sedrickkeh's picture
End of training
dfb78a1 verified
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 2.9956108266276518,
"eval_steps": 500,
"global_step": 1023,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.029261155815654718,
"grad_norm": 63.33159396582597,
"learning_rate": 5e-06,
"loss": 0.7981,
"step": 10
},
{
"epoch": 0.058522311631309436,
"grad_norm": 1.2985053942795166,
"learning_rate": 5e-06,
"loss": 0.7394,
"step": 20
},
{
"epoch": 0.08778346744696415,
"grad_norm": 1.2994972414920418,
"learning_rate": 5e-06,
"loss": 0.7076,
"step": 30
},
{
"epoch": 0.11704462326261887,
"grad_norm": 1.09154200562291,
"learning_rate": 5e-06,
"loss": 0.6974,
"step": 40
},
{
"epoch": 0.14630577907827358,
"grad_norm": 1.1635059184964134,
"learning_rate": 5e-06,
"loss": 0.6882,
"step": 50
},
{
"epoch": 0.1755669348939283,
"grad_norm": 0.9663792155845926,
"learning_rate": 5e-06,
"loss": 0.6781,
"step": 60
},
{
"epoch": 0.20482809070958302,
"grad_norm": 0.6939099177117586,
"learning_rate": 5e-06,
"loss": 0.6713,
"step": 70
},
{
"epoch": 0.23408924652523774,
"grad_norm": 0.4554483136685933,
"learning_rate": 5e-06,
"loss": 0.6655,
"step": 80
},
{
"epoch": 0.26335040234089246,
"grad_norm": 0.43998686977438867,
"learning_rate": 5e-06,
"loss": 0.6624,
"step": 90
},
{
"epoch": 0.29261155815654716,
"grad_norm": 0.4681462289171194,
"learning_rate": 5e-06,
"loss": 0.6547,
"step": 100
},
{
"epoch": 0.3218727139722019,
"grad_norm": 0.4088427127266188,
"learning_rate": 5e-06,
"loss": 0.6635,
"step": 110
},
{
"epoch": 0.3511338697878566,
"grad_norm": 0.36705252012743494,
"learning_rate": 5e-06,
"loss": 0.6683,
"step": 120
},
{
"epoch": 0.38039502560351135,
"grad_norm": 0.374209969368988,
"learning_rate": 5e-06,
"loss": 0.6522,
"step": 130
},
{
"epoch": 0.40965618141916604,
"grad_norm": 0.3565722412862572,
"learning_rate": 5e-06,
"loss": 0.6529,
"step": 140
},
{
"epoch": 0.4389173372348208,
"grad_norm": 0.34204106396539435,
"learning_rate": 5e-06,
"loss": 0.6483,
"step": 150
},
{
"epoch": 0.4681784930504755,
"grad_norm": 0.38615414130675385,
"learning_rate": 5e-06,
"loss": 0.6502,
"step": 160
},
{
"epoch": 0.49743964886613024,
"grad_norm": 0.3634230078003702,
"learning_rate": 5e-06,
"loss": 0.6552,
"step": 170
},
{
"epoch": 0.5267008046817849,
"grad_norm": 0.3340266332324882,
"learning_rate": 5e-06,
"loss": 0.6549,
"step": 180
},
{
"epoch": 0.5559619604974396,
"grad_norm": 0.35590307178082214,
"learning_rate": 5e-06,
"loss": 0.6434,
"step": 190
},
{
"epoch": 0.5852231163130943,
"grad_norm": 0.3448405016272299,
"learning_rate": 5e-06,
"loss": 0.6446,
"step": 200
},
{
"epoch": 0.6144842721287491,
"grad_norm": 0.3385609115523372,
"learning_rate": 5e-06,
"loss": 0.6511,
"step": 210
},
{
"epoch": 0.6437454279444038,
"grad_norm": 0.3534390098067362,
"learning_rate": 5e-06,
"loss": 0.6392,
"step": 220
},
{
"epoch": 0.6730065837600585,
"grad_norm": 0.3619673231162035,
"learning_rate": 5e-06,
"loss": 0.6391,
"step": 230
},
{
"epoch": 0.7022677395757132,
"grad_norm": 0.3367671267006255,
"learning_rate": 5e-06,
"loss": 0.646,
"step": 240
},
{
"epoch": 0.731528895391368,
"grad_norm": 0.34227548100303745,
"learning_rate": 5e-06,
"loss": 0.6443,
"step": 250
},
{
"epoch": 0.7607900512070227,
"grad_norm": 0.3408993257912956,
"learning_rate": 5e-06,
"loss": 0.6469,
"step": 260
},
{
"epoch": 0.7900512070226774,
"grad_norm": 0.3790106895262509,
