llama5pEnt / trainer_state.json
terry69's picture
Model save
30811ac verified
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.9697529438928654,
"eval_steps": 500,
"global_step": 525,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0018471484645578389,
"grad_norm": 0.2646295130252838,
"learning_rate": 3.636363636363636e-06,
"loss": 0.9117,
"step": 1
},
{
"epoch": 0.009235742322789195,
"grad_norm": 0.29179853200912476,
"learning_rate": 1.8181818181818182e-05,
"loss": 0.9579,
"step": 5
},
{
"epoch": 0.01847148464557839,
"grad_norm": 0.30512726306915283,
"learning_rate": 3.6363636363636364e-05,
"loss": 0.9489,
"step": 10
},
{
"epoch": 0.027707226968367582,
"grad_norm": 0.08514489233493805,
"learning_rate": 5.4545454545454546e-05,
"loss": 0.9274,
"step": 15
},
{
"epoch": 0.03694296929115678,
"grad_norm": 0.07719692587852478,
"learning_rate": 7.272727272727273e-05,
"loss": 0.9257,
"step": 20
},
{
"epoch": 0.04617871161394597,
"grad_norm": 0.08602738380432129,
"learning_rate": 9.090909090909092e-05,
"loss": 0.9255,
"step": 25
},
{
"epoch": 0.055414453936735164,
"grad_norm": 0.08664494752883911,
"learning_rate": 0.00010909090909090909,
"loss": 0.8889,
"step": 30
},
{
"epoch": 0.06465019625952435,
"grad_norm": 0.07322084158658981,
"learning_rate": 0.00012727272727272728,
"loss": 0.8809,
"step": 35
},
{
"epoch": 0.07388593858231356,
"grad_norm": 0.08361256867647171,
"learning_rate": 0.00014545454545454546,
"loss": 0.8901,
"step": 40
},
{
"epoch": 0.08312168090510275,
"grad_norm": 0.0712098553776741,
"learning_rate": 0.00016363636363636366,
"loss": 0.8712,
"step": 45
},
{
"epoch": 0.09235742322789194,
"grad_norm": 0.07038867473602295,
"learning_rate": 0.00018181818181818183,
"loss": 0.8779,
"step": 50
},
{
"epoch": 0.10159316555068114,
"grad_norm": 0.08133558183908463,
"learning_rate": 0.0002,
"loss": 0.8549,
"step": 55
},
{
"epoch": 0.11082890787347033,
"grad_norm": 0.07199383527040482,
"learning_rate": 0.00019994777247895855,
"loss": 0.8655,
"step": 60
},
{
"epoch": 0.12006465019625952,
"grad_norm": 0.08243429660797119,
"learning_rate": 0.00019979114447011323,
"loss": 0.891,
"step": 65
},
{
"epoch": 0.1293003925190487,
"grad_norm": 0.0855724886059761,
"learning_rate": 0.00019953027957931658,
"loss": 0.8675,
"step": 70
},
{
"epoch": 0.1385361348418379,
"grad_norm": 0.0847950428724289,
"learning_rate": 0.00019916545029310012,
"loss": 0.8579,
"step": 75
},
{
"epoch": 0.1477718771646271,
"grad_norm": 0.07739102840423584,
"learning_rate": 0.00019869703769404828,
"loss": 0.8643,
"step": 80
},
{
"epoch": 0.1570076194874163,
"grad_norm": 0.08714427053928375,
"learning_rate": 0.00019812553106273847,
"loss": 0.8766,
"step": 85
},
{
"epoch": 0.1662433618102055,
"grad_norm": 0.06926970183849335,
"learning_rate": 0.00019745152736666302,
"loss": 0.8539,
"step": 90
},
{
"epoch": 0.1754791041329947,
"grad_norm": 0.08098109066486359,
"learning_rate": 0.0001966757306366662,
"loss": 0.8823,
"step": 95
},
{
"epoch": 0.18471484645578387,
"grad_norm": 0.09138432890176773,
"learning_rate": 0.0001957989512315489,
