codellama-7b-ultrachat / trainer_state.json
kykim0's picture
Model save
000fc5c verified
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 1.9990995047276003,
"eval_steps": 500,
"global_step": 1110,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.018009905447996397,
"grad_norm": 3.8927660727308404,
"learning_rate": 1.801801801801802e-06,
"loss": 1.3686,
"step": 10
},
{
"epoch": 0.03601981089599279,
"grad_norm": 1.2914340039825571,
"learning_rate": 3.603603603603604e-06,
"loss": 1.2618,
"step": 20
},
{
"epoch": 0.0540297163439892,
"grad_norm": 0.5117572348414451,
"learning_rate": 5.405405405405406e-06,
"loss": 1.1941,
"step": 30
},
{
"epoch": 0.07203962179198559,
"grad_norm": 0.3415375067191979,
"learning_rate": 7.207207207207208e-06,
"loss": 1.157,
"step": 40
},
{
"epoch": 0.090049527239982,
"grad_norm": 0.3050099720948823,
"learning_rate": 9.00900900900901e-06,
"loss": 1.1486,
"step": 50
},
{
"epoch": 0.1080594326879784,
"grad_norm": 0.2628693083106922,
"learning_rate": 1.0810810810810812e-05,
"loss": 1.12,
"step": 60
},
{
"epoch": 0.1260693381359748,
"grad_norm": 0.23606513506788576,
"learning_rate": 1.2612612612612613e-05,
"loss": 1.1014,
"step": 70
},
{
"epoch": 0.14407924358397117,
"grad_norm": 0.24000278532244093,
"learning_rate": 1.4414414414414416e-05,
"loss": 1.1087,
"step": 80
},
{
"epoch": 0.16208914903196758,
"grad_norm": 0.21377676080662794,
"learning_rate": 1.6216216216216218e-05,
"loss": 1.0764,
"step": 90
},
{
"epoch": 0.180099054479964,
"grad_norm": 0.1794631392027073,
"learning_rate": 1.801801801801802e-05,
"loss": 1.0757,
"step": 100
},
{
"epoch": 0.19810895992796038,
"grad_norm": 0.19497374657829,
"learning_rate": 1.981981981981982e-05,
"loss": 1.0904,
"step": 110
},
{
"epoch": 0.2161188653759568,
"grad_norm": 0.21533071811896856,
"learning_rate": 1.999599507118322e-05,
"loss": 1.0765,
"step": 120
},
{
"epoch": 0.23412877082395317,
"grad_norm": 0.19307788030973838,
"learning_rate": 1.998215499120161e-05,
"loss": 1.0702,
"step": 130
},
{
"epoch": 0.2521386762719496,
"grad_norm": 0.18836481163407426,
"learning_rate": 1.99584439990734e-05,
"loss": 1.0766,
"step": 140
},
{
"epoch": 0.270148581719946,
"grad_norm": 0.21511735322364797,
"learning_rate": 1.992488554155135e-05,
"loss": 1.0811,
"step": 150
},
{
"epoch": 0.28815848716794235,
"grad_norm": 0.195309434953766,
"learning_rate": 1.9881512803111797e-05,
"loss": 1.0658,
"step": 160
},
{
"epoch": 0.30616839261593876,
"grad_norm": 0.2421941539327756,
"learning_rate": 1.9828368673139947e-05,
"loss": 1.0672,
"step": 170
},
{
"epoch": 0.32417829806393517,
"grad_norm": 0.1860112417998846,
"learning_rate": 1.9765505703518494e-05,
"loss": 1.0611,
"step": 180
},
{
"epoch": 0.3421882035119316,
"grad_norm": 0.17818304810152363,
"learning_rate": 1.9692986056661355e-05,
"loss": 1.066,
"step": 190
},
{
"epoch": 0.360198108959928,
"grad_norm": 0.2397415845301638,
"learning_rate": 1.961088144404403e-05,
"loss": 1.0634,
"step": 200
},
{
"epoch": 0.37820801440792434,
"grad_norm": 0.18967589505610974,
"learning_rate": 1.9519273055291266e-05,
"loss": 1.0613,
"step": 210
},
{
"epoch": 0.39621791985592075,
"grad_norm": 0.2130198103194759,
"learning_rate": 1.941825147789225e-05,
"loss": 1.0668,
"step": 220
