sedrickkeh's picture
End of training
bce774d 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": 1.2657584959387165,
"learning_rate": 5e-06,
"loss": 0.7988,
"step": 10
},
{
"epoch": 0.058522311631309436,
"grad_norm": 1.2704339798031188,
"learning_rate": 5e-06,
"loss": 0.7387,
"step": 20
},
{
"epoch": 0.08778346744696415,
"grad_norm": 1.2825229206893989,
"learning_rate": 5e-06,
"loss": 0.7076,
"step": 30
},
{
"epoch": 0.11704462326261887,
"grad_norm": 1.0912743927578166,
"learning_rate": 5e-06,
"loss": 0.6977,
"step": 40
},
{
"epoch": 0.14630577907827358,
"grad_norm": 1.1685966080875834,
"learning_rate": 5e-06,
"loss": 0.6889,
"step": 50
},
{
"epoch": 0.1755669348939283,
"grad_norm": 1.0079311478650694,
"learning_rate": 5e-06,
"loss": 0.6791,
"step": 60
},
{
"epoch": 0.20482809070958302,
"grad_norm": 0.7674137568239411,
"learning_rate": 5e-06,
"loss": 0.6724,
"step": 70
},
{
"epoch": 0.23408924652523774,
"grad_norm": 0.503890866518598,
"learning_rate": 5e-06,
"loss": 0.6662,
"step": 80
},
{
"epoch": 0.26335040234089246,
"grad_norm": 0.45260557253554135,
"learning_rate": 5e-06,
"loss": 0.6628,
"step": 90
},
{
"epoch": 0.29261155815654716,
"grad_norm": 0.4646839382593406,
"learning_rate": 5e-06,
"loss": 0.6549,
"step": 100
},
{
"epoch": 0.3218727139722019,
"grad_norm": 0.4222200033184708,
"learning_rate": 5e-06,
"loss": 0.6636,
"step": 110
},
{
"epoch": 0.3511338697878566,
"grad_norm": 0.38506546794145063,
"learning_rate": 5e-06,
"loss": 0.6685,
"step": 120
},
{
"epoch": 0.38039502560351135,
"grad_norm": 0.38917663709800554,
"learning_rate": 5e-06,
"loss": 0.6523,
"step": 130
},
{
"epoch": 0.40965618141916604,
"grad_norm": 0.3691000175174078,
"learning_rate": 5e-06,
"loss": 0.653,
"step": 140
},
{
"epoch": 0.4389173372348208,
"grad_norm": 0.3439949566786501,
"learning_rate": 5e-06,
"loss": 0.6484,
"step": 150
},
{
"epoch": 0.4681784930504755,
"grad_norm": 0.387115974740115,
"learning_rate": 5e-06,
"loss": 0.6503,
"step": 160
},
{
"epoch": 0.49743964886613024,
"grad_norm": 0.36854195191157635,
"learning_rate": 5e-06,
"loss": 0.6554,
"step": 170
},
{
"epoch": 0.5267008046817849,
"grad_norm": 0.33925721666948355,
"learning_rate": 5e-06,
"loss": 0.655,
"step": 180
},
{
"epoch": 0.5559619604974396,
"grad_norm": 0.35725360333114753,
"learning_rate": 5e-06,
"loss": 0.6435,
"step": 190
},
{
"epoch": 0.5852231163130943,
"grad_norm": 0.34358088975665035,
"learning_rate": 5e-06,
"loss": 0.6447,
"step": 200
},
{
"epoch": 0.6144842721287491,
"grad_norm": 0.3416273694800734,
"learning_rate": 5e-06,
"loss": 0.6513,
"step": 210
},
{
"epoch": 0.6437454279444038,
"grad_norm": 0.35393765203156324,
"learning_rate": 5e-06,
"loss": 0.6393,
"step": 220
},
{
"epoch": 0.6730065837600585,
"grad_norm": 0.354923204859383,
"learning_rate": 5e-06,
"loss": 0.6392,
"step": 230
},
{
"epoch": 0.7022677395757132,
"grad_norm": 0.33619322547857633,
"learning_rate": 5e-06,
"loss": 0.6461,
"step": 240
},
{
"epoch": 0.731528895391368,
"grad_norm": 0.34788354830459295,
"learning_rate": 5e-06,
"loss": 0.6445,
"step": 250
},
{
"epoch": 0.7607900512070227,
"grad_norm": 0.3443841612737956,
"learning_rate": 5e-06,
"loss": 0.6471,
"step": 260
},
{
"epoch": 0.7900512070226774,
"grad_norm": 0.37385459607166616,
"learning_rate": 5e-06,