"learning_rate": 5e-06,
"loss": 0.6492,
"step": 270
},
{
"epoch": 0.8193123628383321,
"grad_norm": 0.3429230463611281,
"learning_rate": 5e-06,
"loss": 0.648,
"step": 280
},
{
"epoch": 0.8485735186539868,
"grad_norm": 0.33672564322042764,
"learning_rate": 5e-06,
"loss": 0.6402,
"step": 290
},
{
"epoch": 0.8778346744696416,
"grad_norm": 0.34609131513685526,
"learning_rate": 5e-06,
"loss": 0.6321,
"step": 300
},
{
"epoch": 0.9070958302852963,
"grad_norm": 0.34957045568901657,
"learning_rate": 5e-06,
"loss": 0.6452,
"step": 310
},
{
"epoch": 0.936356986100951,
"grad_norm": 0.33437406171376016,
"learning_rate": 5e-06,
"loss": 0.6372,
"step": 320
},
{
"epoch": 0.9656181419166057,
"grad_norm": 0.371533848542589,
"learning_rate": 5e-06,
"loss": 0.6404,
"step": 330
},
{
"epoch": 0.9948792977322605,
"grad_norm": 0.35130336452701355,
"learning_rate": 5e-06,
"loss": 0.6364,
"step": 340
},
{
"epoch": 0.9978054133138259,
"eval_loss": 0.640914797782898,
"eval_runtime": 345.6005,
"eval_samples_per_second": 26.646,
"eval_steps_per_second": 0.417,
"step": 341
},
{
"epoch": 1.025237746891002,
"grad_norm": 0.3692130458298124,
"learning_rate": 5e-06,
"loss": 0.6524,
"step": 350
},
{
"epoch": 1.054498902706657,
"grad_norm": 0.3553119811115944,
"learning_rate": 5e-06,
"loss": 0.6067,
"step": 360
},
{
"epoch": 1.0837600585223117,
"grad_norm": 0.40346799763343727,
"learning_rate": 5e-06,
"loss": 0.6101,
"step": 370
},
{
"epoch": 1.1130212143379663,
"grad_norm": 0.3197358067569843,
"learning_rate": 5e-06,
"loss": 0.6075,
"step": 380
},
{
"epoch": 1.142282370153621,
"grad_norm": 0.40727251989734975,
"learning_rate": 5e-06,
"loss": 0.6127,
"step": 390
},
{
"epoch": 1.1715435259692757,
"grad_norm": 0.3623259217407185,
"learning_rate": 5e-06,
"loss": 0.6142,
"step": 400
},
{
"epoch": 1.2008046817849305,
"grad_norm": 0.3581625285849574,
"learning_rate": 5e-06,
"loss": 0.6139,
"step": 410
},
{
"epoch": 1.2300658376005853,
"grad_norm": 0.36695340881733157,
"learning_rate": 5e-06,
"loss": 0.6021,
"step": 420
},
{
"epoch": 1.2593269934162399,
"grad_norm": 0.3572031445012923,
"learning_rate": 5e-06,
"loss": 0.6122,
"step": 430
},
{
"epoch": 1.2885881492318947,
"grad_norm": 0.34962802000050813,
"learning_rate": 5e-06,
"loss": 0.6048,
"step": 440
},
{
"epoch": 1.3178493050475493,
"grad_norm": 0.35755536354001544,
"learning_rate": 5e-06,
"loss": 0.6086,
"step": 450
},
{
"epoch": 1.347110460863204,
"grad_norm": 0.37570399727715037,
"learning_rate": 5e-06,
"loss": 0.6062,
"step": 460
},
{
"epoch": 1.3763716166788589,
"grad_norm": 0.3228789388457264,
"learning_rate": 5e-06,
"loss": 0.6035,
"step": 470
},
{
"epoch": 1.4056327724945135,
"grad_norm": 0.3634169062758459,
"learning_rate": 5e-06,
"loss": 0.6044,
"step": 480
},
{
"epoch": 1.4348939283101683,
"grad_norm": 0.3502861798040734,
"learning_rate": 5e-06,
"loss": 0.6108,
"step": 490
},
{
"epoch": 1.464155084125823,
"grad_norm": 0.34093533114824826,
"learning_rate": 5e-06,
"loss": 0.5991,
"step": 500
},
{
"epoch": 1.4934162399414777,
"grad_norm": 0.3463701037893924,
"learning_rate": 5e-06,
"loss": 0.6071,
"step": 510
},
{
"epoch": 1.5226773957571325,
"grad_norm": 0.38931672754338176,
"learning_rate": 5e-06,
"loss": 0.6095,
"step": 520
},
{
"epoch": 1.5519385515727873,
"grad_norm": 0.34906068590167777,
"learning_rate": 5e-06,
"loss": 0.608,
"step": 530
},
{
"epoch": 1.5811997073884418,
"grad_norm": 0.3701456606885589,