},
{
"epoch": 0.41422782530391716,
"grad_norm": 0.18395677854198192,
"learning_rate": 1.930791660762262e-05,
"loss": 1.0619,
"step": 230
},
{
"epoch": 0.4322377307519136,
"grad_norm": 0.1865388893667488,
"learning_rate": 1.9188377549761962e-05,
"loss": 1.0636,
"step": 240
},
{
"epoch": 0.45024763619990993,
"grad_norm": 0.18776091331352887,
"learning_rate": 1.90597525112044e-05,
"loss": 1.0514,
"step": 250
},
{
"epoch": 0.46825754164790634,
"grad_norm": 0.18243911298468143,
"learning_rate": 1.8922168683569038e-05,
"loss": 1.0519,
"step": 260
},
{
"epoch": 0.48626744709590275,
"grad_norm": 0.18797424109818742,
"learning_rate": 1.8775762117425777e-05,
"loss": 1.0544,
"step": 270
},
{
"epoch": 0.5042773525438992,
"grad_norm": 0.18906571186079896,
"learning_rate": 1.8620677587760916e-05,
"loss": 1.0567,
"step": 280
},
{
"epoch": 0.5222872579918956,
"grad_norm": 0.18393014193254023,
"learning_rate": 1.8457068450815562e-05,
"loss": 1.0567,
"step": 290
},
{
"epoch": 0.540297163439892,
"grad_norm": 0.19328266852176415,
"learning_rate": 1.8285096492438424e-05,
"loss": 1.055,
"step": 300
},
{
"epoch": 0.5583070688878884,
"grad_norm": 0.18341259531456078,
"learning_rate": 1.810493176810292e-05,
"loss": 1.0537,
"step": 310
},
{
"epoch": 0.5763169743358847,
"grad_norm": 0.19070534973528358,
"learning_rate": 1.7916752434746856e-05,
"loss": 1.0388,
"step": 320
},
{
"epoch": 0.5943268797838811,
"grad_norm": 0.18919026633766578,
"learning_rate": 1.7720744574600865e-05,
"loss": 1.054,
"step": 330
},
{
"epoch": 0.6123367852318775,
"grad_norm": 0.2324366480100927,
"learning_rate": 1.7517102011179935e-05,
"loss": 1.0521,
"step": 340
},
{
"epoch": 0.6303466906798739,
"grad_norm": 0.18954585112855193,
"learning_rate": 1.730602611761989e-05,
"loss": 1.0442,
"step": 350
},
{
"epoch": 0.6483565961278703,
"grad_norm": 0.19841527530500794,
"learning_rate": 1.7087725617548385e-05,
"loss": 1.0461,
"step": 360
},
{
"epoch": 0.6663665015758667,
"grad_norm": 0.18229995357270518,
"learning_rate": 1.686241637868734e-05,
"loss": 1.0416,
"step": 370
},
{
"epoch": 0.6843764070238632,
"grad_norm": 0.18303392694652232,
"learning_rate": 1.6630321199390868e-05,
"loss": 1.0444,
"step": 380
},
{
"epoch": 0.7023863124718596,
"grad_norm": 0.18252570663629492,
"learning_rate": 1.639166958832985e-05,
"loss": 1.0382,
"step": 390
},
{
"epoch": 0.720396217919856,
"grad_norm": 0.1920090778916924,
"learning_rate": 1.6146697537540926e-05,
"loss": 1.0451,
"step": 400
},
{
"epoch": 0.7384061233678523,
"grad_norm": 0.182014948065573,
"learning_rate": 1.5895647289064395e-05,
"loss": 1.0459,
"step": 410
},
{
"epoch": 0.7564160288158487,
"grad_norm": 0.18069534485817443,
"learning_rate": 1.5638767095401778e-05,
"loss": 1.0414,
"step": 420
},
{
"epoch": 0.7744259342638451,
"grad_norm": 0.1970419139073729,
"learning_rate": 1.5376310974029872e-05,
"loss": 1.0379,
"step": 430
},
{
"epoch": 0.7924358397118415,
"grad_norm": 0.1831244876657377,
"learning_rate": 1.5108538456214088e-05,
"loss": 1.038,
"step": 440
},
{
"epoch": 0.8104457451598379,
"grad_norm": 0.18766104093935895,
"learning_rate": 1.4835714330369445e-05,
"loss": 1.0396,
"step": 450
},
{
"epoch": 0.8284556506078343,
"grad_norm": 0.20205755605640713,