"loss": 0.6493,
"step": 270
},
{
"epoch": 0.8193123628383321,
"grad_norm": 0.338394354503849,
"learning_rate": 5e-06,
"loss": 0.6481,
"step": 280
},
{
"epoch": 0.8485735186539868,
"grad_norm": 0.336513498864063,
"learning_rate": 5e-06,
"loss": 0.6404,
"step": 290
},
{
"epoch": 0.8778346744696416,
"grad_norm": 0.34543333996062836,
"learning_rate": 5e-06,
"loss": 0.6322,
"step": 300
},
{
"epoch": 0.9070958302852963,
"grad_norm": 0.3495739473223406,
"learning_rate": 5e-06,
"loss": 0.6454,
"step": 310
},
{
"epoch": 0.936356986100951,
"grad_norm": 0.33400664858087803,
"learning_rate": 5e-06,
"loss": 0.6373,
"step": 320
},
{
"epoch": 0.9656181419166057,
"grad_norm": 0.372391618489491,
"learning_rate": 5e-06,
"loss": 0.6406,
"step": 330
},
{
"epoch": 0.9948792977322605,
"grad_norm": 0.3496489692225867,
"learning_rate": 5e-06,
"loss": 0.6366,
"step": 340
},
{
"epoch": 0.9978054133138259,
"eval_loss": 0.6410567760467529,
"eval_runtime": 349.4713,
"eval_samples_per_second": 26.351,
"eval_steps_per_second": 0.412,
"step": 341
},
{
"epoch": 1.025237746891002,
"grad_norm": 0.3645378995154809,
"learning_rate": 5e-06,
"loss": 0.6529,
"step": 350
},
{
"epoch": 1.054498902706657,
"grad_norm": 0.3584256753081636,
"learning_rate": 5e-06,
"loss": 0.6072,
"step": 360
},
{
"epoch": 1.0837600585223117,
"grad_norm": 0.4013272730229709,
"learning_rate": 5e-06,
"loss": 0.6106,
"step": 370
},
{
"epoch": 1.1130212143379663,
"grad_norm": 0.3118543677090741,
"learning_rate": 5e-06,
"loss": 0.608,
"step": 380
},
{
"epoch": 1.142282370153621,
"grad_norm": 0.40549978724467745,
"learning_rate": 5e-06,
"loss": 0.6132,
"step": 390
},
{
"epoch": 1.1715435259692757,
"grad_norm": 0.3801789350020663,
"learning_rate": 5e-06,
"loss": 0.6147,
"step": 400
},
{
"epoch": 1.2008046817849305,
"grad_norm": 0.35753776804567533,
"learning_rate": 5e-06,
"loss": 0.6144,
"step": 410
},
{
"epoch": 1.2300658376005853,
"grad_norm": 0.35686093192236534,
"learning_rate": 5e-06,
"loss": 0.6026,
"step": 420
},
{
"epoch": 1.2593269934162399,
"grad_norm": 0.3542099455933607,
"learning_rate": 5e-06,
"loss": 0.6127,
"step": 430
},
{
"epoch": 1.2885881492318947,
"grad_norm": 0.3674357125440883,
"learning_rate": 5e-06,
"loss": 0.6053,
"step": 440
},
{
"epoch": 1.3178493050475493,
"grad_norm": 0.347417493921015,
"learning_rate": 5e-06,
"loss": 0.6091,
"step": 450
},
{
"epoch": 1.347110460863204,
"grad_norm": 0.3755065229307844,
"learning_rate": 5e-06,
"loss": 0.6066,
"step": 460
},
{
"epoch": 1.3763716166788589,
"grad_norm": 0.32563745811295164,
"learning_rate": 5e-06,
"loss": 0.604,
"step": 470
},
{
"epoch": 1.4056327724945135,
"grad_norm": 0.3775969105850567,
"learning_rate": 5e-06,
"loss": 0.6049,
"step": 480
},
{
"epoch": 1.4348939283101683,
"grad_norm": 0.3523952180274144,
"learning_rate": 5e-06,
"loss": 0.6113,
"step": 490
},
{
"epoch": 1.464155084125823,
"grad_norm": 0.34210469286447265,
"learning_rate": 5e-06,
"loss": 0.5996,
"step": 500
},
{
"epoch": 1.4934162399414777,
"grad_norm": 0.34388740461109346,
"learning_rate": 5e-06,
"loss": 0.6076,
"step": 510
},
{
"epoch": 1.5226773957571325,
"grad_norm": 0.38141624225595383,
"learning_rate": 5e-06,
"loss": 0.6099,
"step": 520
},
{
"epoch": 1.5519385515727873,
"grad_norm": 0.3500298323225119,