"loss": 0.8601,
"step": 100
},
{
"epoch": 0.19395058877857307,
"grad_norm": 0.09067590534687042,
"learning_rate": 0.00019482210499160765,
"loss": 0.8551,
"step": 105
},
{
"epoch": 0.20318633110136228,
"grad_norm": 0.08557430654764175,
"learning_rate": 0.0001937462122819935,
"loss": 0.8582,
"step": 110
},
{
"epoch": 0.21242207342415145,
"grad_norm": 0.07583803683519363,
"learning_rate": 0.00019257239692688907,
"loss": 0.8468,
"step": 115
},
{
"epoch": 0.22165781574694066,
"grad_norm": 0.10376883298158646,
"learning_rate": 0.00019130188503561741,
"loss": 0.8722,
"step": 120
},
{
"epoch": 0.23089355806972986,
"grad_norm": 0.07388196140527725,
"learning_rate": 0.00018993600372190932,
"loss": 0.8715,
"step": 125
},
{
"epoch": 0.24012930039251904,
"grad_norm": 0.07768449187278748,
"learning_rate": 0.00018847617971766577,
"loss": 0.8632,
"step": 130
},
{
"epoch": 0.24936504271530824,
"grad_norm": 0.07902154326438904,
"learning_rate": 0.00018692393788266479,
"loss": 0.8746,
"step": 135
},
{
"epoch": 0.2586007850380974,
"grad_norm": 0.07420278340578079,
"learning_rate": 0.0001852808996117683,
"loss": 0.8676,
"step": 140
},
{
"epoch": 0.26783652736088664,
"grad_norm": 0.07562129944562912,
"learning_rate": 0.00018354878114129367,
"loss": 0.8829,
"step": 145
},
{
"epoch": 0.2770722696836758,
"grad_norm": 0.07764951884746552,
"learning_rate": 0.00018172939175631808,
"loss": 0.8766,
"step": 150
},
{
"epoch": 0.286308012006465,
"grad_norm": 0.07714657485485077,
"learning_rate": 0.0001798246319007893,
"loss": 0.8802,
"step": 155
},
{
"epoch": 0.2955437543292542,
"grad_norm": 0.08166080713272095,
"learning_rate": 0.00017783649119241602,
"loss": 0.8504,
"step": 160
},
{
"epoch": 0.3047794966520434,
"grad_norm": 0.07295581698417664,
"learning_rate": 0.0001757670463444118,
"loss": 0.8578,
"step": 165
},
{
"epoch": 0.3140152389748326,
"grad_norm": 0.08463213592767715,
"learning_rate": 0.00017361845899626355,
"loss": 0.8465,
"step": 170
},
{
"epoch": 0.3232509812976218,
"grad_norm": 0.07825674116611481,
"learning_rate": 0.00017139297345578994,
"loss": 0.8581,
"step": 175
},
{
"epoch": 0.332486723620411,
"grad_norm": 0.08157385140657425,
"learning_rate": 0.0001690929143548488,
"loss": 0.8598,
"step": 180
},
{
"epoch": 0.34172246594320016,
"grad_norm": 0.08528893440961838,
"learning_rate": 0.00016672068422114196,
"loss": 0.85,
"step": 185
},
{
"epoch": 0.3509582082659894,
"grad_norm": 0.07343052327632904,
"learning_rate": 0.00016427876096865394,
"loss": 0.8515,
"step": 190
},
{
"epoch": 0.36019395058877857,
"grad_norm": 0.07675015926361084,
"learning_rate": 0.00016176969530934572,
"loss": 0.8628,
"step": 195
},
{
"epoch": 0.36942969291156774,
"grad_norm": 0.07422856241464615,
"learning_rate": 0.0001591961080888076,
"loss": 0.866,
"step": 200
},
{
"epoch": 0.378665435234357,
"grad_norm": 0.0853537917137146,
"learning_rate": 0.00015656068754865387,
"loss": 0.8652,
"step": 205
},
{
"epoch": 0.38790117755714615,
"grad_norm": 0.07879694551229477,
"learning_rate": 0.0001538661865185188,