"learning_rate": 1.4558108380223013e-05,
"loss": 1.0312,
"step": 460
},
{
"epoch": 0.8464655560558307,
"grad_norm": 0.1958644336406297,
"learning_rate": 1.4275995118036694e-05,
"loss": 1.0333,
"step": 470
},
{
"epoch": 0.8644754615038271,
"grad_norm": 0.17729339680552028,
"learning_rate": 1.3989653513154165e-05,
"loss": 1.0429,
"step": 480
},
{
"epoch": 0.8824853669518236,
"grad_norm": 0.17561113142823978,
"learning_rate": 1.3699366716140434e-05,
"loss": 1.0419,
"step": 490
},
{
"epoch": 0.9004952723998199,
"grad_norm": 0.18385449880212262,
"learning_rate": 1.3405421778786738e-05,
"loss": 1.0482,
"step": 500
},
{
"epoch": 0.9185051778478163,
"grad_norm": 0.18110333089233704,
"learning_rate": 1.3108109370257714e-05,
"loss": 1.0539,
"step": 510
},
{
"epoch": 0.9365150832958127,
"grad_norm": 0.18232534077561582,
"learning_rate": 1.2807723489661497e-05,
"loss": 1.0371,
"step": 520
},
{
"epoch": 0.9545249887438091,
"grad_norm": 0.17118485590371502,
"learning_rate": 1.2504561175326986e-05,
"loss": 1.034,
"step": 530
},
{
"epoch": 0.9725348941918055,
"grad_norm": 0.17973311103034112,
"learning_rate": 1.2198922211075779e-05,
"loss": 1.0374,
"step": 540
},
{
"epoch": 0.9905447996398019,
"grad_norm": 0.18714581161352067,
"learning_rate": 1.1891108829779166e-05,
"loss": 1.0317,
"step": 550
},
{
"epoch": 0.9995497523638001,
"eval_loss": 1.039684534072876,
"eval_runtime": 662.247,
"eval_samples_per_second": 11.876,
"eval_steps_per_second": 0.743,
"step": 555
},
{
"epoch": 1.0085547050877983,
"grad_norm": 0.20548749168254346,
"learning_rate": 1.158142541449341e-05,
"loss": 1.0109,
"step": 560
},
{
"epoch": 1.0265646105357946,
"grad_norm": 0.201629063500619,
"learning_rate": 1.1270178197468788e-05,
"loss": 0.9851,
"step": 570
},
{
"epoch": 1.0445745159837911,
"grad_norm": 0.19089253248727517,
"learning_rate": 1.0957674957330043e-05,
"loss": 0.9736,
"step": 580
},
{
"epoch": 1.0625844214317874,
"grad_norm": 0.18420339461964708,
"learning_rate": 1.0644224714727683e-05,
"loss": 0.9918,
"step": 590
},
{
"epoch": 1.080594326879784,
"grad_norm": 0.18488819323563785,
"learning_rate": 1.0330137426761136e-05,
"loss": 0.9832,
"step": 600
},
{
"epoch": 1.0986042323277803,
"grad_norm": 0.18816750176239094,
"learning_rate": 1.0015723680475847e-05,
"loss": 0.9894,
"step": 610
},
{
"epoch": 1.1166141377757768,
"grad_norm": 0.17955677324222288,
"learning_rate": 9.701294385737471e-06,
"loss": 0.9915,
"step": 620
},
{
"epoch": 1.134624043223773,
"grad_norm": 0.1860265308958814,
"learning_rate": 9.38716046778684e-06,
"loss": 0.9702,
"step": 630
},
{
"epoch": 1.1526339486717694,
"grad_norm": 0.18388832465015095,
"learning_rate": 9.073632559779731e-06,
"loss": 0.981,
"step": 640
},
{
"epoch": 1.170643854119766,
"grad_norm": 0.1788808037903788,
"learning_rate": 8.76102069561545e-06,
"loss": 0.9812,
"step": 650
},
{
"epoch": 1.1886537595677622,
"grad_norm": 0.178574258193573,
"learning_rate": 8.449634003358022e-06,
"loss": 0.9836,
"step": 660
},
{
"epoch": 1.2066636650157587,
"grad_norm": 0.18306275208917874,
"learning_rate": 8.13978039955308e-06,
"loss": 0.9831,
"step": 670
},
{
"epoch": 1.224673570463755,
"grad_norm": 0.18918449761657866,
"learning_rate": 7.831766284742807e-06,
"loss": 0.9846,
"step": 680
},
{