"learning_rate": 5e-06,
"loss": 0.6086,
"step": 530
},
{
"epoch": 1.5811997073884418,
"grad_norm": 0.3691197474525604,
"learning_rate": 5e-06,
"loss": 0.6151,
"step": 540
},
{
"epoch": 1.6104608632040964,
"grad_norm": 0.3534428783694921,
"learning_rate": 5e-06,
"loss": 0.6061,
"step": 550
},
{
"epoch": 1.6397220190197512,
"grad_norm": 0.3517980788142071,
"learning_rate": 5e-06,
"loss": 0.6078,
"step": 560
},
{
"epoch": 1.668983174835406,
"grad_norm": 0.3223935169033611,
"learning_rate": 5e-06,
"loss": 0.5984,
"step": 570
},
{
"epoch": 1.6982443306510606,
"grad_norm": 0.35022080709149067,
"learning_rate": 5e-06,
"loss": 0.6078,
"step": 580
},
{
"epoch": 1.7275054864667154,
"grad_norm": 0.4119325147732061,
"learning_rate": 5e-06,
"loss": 0.608,
"step": 590
},
{
"epoch": 1.7567666422823702,
"grad_norm": 0.34243774786649567,
"learning_rate": 5e-06,
"loss": 0.6049,
"step": 600
},
{
"epoch": 1.7860277980980248,
"grad_norm": 0.34892175870413145,
"learning_rate": 5e-06,
"loss": 0.6074,
"step": 610
},
{
"epoch": 1.8152889539136796,
"grad_norm": 0.3229538393877461,
"learning_rate": 5e-06,
"loss": 0.6077,
"step": 620
},
{
"epoch": 1.8445501097293344,
"grad_norm": 0.35234082559415686,
"learning_rate": 5e-06,
"loss": 0.6068,
"step": 630
},
{
"epoch": 1.873811265544989,
"grad_norm": 0.33202419310819015,
"learning_rate": 5e-06,
"loss": 0.6017,
"step": 640
},
{
"epoch": 1.9030724213606436,
"grad_norm": 0.36013370220716057,
"learning_rate": 5e-06,
"loss": 0.6073,
"step": 650
},
{
"epoch": 1.9323335771762986,
"grad_norm": 0.3421181986649472,
"learning_rate": 5e-06,
"loss": 0.6078,
"step": 660
},
{
"epoch": 1.9615947329919532,
"grad_norm": 0.3374053336625735,
"learning_rate": 5e-06,
"loss": 0.5967,
"step": 670
},
{
"epoch": 1.9908558888076078,
"grad_norm": 0.35619638919255014,
"learning_rate": 5e-06,
"loss": 0.6112,
"step": 680
},
{
"epoch": 1.9967081199707388,
"eval_loss": 0.6334938406944275,
"eval_runtime": 348.5476,
"eval_samples_per_second": 26.421,
"eval_steps_per_second": 0.413,
"step": 682
},
{
"epoch": 2.0212143379663496,
"grad_norm": 0.37926282562581526,
"learning_rate": 5e-06,
"loss": 0.622,
"step": 690
},
{
"epoch": 2.050475493782004,
"grad_norm": 0.3929793627802926,
"learning_rate": 5e-06,
"loss": 0.57,
"step": 700
},
{
"epoch": 2.0797366495976592,
"grad_norm": 0.3706776754114613,
"learning_rate": 5e-06,
"loss": 0.5719,
"step": 710
},
{
"epoch": 2.108997805413314,
"grad_norm": 0.37089264459896026,
"learning_rate": 5e-06,
"loss": 0.573,
"step": 720
},
{
"epoch": 2.1382589612289684,
"grad_norm": 0.3594270099758778,
"learning_rate": 5e-06,
"loss": 0.5748,
"step": 730
},
{
"epoch": 2.1675201170446234,
"grad_norm": 0.3461294399053881,
"learning_rate": 5e-06,
"loss": 0.58,
"step": 740
},
{
"epoch": 2.196781272860278,
"grad_norm": 0.3181893986886622,
"learning_rate": 5e-06,
"loss": 0.5787,
"step": 750
},
{
"epoch": 2.2260424286759326,
"grad_norm": 0.36416345109556053,
"learning_rate": 5e-06,
"loss": 0.5762,
"step": 760
},
{
"epoch": 2.255303584491587,
"grad_norm": 0.3419575420494747,
"learning_rate": 5e-06,
"loss": 0.5749,
"step": 770
},
{
"epoch": 2.284564740307242,
"grad_norm": 0.3372677940038377,
"learning_rate": 5e-06,
"loss": 0.574,
"step": 780
},
{
"epoch": 2.313825896122897,
"grad_norm": 0.3300673182470663,