"loss": 0.8613,
"step": 210
},
{
"epoch": 0.3971369198799353,
"grad_norm": 0.0811944380402565,
"learning_rate": 0.00015111541954058734,
"loss": 0.8723,
"step": 215
},
{
"epoch": 0.40637266220272455,
"grad_norm": 0.08038283884525299,
"learning_rate": 0.00014831125992966385,
"loss": 0.8709,
"step": 220
},
{
"epoch": 0.41560840452551373,
"grad_norm": 0.07797664403915405,
"learning_rate": 0.00014545663677185006,
"loss": 0.8583,
"step": 225
},
{
"epoch": 0.4248441468483029,
"grad_norm": 0.08307761698961258,
"learning_rate": 0.00014255453186496673,
"loss": 0.8467,
"step": 230
},
{
"epoch": 0.43407988917109214,
"grad_norm": 0.07691530138254166,
"learning_rate": 0.0001396079766039157,
"loss": 0.8435,
"step": 235
},
{
"epoch": 0.4433156314938813,
"grad_norm": 0.09281527251005173,
"learning_rate": 0.0001366200488142348,
"loss": 0.8605,
"step": 240
},
{
"epoch": 0.4525513738166705,
"grad_norm": 0.07799265533685684,
"learning_rate": 0.00013359386953715421,
"loss": 0.85,
"step": 245
},
{
"epoch": 0.4617871161394597,
"grad_norm": 0.07675463706254959,
"learning_rate": 0.00013053259976951133,
"loss": 0.8434,
"step": 250
},
{
"epoch": 0.4710228584622489,
"grad_norm": 0.09029978513717651,
"learning_rate": 0.00012743943716193016,
"loss": 0.8587,
"step": 255
},
{
"epoch": 0.48025860078503807,
"grad_norm": 0.07997170835733414,
"learning_rate": 0.00012431761267871417,
"loss": 0.8436,
"step": 260
},
{
"epoch": 0.4894943431078273,
"grad_norm": 0.08595962822437286,
"learning_rate": 0.0001211703872229411,
"loss": 0.8682,
"step": 265
},
{
"epoch": 0.4987300854306165,
"grad_norm": 0.08949116617441177,
"learning_rate": 0.00011800104823028515,
"loss": 0.8663,
"step": 270
},
{
"epoch": 0.5079658277534057,
"grad_norm": 0.08509814739227295,
"learning_rate": 0.0001148129062351249,
"loss": 0.8359,
"step": 275
},
{
"epoch": 0.5172015700761948,
"grad_norm": 0.08007588982582092,
"learning_rate": 0.00011160929141252303,
"loss": 0.8494,
"step": 280
},
{
"epoch": 0.5264373123989841,
"grad_norm": 0.07660423964262009,
"learning_rate": 0.00010839355009969068,
"loss": 0.8403,
"step": 285
},
{
"epoch": 0.5356730547217733,
"grad_norm": 0.07779201865196228,
"learning_rate": 0.00010516904130056946,
"loss": 0.863,
"step": 290
},
{
"epoch": 0.5449087970445624,
"grad_norm": 0.07425861805677414,
"learning_rate": 0.00010193913317718244,
"loss": 0.8732,
"step": 295
},
{
"epoch": 0.5541445393673516,
"grad_norm": 0.08780544251203537,
"learning_rate": 9.870719953141917e-05,
"loss": 0.856,
"step": 300
},
{
"epoch": 0.5633802816901409,
"grad_norm": 0.07546891272068024,
"learning_rate": 9.547661628092937e-05,
"loss": 0.8418,
"step": 305
},
{
"epoch": 0.57261602401293,
"grad_norm": 0.08621185272932053,
"learning_rate": 9.225075793280692e-05,
"loss": 0.8463,
"step": 310
},
{
"epoch": 0.5818517663357192,
"grad_norm": 0.07618840038776398,
"learning_rate": 8.903299405874684e-05,
"loss": 0.8257,
"step": 315
},
{
"epoch": 0.5910875086585085,
"grad_norm": 0.07749903202056885,
"learning_rate": 8.582668577535797e-05,
"loss": 0.8442,
"step": 320
},
{
