mistral5p / trainer_state.json
terry69's picture
Model save
fe4b622 verified
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 1.0,
"eval_steps": 500,
"global_step": 406,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0024630541871921183,
"grad_norm": 0.13191281259059906,
"learning_rate": 4.8780487804878055e-06,
"loss": 0.7592,
"step": 1
},
{
"epoch": 0.012315270935960592,
"grad_norm": 0.1377675086259842,
"learning_rate": 2.4390243902439026e-05,
"loss": 0.7841,
"step": 5
},
{
"epoch": 0.024630541871921183,
"grad_norm": 0.11912436038255692,
"learning_rate": 4.878048780487805e-05,
"loss": 0.7896,
"step": 10
},
{
"epoch": 0.03694581280788178,
"grad_norm": 0.12670652568340302,
"learning_rate": 7.317073170731707e-05,
"loss": 0.7835,
"step": 15
},
{
"epoch": 0.04926108374384237,
"grad_norm": 0.10752439498901367,
"learning_rate": 9.75609756097561e-05,
"loss": 0.7806,
"step": 20
},
{
"epoch": 0.06157635467980296,
"grad_norm": 0.10097604990005493,
"learning_rate": 0.00012195121951219512,
"loss": 0.7567,
"step": 25
},
{
"epoch": 0.07389162561576355,
"grad_norm": 0.10570556670427322,
"learning_rate": 0.00014634146341463414,
"loss": 0.7527,
"step": 30
},
{
"epoch": 0.08620689655172414,
"grad_norm": 0.11552459746599197,
"learning_rate": 0.0001707317073170732,
"loss": 0.7514,
"step": 35
},
{
"epoch": 0.09852216748768473,
"grad_norm": 0.11246493458747864,
"learning_rate": 0.0001951219512195122,
"loss": 0.75,
"step": 40
},
{
"epoch": 0.11083743842364532,
"grad_norm": 0.10585317760705948,
"learning_rate": 0.0001999407400739705,
"loss": 0.738,
"step": 45
},
{
"epoch": 0.12315270935960591,
"grad_norm": 0.10629229247570038,
"learning_rate": 0.00019970011699250152,
"loss": 0.7748,
"step": 50
},
{
"epoch": 0.1354679802955665,
"grad_norm": 0.09721837192773819,
"learning_rate": 0.00019927487224577402,
"loss": 0.7469,
"step": 55
},
{
"epoch": 0.1477832512315271,
"grad_norm": 0.10196585208177567,
"learning_rate": 0.0001986657932891657,
"loss": 0.7435,
"step": 60
},
{
"epoch": 0.16009852216748768,
"grad_norm": 0.09946288913488388,
"learning_rate": 0.00019787400799669154,
"loss": 0.7482,
"step": 65
},
{
"epoch": 0.1724137931034483,
"grad_norm": 0.10071098804473877,
"learning_rate": 0.00019690098257244064,
"loss": 0.7551,
"step": 70
},
{
"epoch": 0.18472906403940886,
"grad_norm": 0.10068488121032715,
"learning_rate": 0.00019574851883550395,
"loss": 0.7502,
"step": 75
},
{
"epoch": 0.19704433497536947,
"grad_norm": 0.10105381160974503,
"learning_rate": 0.00019441875088341997,
"loss": 0.7438,
"step": 80
},
{
"epoch": 0.20935960591133004,
"grad_norm": 0.09201391041278839,
"learning_rate": 0.00019291414114031743,
"loss": 0.7374,
"step": 85
},
{
"epoch": 0.22167487684729065,
"grad_norm": 0.09724140912294388,
"learning_rate": 0.00019123747579707275,
"loss": 0.7588,
"step": 90
},
{
"epoch": 0.23399014778325122,
"grad_norm": 0.09425447881221771,
"learning_rate": 0.0001893918596519257,
"loss": 0.7631,
"step": 95
},
{
"epoch": 0.24630541871921183,
