05,
"loss": 1.102,
"step": 385
},
{
"epoch": 0.8590308370044053,
"grad_norm": 0.06852988441116457,
"learning_rate": 1.1898780571421552e-05,
"loss": 1.0987,
"step": 390
},
{
"epoch": 0.8700440528634361,
"grad_norm": 0.07030606192740464,
"learning_rate": 1.0143110210730312e-05,
"loss": 1.1009,
"step": 395
},
{
"epoch": 0.8810572687224669,
"grad_norm": 0.06918150786144664,
"learning_rate": 8.520613151197898e-06,
"loss": 1.1123,
"step": 400
},
{
"epoch": 0.8920704845814978,
"grad_norm": 0.07235026662376734,
"learning_rate": 7.033694030799787e-06,
"loss": 1.098,
"step": 405
},
{
"epoch": 0.9030837004405287,
"grad_norm": 0.06951148470437334,
"learning_rate": 5.684556552872256e-06,
"loss": 1.0975,
"step": 410
},
{
"epoch": 0.9140969162995595,
"grad_norm": 0.06872065759217105,
"learning_rate": 4.475200220092002e-06,
"loss": 1.0896,
"step": 415
},
{
"epoch": 0.9251101321585903,
"grad_norm": 0.06809525315060157,
"learning_rate": 3.40741737109318e-06,
"loss": 1.1158,
"step": 420
},
{
"epoch": 0.9361233480176211,
"grad_norm": 0.07195357579119646,
"learning_rate": 2.482790524113998e-06,
"loss": 1.1097,
"step": 425
},
{
"epoch": 0.947136563876652,
"grad_norm": 0.06817141756269769,
"learning_rate": 1.7026900316098215e-06,
"loss": 1.1113,
"step": 430
},
{
"epoch": 0.9581497797356828,
"grad_norm": 0.07218201005051188,
"learning_rate": 1.0682720493085607e-06,
"loss": 1.1049,
"step": 435
},
{
"epoch": 0.9691629955947136,
"grad_norm": 0.06979182488846636,
"learning_rate": 5.804768227185565e-07,
"loss": 1.1125,
"step": 440
},
{
"epoch": 0.9801762114537445,
"grad_norm": 0.06813879663952505,
"learning_rate": 2.400272936283088e-07,
"loss": 1.1002,
"step": 445
},
{
"epoch": 0.9911894273127754,
"grad_norm": 0.06852453577386937,
"learning_rate": 4.74280286634099e-08,
"loss": 1.094,
"step": 450
},
{
"epoch": 1.0,
"eval_loss": 1.0366965532302856,
"eval_runtime": 2.2297,
"eval_samples_per_second": 3.139,
"eval_steps_per_second": 0.897,
"step": 454
},
{
"epoch": 1.0,
"step": 454,
"total_flos": 1.4408836278386688e+16,
"train_loss": 1.131237539688396,
"train_runtime": 17315.3957,
"train_samples_per_second": 3.356,
"train_steps_per_second": 0.026
}
],
"logging_steps": 5,
"max_steps": 454,
"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": 1.4408836278386688e+16,
"train_batch_size": 8,
"trial_name": null,
"trial_params": null
}