"epoch": 0.6003232509812976,
"grad_norm": 0.08626076579093933,
"learning_rate": 8.263518223330697e-05,
"loss": 0.8739,
"step": 325
},
{
"epoch": 0.6095589933040868,
"grad_norm": 0.07912192493677139,
"learning_rate": 7.94618171189618e-05,
"loss": 0.816,
"step": 330
},
{
"epoch": 0.618794735626876,
"grad_norm": 0.08324088156223297,
"learning_rate": 7.630990517218808e-05,
"loss": 0.853,
"step": 335
},
{
"epoch": 0.6280304779496652,
"grad_norm": 0.08155480027198792,
"learning_rate": 7.318273872393625e-05,
"loss": 0.86,
"step": 340
},
{
"epoch": 0.6372662202724544,
"grad_norm": 0.10541233420372009,
"learning_rate": 7.008358425723585e-05,
"loss": 0.8674,
"step": 345
},
{
"epoch": 0.6465019625952436,
"grad_norm": 0.08856544643640518,
"learning_rate": 6.701567899518924e-05,
"loss": 0.8542,
"step": 350
},
{
"epoch": 0.6557377049180327,
"grad_norm": 0.0975981131196022,
"learning_rate": 6.398222751952899e-05,
"loss": 0.8441,
"step": 355
},
{
"epoch": 0.664973447240822,
"grad_norm": 0.0872284546494484,
"learning_rate": 6.098639842327052e-05,
"loss": 0.8661,
"step": 360
},
{
"epoch": 0.6742091895636112,
"grad_norm": 0.08079314976930618,
"learning_rate": 5.80313210009571e-05,
"loss": 0.8465,
"step": 365
},
{
"epoch": 0.6834449318864003,
"grad_norm": 0.08326224982738495,
"learning_rate": 5.5120081979953785e-05,
"loss": 0.8551,
"step": 370
},
{
"epoch": 0.6926806742091896,
"grad_norm": 0.07942940294742584,
"learning_rate": 5.22557222962051e-05,
"loss": 0.8673,
"step": 375
},
{
"epoch": 0.7019164165319788,
"grad_norm": 0.08273901045322418,
"learning_rate": 4.9441233917824106e-05,
"loss": 0.8424,
"step": 380
},
{
"epoch": 0.7111521588547679,
"grad_norm": 0.07538026571273804,
"learning_rate": 4.66795567198309e-05,
"loss": 0.8448,
"step": 385
},
{
"epoch": 0.7203879011775571,
"grad_norm": 0.07920888811349869,
"learning_rate": 4.397357541330476e-05,
"loss": 0.874,
"step": 390
},
{
"epoch": 0.7296236435003464,
"grad_norm": 0.09400610625743866,
"learning_rate": 4.132611653215822e-05,
"loss": 0.8487,
"step": 395
},
{
"epoch": 0.7388593858231355,
"grad_norm": 0.08254476636648178,
"learning_rate": 3.873994548067972e-05,
"loss": 0.836,
"step": 400
},
{
"epoch": 0.7480951281459247,
"grad_norm": 0.08779824525117874,
"learning_rate": 3.621776364492939e-05,
"loss": 0.8621,
"step": 405
},
{
"epoch": 0.757330870468714,
"grad_norm": 0.08715569227933884,
"learning_rate": 3.376220557100523e-05,
"loss": 0.8413,
"step": 410
},
{
"epoch": 0.7665666127915031,
"grad_norm": 0.08005767315626144,
"learning_rate": 3.137583621312665e-05,
"loss": 0.8563,
"step": 415
},
{
"epoch": 0.7758023551142923,
"grad_norm": 0.08362836390733719,
"learning_rate": 2.906114825441072e-05,
"loss": 0.8431,
"step": 420
},
{
"epoch": 0.7850380974370815,
"grad_norm": 0.08525697141885757,
"learning_rate": 2.6820559503138797e-05,
"loss": 0.8619,
"step": 425
},
{
"epoch": 0.7942738397598706,
"grad_norm": 0.08668297529220581,
"learning_rate": 2.465641036723393e-05,
"loss": 0.8525,
"step": 430
},
{
"epoch": 0.8035095820826599,
"grad_norm": 0.08923713862895966,