"learning_rate": 5e-06,
"loss": 0.5723,
"step": 790
},
{
"epoch": 2.3430870519385514,
"grad_norm": 0.3650778084840146,
"learning_rate": 5e-06,
"loss": 0.58,
"step": 800
},
{
"epoch": 2.3723482077542064,
"grad_norm": 0.32963356969240615,
"learning_rate": 5e-06,
"loss": 0.5753,
"step": 810
},
{
"epoch": 2.401609363569861,
"grad_norm": 0.3364991693592673,
"learning_rate": 5e-06,
"loss": 0.5816,
"step": 820
},
{
"epoch": 2.4308705193855156,
"grad_norm": 0.31936137004871046,
"learning_rate": 5e-06,
"loss": 0.5737,
"step": 830
},
{
"epoch": 2.4601316752011706,
"grad_norm": 0.358418877386848,
"learning_rate": 5e-06,
"loss": 0.5742,
"step": 840
},
{
"epoch": 2.489392831016825,
"grad_norm": 0.3403922615121142,
"learning_rate": 5e-06,
"loss": 0.5764,
"step": 850
},
{
"epoch": 2.5186539868324798,
"grad_norm": 0.3548961641103604,
"learning_rate": 5e-06,
"loss": 0.5695,
"step": 860
},
{
"epoch": 2.547915142648135,
"grad_norm": 0.40628196038677905,
"learning_rate": 5e-06,
"loss": 0.5719,
"step": 870
},
{
"epoch": 2.5771762984637894,
"grad_norm": 0.3378466077976212,
"learning_rate": 5e-06,
"loss": 0.5761,
"step": 880
},
{
"epoch": 2.606437454279444,
"grad_norm": 0.35528813258823905,
"learning_rate": 5e-06,
"loss": 0.5755,
"step": 890
},
{
"epoch": 2.6356986100950985,
"grad_norm": 0.35270724056367453,
"learning_rate": 5e-06,
"loss": 0.58,
"step": 900
},
{
"epoch": 2.6649597659107536,
"grad_norm": 0.34198851107656175,
"learning_rate": 5e-06,
"loss": 0.5801,
"step": 910
},
{
"epoch": 2.694220921726408,
"grad_norm": 0.3659554553870692,
"learning_rate": 5e-06,
"loss": 0.5745,
"step": 920
},
{
"epoch": 2.723482077542063,
"grad_norm": 0.34754788190335867,
"learning_rate": 5e-06,
"loss": 0.5781,
"step": 930
},
{
"epoch": 2.7527432333577178,
"grad_norm": 0.350021186722354,
"learning_rate": 5e-06,
"loss": 0.5771,
"step": 940
},
{
"epoch": 2.7820043891733723,
"grad_norm": 0.3261901482406052,
"learning_rate": 5e-06,
"loss": 0.5758,
"step": 950
},
{
"epoch": 2.811265544989027,
"grad_norm": 0.3760979083321526,
"learning_rate": 5e-06,
"loss": 0.5796,
"step": 960
},
{
"epoch": 2.840526700804682,
"grad_norm": 0.3382336346055005,
"learning_rate": 5e-06,
"loss": 0.5753,
"step": 970
},
{
"epoch": 2.8697878566203365,
"grad_norm": 0.3727676460972042,
"learning_rate": 5e-06,
"loss": 0.5752,
"step": 980
},
{
"epoch": 2.899049012435991,
"grad_norm": 0.32827466587432536,
"learning_rate": 5e-06,
"loss": 0.5681,
"step": 990
},
{
"epoch": 2.928310168251646,
"grad_norm": 0.3221607307507683,
"learning_rate": 5e-06,
"loss": 0.5759,
"step": 1000
},
{
"epoch": 2.9575713240673007,
"grad_norm": 0.33746812770770435,
"learning_rate": 5e-06,
"loss": 0.5677,
"step": 1010
},
{
"epoch": 2.9868324798829553,
"grad_norm": 0.33814683138644874,
"learning_rate": 5e-06,
"loss": 0.5794,
"step": 1020
},
{
"epoch": 2.9956108266276518,
"eval_loss": 0.6330721378326416,
"eval_runtime": 348.0778,
"eval_samples_per_second": 26.457,
"eval_steps_per_second": 0.414,
"step": 1023
},
{
"epoch": 2.9956108266276518,
"step": 1023,
"total_flos": 2144987064041472.0,
"train_loss": 0.6156940991577864,
"train_runtime": 55371.9067,
"train_samples_per_second": 9.479,
"train_steps_per_second": 0.018
}
],
"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
}