"learning_rate": 5e-06,
"loss": 0.6146,
"step": 540
},
{
"epoch": 1.6104608632040964,
"grad_norm": 0.3506849233715239,
"learning_rate": 5e-06,
"loss": 0.6056,
"step": 550
},
{
"epoch": 1.6397220190197512,
"grad_norm": 0.35747637775205837,
"learning_rate": 5e-06,
"loss": 0.6074,
"step": 560
},
{
"epoch": 1.668983174835406,
"grad_norm": 0.32158800583674974,
"learning_rate": 5e-06,
"loss": 0.5979,
"step": 570
},
{
"epoch": 1.6982443306510606,
"grad_norm": 0.3528344876208815,
"learning_rate": 5e-06,
"loss": 0.6072,
"step": 580
},
{
"epoch": 1.7275054864667154,
"grad_norm": 0.41985622969098957,
"learning_rate": 5e-06,
"loss": 0.6075,
"step": 590
},
{
"epoch": 1.7567666422823702,
"grad_norm": 0.34105125625368893,
"learning_rate": 5e-06,
"loss": 0.6044,
"step": 600
},
{
"epoch": 1.7860277980980248,
"grad_norm": 0.34589478838024207,
"learning_rate": 5e-06,
"loss": 0.6069,
"step": 610
},
{
"epoch": 1.8152889539136796,
"grad_norm": 0.32553452148702333,
"learning_rate": 5e-06,
"loss": 0.6073,
"step": 620
},
{
"epoch": 1.8445501097293344,
"grad_norm": 0.3583629997971789,
"learning_rate": 5e-06,
"loss": 0.6063,
"step": 630
},
{
"epoch": 1.873811265544989,
"grad_norm": 0.3258762553733896,
"learning_rate": 5e-06,
"loss": 0.6012,
"step": 640
},
{
"epoch": 1.9030724213606436,
"grad_norm": 0.36542172455522237,
"learning_rate": 5e-06,
"loss": 0.6068,
"step": 650
},
{
"epoch": 1.9323335771762986,
"grad_norm": 0.3468540827688793,
"learning_rate": 5e-06,
"loss": 0.6074,
"step": 660
},
{
"epoch": 1.9615947329919532,
"grad_norm": 0.3301547297982186,
"learning_rate": 5e-06,
"loss": 0.5962,
"step": 670
},
{
"epoch": 1.9908558888076078,
"grad_norm": 0.353549889665131,
"learning_rate": 5e-06,
"loss": 0.6107,
"step": 680
},
{
"epoch": 1.9967081199707388,
"eval_loss": 0.6333796381950378,
"eval_runtime": 344.9128,
"eval_samples_per_second": 26.699,
"eval_steps_per_second": 0.417,
"step": 682
},
{
"epoch": 2.0212143379663496,
"grad_norm": 0.38522584628732043,
"learning_rate": 5e-06,
"loss": 0.6214,
"step": 690
},
{
"epoch": 2.050475493782004,
"grad_norm": 0.4092650769853219,
"learning_rate": 5e-06,
"loss": 0.5693,
"step": 700
},
{
"epoch": 2.0797366495976592,
"grad_norm": 0.36395160858693937,
"learning_rate": 5e-06,
"loss": 0.5712,
"step": 710
},
{
"epoch": 2.108997805413314,
"grad_norm": 0.36987210556495065,
"learning_rate": 5e-06,
"loss": 0.5723,
"step": 720
},
{
"epoch": 2.1382589612289684,
"grad_norm": 0.3530004884029566,
"learning_rate": 5e-06,
"loss": 0.5741,
"step": 730
},
{
"epoch": 2.1675201170446234,
"grad_norm": 0.3474755062723768,
"learning_rate": 5e-06,
"loss": 0.5794,
"step": 740
},
{
"epoch": 2.196781272860278,
"grad_norm": 0.3203177683782767,
"learning_rate": 5e-06,
"loss": 0.5781,
"step": 750
},
{
"epoch": 2.2260424286759326,
"grad_norm": 0.3661319457012291,
"learning_rate": 5e-06,
"loss": 0.5755,
"step": 760
},
{
"epoch": 2.255303584491587,
"grad_norm": 0.3359421587293624,
"learning_rate": 5e-06,
"loss": 0.5742,
"step": 770
},
{
"epoch": 2.284564740307242,
"grad_norm": 0.3332624185731929,
"learning_rate": 5e-06,
"loss": 0.5733,
"step": 780
},
{
"epoch": 2.313825896122897,
"grad_norm": 0.3326965713448184,
"learning_rate": 5e-06,
"loss": 0.5716,
"step": 790
},
{
"epoch": 2.3430870519385514,
"grad_norm": 0.3617619945788122,
"learning_rate": 5e-06,
"loss": 0.5794,
"step": 800
},
{
"epoch": 2.3723482077542064,