"epoch": 1.2426834759117515,
"grad_norm": 0.1855294097858111,
"learning_rate": 7.525896240479977e-06,
"loss": 0.9744,
"step": 690
},
{
"epoch": 1.2606933813597478,
"grad_norm": 0.20185722671737671,
"learning_rate": 7.222472728140695e-06,
"loss": 0.9859,
"step": 700
},
{
"epoch": 1.2787032868077444,
"grad_norm": 0.192482680701874,
"learning_rate": 6.921795789833723e-06,
"loss": 0.9886,
"step": 710
},
{
"epoch": 1.2967131922557407,
"grad_norm": 0.1795264410436132,
"learning_rate": 6.624162751702077e-06,
"loss": 0.9854,
"step": 720
},
{
"epoch": 1.314723097703737,
"grad_norm": 0.18175034374465715,
"learning_rate": 6.329867929910347e-06,
"loss": 0.9836,
"step": 730
},
{
"epoch": 1.3327330031517335,
"grad_norm": 0.18874834848612046,
"learning_rate": 6.039202339608432e-06,
"loss": 0.9792,
"step": 740
},
{
"epoch": 1.3507429085997298,
"grad_norm": 0.1843992268180646,
"learning_rate": 5.752453407159521e-06,
"loss": 0.9794,
"step": 750
},
{
"epoch": 1.3687528140477263,
"grad_norm": 0.17887298529702766,
"learning_rate": 5.469904685916861e-06,
"loss": 0.985,
"step": 760
},
{
"epoch": 1.3867627194957226,
"grad_norm": 0.17428526152643262,
"learning_rate": 5.1918355758303515e-06,
"loss": 0.9793,
"step": 770
},
{
"epoch": 1.4047726249437191,
"grad_norm": 0.17579200186681246,
"learning_rate": 4.918521047160309e-06,
"loss": 0.9797,
"step": 780
},
{
"epoch": 1.4227825303917154,
"grad_norm": 0.17482239840906302,
"learning_rate": 4.650231368571486e-06,
"loss": 0.967,
"step": 790
},
{
"epoch": 1.440792435839712,
"grad_norm": 0.1846232978343707,
"learning_rate": 4.387231839876349e-06,
"loss": 0.9795,
"step": 800
},
{
"epoch": 1.4588023412877082,
"grad_norm": 0.17722992573590463,
"learning_rate": 4.1297825296918145e-06,
"loss": 0.9749,
"step": 810
},
{
"epoch": 1.4768122467357045,
"grad_norm": 0.18740341744136668,
"learning_rate": 3.878138018268867e-06,
"loss": 0.9846,
"step": 820
},
{
"epoch": 1.494822152183701,
"grad_norm": 0.18222853458564284,
"learning_rate": 3.6325471457493956e-06,
"loss": 0.9869,
"step": 830
},
{
"epoch": 1.5128320576316976,
"grad_norm": 0.17604705966690576,
"learning_rate": 3.3932527660991877e-06,
"loss": 0.9791,
"step": 840
},
{
"epoch": 1.530841963079694,
"grad_norm": 0.17129760493794519,
"learning_rate": 3.160491506960344e-06,
"loss": 0.9803,
"step": 850
},
{
"epoch": 1.5488518685276902,
"grad_norm": 0.17743331668980605,
"learning_rate": 2.934493535660677e-06,
"loss": 0.9849,
"step": 860
},
{
"epoch": 1.5668617739756865,
"grad_norm": 0.17206764149200166,
"learning_rate": 2.715482331611393e-06,
"loss": 0.9799,
"step": 870
},
{
"epoch": 1.584871679423683,
"grad_norm": 0.17401212500860314,
"learning_rate": 2.5036744653181755e-06,
"loss": 0.977,
"step": 880
},
{
"epoch": 1.6028815848716795,
"grad_norm": 0.1864847419394007,
"learning_rate": 2.29927938422419e-06,
"loss": 0.9824,
"step": 890
},
{
"epoch": 1.6208914903196758,
"grad_norm": 0.1794444567972596,
"learning_rate": 2.102499205596743e-06,
"loss": 0.9711,
"step": 900
},
{
"epoch": 1.6389013957676721,
"grad_norm": 0.17867218012908836,
"learning_rate": 1.913528516662452e-06,
"loss": 0.9904,
"step": 910
},
{
"epoch": 1.6569113012156687,
"grad_norm": 0.17632077595030837,
"learning_rate": 1.7325541821885383e-06,