"learning_rate": 2.2570961409586754e-05,
"loss": 0.854,
"step": 435
},
{
"epoch": 0.8127453244054491,
"grad_norm": 0.08294524252414703,
"learning_rate": 2.0566390986783646e-05,
"loss": 0.867,
"step": 440
},
{
"epoch": 0.8219810667282382,
"grad_norm": 0.08329101651906967,
"learning_rate": 1.864479297370325e-05,
"loss": 0.8454,
"step": 445
},
{
"epoch": 0.8312168090510275,
"grad_norm": 0.08947557955980301,
"learning_rate": 1.6808174576358848e-05,
"loss": 0.8663,
"step": 450
},
{
"epoch": 0.8404525513738167,
"grad_norm": 0.08388309925794601,
"learning_rate": 1.505845423527027e-05,
"loss": 0.8663,
"step": 455
},
{
"epoch": 0.8496882936966058,
"grad_norm": 0.07886148244142532,
"learning_rate": 1.339745962155613e-05,
"loss": 0.8484,
"step": 460
},
{
"epoch": 0.858924036019395,
"grad_norm": 0.08021736145019531,
"learning_rate": 1.18269257278392e-05,
"loss": 0.8613,
"step": 465
},
{
"epoch": 0.8681597783421843,
"grad_norm": 0.0815872773528099,
"learning_rate": 1.0348493055959062e-05,
"loss": 0.8361,
"step": 470
},
{
"epoch": 0.8773955206649734,
"grad_norm": 0.08512595295906067,
"learning_rate": 8.963705903385345e-06,
"loss": 0.8236,
"step": 475
},
{
"epoch": 0.8866312629877626,
"grad_norm": 0.07890893518924713,
"learning_rate": 7.674010750120964e-06,
"loss": 0.8398,
"step": 480
},
{
"epoch": 0.8958670053105519,
"grad_norm": 0.08390273153781891,
"learning_rate": 6.480754747781037e-06,
"loss": 0.8392,
"step": 485
},
{
"epoch": 0.905102747633341,
"grad_norm": 0.07941864430904388,
"learning_rate": 5.385184312424974e-06,
"loss": 0.8344,
"step": 490
},
{
"epoch": 0.9143384899561302,
"grad_norm": 0.0822635293006897,
"learning_rate": 4.3884438226120424e-06,
"loss": 0.8492,
"step": 495
},
{
"epoch": 0.9235742322789194,
"grad_norm": 0.07574615627527237,
"learning_rate": 3.4915744240403558e-06,
"loss": 0.8499,
"step": 500
},
{
"epoch": 0.9328099746017086,
"grad_norm": 0.08466410636901855,
"learning_rate": 2.6955129420176196e-06,
"loss": 0.848,
"step": 505
},
{
"epoch": 0.9420457169244978,
"grad_norm": 0.0831914022564888,
"learning_rate": 2.0010909028998827e-06,
"loss": 0.8442,
"step": 510
},
{
"epoch": 0.951281459247287,
"grad_norm": 0.08431433886289597,
"learning_rate": 1.409033665520354e-06,
"loss": 0.8491,
"step": 515
},
{
"epoch": 0.9605172015700761,
"grad_norm": 0.08757560700178146,
"learning_rate": 9.199596635154683e-07,
"loss": 0.8526,
"step": 520
},
{
"epoch": 0.9697529438928654,
"grad_norm": 0.0839475765824318,
"learning_rate": 5.343797593398536e-07,
"loss": 0.8663,
"step": 525
},
{
"epoch": 0.9697529438928654,
"step": 525,
"total_flos": 2.6262151886631076e+18,
"train_loss": 0.0,
"train_runtime": 0.0108,
"train_samples_per_second": 9612825.336,
"train_steps_per_second": 37551.544
}
],
"logging_steps": 5,
"max_steps": 406,
"num_input_tokens_seen": 0,
"num_train_epochs": 1,
"save_steps": 25,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": false
},
"attributes": {}
}
},
"total_flos": 2.6262151886631076e+18,
"train_batch_size": 8,
"trial_name": null,
"trial_params": null
}