"grad_norm": 0.10638237744569778,
"learning_rate": 0.00018738071036110808,
"loss": 0.7507,
"step": 100
},
{
"epoch": 0.25862068965517243,
"grad_norm": 0.10152100026607513,
"learning_rate": 0.00018520775211013093,
"loss": 0.7503,
"step": 105
},
{
"epoch": 0.270935960591133,
"grad_norm": 0.09959056228399277,
"learning_rate": 0.00018287700871745036,
"loss": 0.7644,
"step": 110
},
{
"epoch": 0.2832512315270936,
"grad_norm": 0.10374708473682404,
"learning_rate": 0.00018039279618328212,
"loss": 0.7601,
"step": 115
},
{
"epoch": 0.2955665024630542,
"grad_norm": 0.10066307336091995,
"learning_rate": 0.0001777597146973627,
"loss": 0.7392,
"step": 120
},
{
"epoch": 0.3078817733990148,
"grad_norm": 0.1052442416548729,
"learning_rate": 0.00017498264012045687,
"loss": 0.7443,
"step": 125
},
{
"epoch": 0.32019704433497537,
"grad_norm": 0.09332836419343948,
"learning_rate": 0.00017206671495538612,
"loss": 0.7407,
"step": 130
},
{
"epoch": 0.33251231527093594,
"grad_norm": 0.10167726129293442,
"learning_rate": 0.0001690173388242972,
"loss": 0.737,
"step": 135
},
{
"epoch": 0.3448275862068966,
"grad_norm": 0.09576103836297989,
"learning_rate": 0.0001658401584698049,
"loss": 0.7467,
"step": 140
},
{
"epoch": 0.35714285714285715,
"grad_norm": 0.09853356331586838,
"learning_rate": 0.00016254105729852464,
"loss": 0.7413,
"step": 145
},
{
"epoch": 0.3694581280788177,
"grad_norm": 0.0903814435005188,
"learning_rate": 0.00015912614448635782,
"loss": 0.7529,
"step": 150
},
{
"epoch": 0.3817733990147783,
"grad_norm": 0.10001372545957565,
"learning_rate": 0.00015560174366570446,
"loss": 0.7396,
"step": 155
},
{
"epoch": 0.39408866995073893,
"grad_norm": 0.10121896117925644,
"learning_rate": 0.0001519743812155516,
"loss": 0.7536,
"step": 160
},
{
"epoch": 0.4064039408866995,
"grad_norm": 0.09317266196012497,
"learning_rate": 0.00014825077417612186,
"loss": 0.754,
"step": 165
},
{
"epoch": 0.4187192118226601,
"grad_norm": 0.09699834138154984,
"learning_rate": 0.00014443781781046136,
"loss": 0.7585,
"step": 170
},
{
"epoch": 0.43103448275862066,
"grad_norm": 0.09634856134653091,
"learning_rate": 0.00014054257283599973,
"loss": 0.7193,
"step": 175
},
{
"epoch": 0.4433497536945813,
"grad_norm": 0.09830117970705032,
"learning_rate": 0.00013657225234972695,
"loss": 0.7346,
"step": 180
},
{
"epoch": 0.45566502463054187,
"grad_norm": 0.09632200747728348,
"learning_rate": 0.00013253420847119803,
"loss": 0.7356,
"step": 185
},
{
"epoch": 0.46798029556650245,
"grad_norm": 0.09794441610574722,
"learning_rate": 0.0001284359187281004,
"loss": 0.7404,
"step": 190
},
{
"epoch": 0.4802955665024631,
"grad_norm": 0.09384100884199142,
"learning_rate": 0.0001242849722095936,
"loss": 0.7282,
"step": 195
},
{
"epoch": 0.49261083743842365,
"grad_norm": 0.09335623681545258,
"learning_rate": 0.00012008905551306356,
"loss": 0.7511,
"step": 200
},
{
"epoch": 0.5049261083743842,
"grad_norm": 0.11448359489440918,
"learning_rate": 0.00011585593851031347,
"loss": 0.7173,
"step": 205
},
{
"epoch": 0.5172413793103449,