"loss": 0.9846,
"step": 920
},
{
"epoch": 1.6749212066636652,
"grad_norm": 0.1749399044308501,
"learning_rate": 1.5597551597004968e-06,
"loss": 0.9759,
"step": 930
},
{
"epoch": 1.6929311121116615,
"grad_norm": 0.1765753424915534,
"learning_rate": 1.3953023225189243e-06,
"loss": 0.9947,
"step": 940
},
{
"epoch": 1.7109410175596578,
"grad_norm": 0.17249203862535814,
"learning_rate": 1.23935829079042e-06,
"loss": 0.987,
"step": 950
},
{
"epoch": 1.728950923007654,
"grad_norm": 0.17083864764712672,
"learning_rate": 1.0920772706797166e-06,
"loss": 0.9655,
"step": 960
},
{
"epoch": 1.7469608284556506,
"grad_norm": 0.17684409732386205,
"learning_rate": 9.536049018820193e-07,
"loss": 0.9741,
"step": 970
},
{
"epoch": 1.7649707339036471,
"grad_norm": 0.17089785327384419,
"learning_rate": 8.240781136063348e-07,
"loss": 0.9699,
"step": 980
},
{
"epoch": 1.7829806393516434,
"grad_norm": 0.17526364109022735,
"learning_rate": 7.03624989172228e-07,
"loss": 0.9884,
"step": 990
},
{
"epoch": 1.8009905447996397,
"grad_norm": 0.17477084621817432,
"learning_rate": 5.923646393538907e-07,
"loss": 0.9762,
"step": 1000
},
{
"epoch": 1.8190004502476362,
"grad_norm": 0.17586999782551316,
"learning_rate": 4.904070845967467e-07,
"loss": 0.9763,
"step": 1010
},
{
"epoch": 1.8370103556956328,
"grad_norm": 0.17351730726317385,
"learning_rate": 3.97853146223105e-07,
"loss": 0.9759,
"step": 1020
},
{
"epoch": 1.855020261143629,
"grad_norm": 0.17245529759955128,
"learning_rate": 3.1479434673440167e-07,
"loss": 0.9762,
"step": 1030
},
{
"epoch": 1.8730301665916254,
"grad_norm": 0.17074305784570018,
"learning_rate": 2.4131281930864006e-07,
"loss": 0.9778,
"step": 1040
},
{
"epoch": 1.8910400720396217,
"grad_norm": 0.1699923858478813,
"learning_rate": 1.7748122658251877e-07,
"loss": 0.9898,
"step": 1050
},
{
"epoch": 1.9090499774876182,
"grad_norm": 0.17560912690299982,
"learning_rate": 1.2336268879856728e-07,
"loss": 0.9764,
"step": 1060
},
{
"epoch": 1.9270598829356147,
"grad_norm": 0.17172727854066605,
"learning_rate": 7.901072138831512e-08,
"loss": 0.9908,
"step": 1070
},
{
"epoch": 1.945069788383611,
"grad_norm": 0.17416282443666103,
"learning_rate": 4.44691820532539e-08,
"loss": 0.9765,
"step": 1080
},
{
"epoch": 1.9630796938316073,
"grad_norm": 0.1719348511857924,
"learning_rate": 1.977222739588891e-08,
"loss": 0.9784,
"step": 1090
},
{
"epoch": 1.9810895992796038,
"grad_norm": 0.1744398791007269,
"learning_rate": 4.944279143784814e-09,
"loss": 0.9841,
"step": 1100
},
{
"epoch": 1.9990995047276003,
"grad_norm": 0.17567978804671192,
"learning_rate": 0.0,
"loss": 0.9698,
"step": 1110
},
{
"epoch": 1.9990995047276003,
"eval_loss": 1.0319925546646118,
"eval_runtime": 659.9931,
"eval_samples_per_second": 11.917,
"eval_steps_per_second": 0.745,
"step": 1110
},
{
"epoch": 1.9990995047276003,
"step": 1110,
"total_flos": 929619498762240.0,
"train_loss": 1.0262996888375497,
"train_runtime": 54210.2289,
"train_samples_per_second": 2.622,
"train_steps_per_second": 0.02
}
],
"logging_steps": 10,
"max_steps": 1110,
"num_input_tokens_seen": 0,
"num_train_epochs": 2,
"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": 929619498762240.0,
"train_batch_size": 1,
"trial_name": null,
"trial_params": null
}