"grad_norm": 0.3298127671828895,
"learning_rate": 5e-06,
"loss": 0.5746,
"step": 810
},
{
"epoch": 2.401609363569861,
"grad_norm": 0.332422697888563,
"learning_rate": 5e-06,
"loss": 0.5808,
"step": 820
},
{
"epoch": 2.4308705193855156,
"grad_norm": 0.3196015454900321,
"learning_rate": 5e-06,
"loss": 0.5731,
"step": 830
},
{
"epoch": 2.4601316752011706,
"grad_norm": 0.35331559179835964,
"learning_rate": 5e-06,
"loss": 0.5736,
"step": 840
},
{
"epoch": 2.489392831016825,
"grad_norm": 0.3392297184730635,
"learning_rate": 5e-06,
"loss": 0.5758,
"step": 850
},
{
"epoch": 2.5186539868324798,
"grad_norm": 0.34620746539815944,
"learning_rate": 5e-06,
"loss": 0.5688,
"step": 860
},
{
"epoch": 2.547915142648135,
"grad_norm": 0.4023501023219655,
"learning_rate": 5e-06,
"loss": 0.5712,
"step": 870
},
{
"epoch": 2.5771762984637894,
"grad_norm": 0.34088303079875626,
"learning_rate": 5e-06,
"loss": 0.5754,
"step": 880
},
{
"epoch": 2.606437454279444,
"grad_norm": 0.36405494880778144,
"learning_rate": 5e-06,
"loss": 0.5748,
"step": 890
},
{
"epoch": 2.6356986100950985,
"grad_norm": 0.34721347207329656,
"learning_rate": 5e-06,
"loss": 0.5794,
"step": 900
},
{
"epoch": 2.6649597659107536,
"grad_norm": 0.3546551945507394,
"learning_rate": 5e-06,
"loss": 0.5795,
"step": 910
},
{
"epoch": 2.694220921726408,
"grad_norm": 0.36027986998179906,
"learning_rate": 5e-06,
"loss": 0.5738,
"step": 920
},
{
"epoch": 2.723482077542063,
"grad_norm": 0.35159220496145294,
"learning_rate": 5e-06,
"loss": 0.5775,
"step": 930
},
{
"epoch": 2.7527432333577178,
"grad_norm": 0.3504714185096239,
"learning_rate": 5e-06,
"loss": 0.5764,
"step": 940
},
{
"epoch": 2.7820043891733723,
"grad_norm": 0.3284737703409318,
"learning_rate": 5e-06,
"loss": 0.5751,
"step": 950
},
{
"epoch": 2.811265544989027,
"grad_norm": 0.37305249319315187,
"learning_rate": 5e-06,
"loss": 0.5789,
"step": 960
},
{
"epoch": 2.840526700804682,
"grad_norm": 0.33884604095107257,
"learning_rate": 5e-06,
"loss": 0.5747,
"step": 970
},
{
"epoch": 2.8697878566203365,
"grad_norm": 0.37958678053577166,
"learning_rate": 5e-06,
"loss": 0.5745,
"step": 980
},
{
"epoch": 2.899049012435991,
"grad_norm": 0.34410472212682797,
"learning_rate": 5e-06,
"loss": 0.5675,
"step": 990
},
{
"epoch": 2.928310168251646,
"grad_norm": 0.3222728501974356,
"learning_rate": 5e-06,
"loss": 0.5753,
"step": 1000
},
{
"epoch": 2.9575713240673007,
"grad_norm": 0.3391773666930332,
"learning_rate": 5e-06,
"loss": 0.567,
"step": 1010
},
{
"epoch": 2.9868324798829553,
"grad_norm": 0.33304266357077267,
"learning_rate": 5e-06,
"loss": 0.5787,
"step": 1020
},
{
"epoch": 2.9956108266276518,
"eval_loss": 0.6330468654632568,
"eval_runtime": 344.0095,
"eval_samples_per_second": 26.77,
"eval_steps_per_second": 0.419,
"step": 1023
},
{
"epoch": 2.9956108266276518,
"step": 1023,
"total_flos": 2144987064041472.0,
"train_loss": 0.6152277645360923,
"train_runtime": 54853.139,
"train_samples_per_second": 9.569,
"train_steps_per_second": 0.019
}
],
"logging_steps": 10,
"max_steps": 1023,
"num_input_tokens_seen": 0,
"num_train_epochs": 3,
"save_steps": 500,
"stateful_callbacks": {
"TrainerControl": {
"args": {
"should_epoch_stop": false,
"should_evaluate": false,
"should_log": false,
"should_save": true,
"should_training_stop": true
},
"attributes": {}
}
},
"total_flos": 2144987064041472.0,
"train_batch_size": 8,
"trial_name": null,
"trial_params": null
}