"grad_norm": 0.11399874091148376,
"learning_rate": 0.00011159345995955006,
"loss": 0.7261,
"step": 210
},
{
"epoch": 0.5295566502463054,
"grad_norm": 0.10711462795734406,
"learning_rate": 0.00010730951298980776,
"loss": 0.7293,
"step": 215
},
{
"epoch": 0.541871921182266,
"grad_norm": 0.11034953594207764,
"learning_rate": 0.00010301203048469083,
"loss": 0.7284,
"step": 220
},
{
"epoch": 0.5541871921182266,
"grad_norm": 0.11478842049837112,
"learning_rate": 9.870897039249911e-05,
"loss": 0.7146,
"step": 225
},
{
"epoch": 0.5665024630541872,
"grad_norm": 0.1125415787100792,
"learning_rate": 9.440830098993969e-05,
"loss": 0.7218,
"step": 230
},
{
"epoch": 0.5788177339901478,
"grad_norm": 0.11621884256601334,
"learning_rate": 9.011798612671286e-05,
"loss": 0.7242,
"step": 235
},
{
"epoch": 0.5911330049261084,
"grad_norm": 0.11078532040119171,
"learning_rate": 8.58459704782957e-05,
"loss": 0.7268,
"step": 240
},
{
"epoch": 0.603448275862069,
"grad_norm": 0.11298695206642151,
"learning_rate": 8.160016483423199e-05,
"loss": 0.7317,
"step": 245
},
{
"epoch": 0.6157635467980296,
"grad_norm": 0.1098036915063858,
"learning_rate": 7.738843144917119e-05,
"loss": 0.7174,
"step": 250
},
{
"epoch": 0.6280788177339901,
"grad_norm": 0.11521229147911072,
"learning_rate": 7.321856948378259e-05,
"loss": 0.7123,
"step": 255
},
{
"epoch": 0.6403940886699507,
"grad_norm": 0.11004281044006348,
"learning_rate": 6.909830056250527e-05,
"loss": 0.7247,
"step": 260
},
{
"epoch": 0.6527093596059114,
"grad_norm": 0.11443013697862625,
"learning_rate": 6.503525447487715e-05,
"loss": 0.7016,
"step": 265
},
{
"epoch": 0.6650246305418719,
"grad_norm": 0.12297528237104416,
"learning_rate": 6.103695504692122e-05,
"loss": 0.7014,
"step": 270
},
{
"epoch": 0.6773399014778325,
"grad_norm": 0.11536238342523575,
"learning_rate": 5.7110806208751655e-05,
"loss": 0.7248,
"step": 275
},
{
"epoch": 0.6896551724137931,
"grad_norm": 0.1148216500878334,
"learning_rate": 5.326407828419979e-05,
"loss": 0.7145,
"step": 280
},
{
"epoch": 0.7019704433497537,
"grad_norm": 0.11440951377153397,
"learning_rate": 4.9503894527847964e-05,
"loss": 0.7127,
"step": 285
},
{
"epoch": 0.7142857142857143,
"grad_norm": 0.11594696342945099,
"learning_rate": 4.583721793440188e-05,
"loss": 0.7491,
"step": 290
},
{
"epoch": 0.7266009852216748,
"grad_norm": 0.1071944460272789,
"learning_rate": 4.227083834482728e-05,
"loss": 0.7173,
"step": 295
},
{
"epoch": 0.7389162561576355,
"grad_norm": 0.11470213532447815,
"learning_rate": 3.881135987312757e-05,
"loss": 0.72,
"step": 300
},
{
"epoch": 0.7512315270935961,
"grad_norm": 0.11714080721139908,
"learning_rate": 3.546518867704499e-05,
"loss": 0.7289,
"step": 305
},
{
"epoch": 0.7635467980295566,
"grad_norm": 0.11828279495239258,
"learning_rate": 3.223852109533112e-05,
"loss": 0.7209,
"step": 310
},
{
"epoch": 0.7758620689655172,
"grad_norm": 0.1184413954615593,
"learning_rate": 2.9137332173554043e-05,
"loss": 0.7071,
"step": 315
},
{
"epoch": 0.7881773399014779,
"grad_norm": 0.12409494072198868,
"learning_rate": 2.616736459968936e-05,
"loss": 0.7104,
"step": 320
},
{
"epoch": 0.8004926108374384,
"grad_norm": 0.1128469705581665,
"learning_rate": 2.33341180699841e-05,
"loss": 0.7128,
"step": 325
},
{
"epoch": 0.812807881773399,
"grad_norm": 0.11089266836643219,
"learning_rate": 2.0642839104785272e-05,
"loss": 0.7166,
"step": 330
},
{
"epoch": 0.8251231527093597,
"grad_norm": 0.11719653755426407,
"learning_rate": 1.8098511333192024e-05,
"loss": 0.7333,
"step": 335
},
{
"epoch": 0.8374384236453202,
"grad_norm": 0.11491881310939789,
"learning_rate": 1.570584626452173e-05,
"loss": 0.7208,
"step": 340
},
{
"epoch": 0.8497536945812808,
"grad_norm": 0.10957372933626175,
"learning_rate": 1.3469274563679402e-05,
"loss": 0.7046,
"step": 345
},
{
"epoch": 0.8620689655172413,
"grad_norm": 0.1188158467411995,
"learning_rate": 1.1392937846586215e-05,
"loss": 0.7178,
"step": 350
},
{
"epoch": 0.874384236453202,
"grad_norm": 0.11511045694351196,
"learning_rate": 9.48068101086026e-06,
"loss": 0.7161,
"step": 355
},
{
"epoch": 0.8866995073891626,
"grad_norm": 0.11417835205793381,
"learning_rate": 7.736045115951251e-06,
"loss": 0.7274,
"step": 360
},
{
"epoch": 0.8990147783251231,
"grad_norm": 0.11594397574663162,
"learning_rate": 6.16226082591359e-06,
"loss": 0.7081,
"step": 365
},
{
"epoch": 0.9113300492610837,
"grad_norm": 0.11524856835603714,
"learning_rate": 4.762242426960262e-06,
"loss": 0.7209,
"step": 370
},
{
"epoch": 0.9236453201970444,
"grad_norm": 0.12005619704723358,
"learning_rate": 3.5385824308756587e-06,
"loss": 0.6989,
"step": 375
},
{
"epoch": 0.9359605911330049,
"grad_norm": 0.12374743819236755,
"learning_rate": 2.493546774280531e-06,
"loss": 0.7276,
"step": 380
},
{
"epoch": 0.9482758620689655,
"grad_norm": 0.11894236505031586,
"learning_rate": 1.6290706226390285e-06,
"loss": 0.7144,
"step": 385
},
{
"epoch": 0.9605911330049262,
"grad_norm": 0.11776220053434372,
"learning_rate": 9.46754786777726e-07,
"loss": 0.7015,
"step": 390
},
{
"epoch": 0.9729064039408867,
"grad_norm": 0.109645314514637,
"learning_rate": 4.4786275855247527e-07,
"loss": 0.7109,
"step": 395
},
{
"epoch": 0.9852216748768473,
"grad_norm": 0.11453017592430115,
"learning_rate": 1.333183711524133e-07,
"loss": 0.7094,
"step": 400
},
{
"epoch": 0.9975369458128078,
"grad_norm": 0.1103915199637413,
"learning_rate": 3.7040883734462683e-09,
"loss": 0.6894,
"step": 405
},
{
"epoch": 1.0,
"eval_loss": NaN,
"eval_runtime": 944.7015,
"eval_samples_per_second": 1.223,
"eval_steps_per_second": 0.306,
"step": 406
},
{
"epoch": 1.0,
"step": 406,
"total_flos": 2.1219407349982167e+18,
"train_loss": 0.3640357888684484,
"train_runtime": 42241.1922,
"train_samples_per_second": 2.46,
"train_steps_per_second": 0.01
}
],
"logging_steps": 5,
"max_steps": 406,
"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": false
},
"attributes": {}
}
},
"total_flos": 2.1219407349982167e+18,
"train_batch_size": 8,
"trial_name": null,
"trial_params": null
}