sedrickkeh's picture
End of training
f453a95 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": 9.167642141908177,
"learning_rate": 9.615384615384617e-07,
"loss": 0.8861,
"step": 10
},
{
"epoch": 0.058522311631309436,
"grad_norm": 2.1214939165936095,
"learning_rate": 1.9230769230769234e-06,
"loss": 0.8326,
"step": 20
},
{
"epoch": 0.08778346744696415,
"grad_norm": 1.4860143285284373,
"learning_rate": 2.8846153846153845e-06,
"loss": 0.7582,
"step": 30
},
{
"epoch": 0.11704462326261887,
"grad_norm": 1.2064803930790473,
"learning_rate": 3.846153846153847e-06,
"loss": 0.7319,
"step": 40
},
{
"epoch": 0.14630577907827358,
"grad_norm": 1.2325754199204868,
"learning_rate": 4.807692307692308e-06,
"loss": 0.715,
"step": 50
},
{
"epoch": 0.1755669348939283,
"grad_norm": 1.1527080287428282,
"learning_rate": 4.999162611434756e-06,
"loss": 0.6993,
"step": 60
},
{
"epoch": 0.20482809070958302,
"grad_norm": 1.0684614331267772,
"learning_rate": 4.995761681803464e-06,
"loss": 0.6873,
"step": 70
},
{
"epoch": 0.23408924652523774,
"grad_norm": 0.7959876001332168,
"learning_rate": 4.989748431344101e-06,
"loss": 0.6767,
"step": 80
},
{
"epoch": 0.26335040234089246,
"grad_norm": 0.5559455782011308,
"learning_rate": 4.981129154142974e-06,
"loss": 0.669,
"step": 90
},
{
"epoch": 0.29261155815654716,
"grad_norm": 0.5142664344625067,
"learning_rate": 4.969912872021965e-06,
"loss": 0.6592,
"step": 100
},
{
"epoch": 0.3218727139722019,
"grad_norm": 0.4068010226081863,
"learning_rate": 4.956111325095368e-06,
"loss": 0.667,
"step": 110
},
{
"epoch": 0.3511338697878566,
"grad_norm": 0.3655542210299479,
"learning_rate": 4.9397389594814805e-06,
"loss": 0.6714,
"step": 120
},
{
"epoch": 0.38039502560351135,
"grad_norm": 0.3768366422263747,
"learning_rate": 4.920812912181809e-06,
"loss": 0.6549,
"step": 130
},
{
"epoch": 0.40965618141916604,
"grad_norm": 0.37566474114009984,
"learning_rate": 4.899352993143717e-06,
"loss": 0.6553,
"step": 140
},
{
"epoch": 0.4389173372348208,
"grad_norm": 0.3612379084735751,
"learning_rate": 4.875381664525304e-06,
"loss": 0.6504,
"step": 150
},
{
"epoch": 0.4681784930504755,
"grad_norm": 0.38526971126907095,
"learning_rate": 4.848924017184196e-06,
"loss": 0.6521,
"step": 160
},
{
"epoch": 0.49743964886613024,
"grad_norm": 0.360842741539703,
"learning_rate": 4.820007744414875e-06,
"loss": 0.6571,
"step": 170
},
{
"epoch": 0.5267008046817849,
"grad_norm": 0.34250107685671755,
"learning_rate": 4.788663112962028e-06,
"loss": 0.6566,
"step": 180
},
{
"epoch": 0.5559619604974396,
"grad_norm": 0.35953985379031617,
"learning_rate": 4.7549229313402636e-06,
"loss": 0.645,
"step": 190
},
{
"epoch": 0.5852231163130943,
"grad_norm": 0.3405632867922337,
"learning_rate": 4.71882251549334e-06,
"loss": 0.6461,
"step": 200
},
{
"epoch": 0.6144842721287491,
"grad_norm": 0.3372605973780747,
"learning_rate": 4.680399651828873e-06,
"loss": 0.6526,
"step": 210
},
{
"epoch": 0.6437454279444038,
"grad_norm": 0.3509569638126989,
"learning_rate": 4.639694557667194e-06,
"loss": 0.6404,
"step": 220
},
{
"epoch": 0.6730065837600585,
"grad_norm": 0.34631382508847197,
"learning_rate": 4.596749839145763e-06,
"loss": 0.6403,
"step": 230
},
{
"epoch": 0.7022677395757132,
"grad_norm": 0.32023219081524656,
"learning_rate": 4.551610446623209e-06,
"loss": 0.6472,
"step": 240
},
{
"epoch": 0.731528895391368,
"grad_norm": 0.3457771028484765,
"learning_rate": 4.504323627629662e-06,
"loss": 0.6455,
"step": 250
},
{
"epoch": 0.7607900512070227,
"grad_norm": 0.34127281622446637,
"learning_rate": 4.454938877412623e-06,
"loss": 0.6481,
"step": 260
},
{
"epoch": 0.7900512070226774,
"grad_norm": 0.3765210731050284,
"learning_rate": 4.4035078871301565e-06,
"loss": 0.6503,
"step": 270
},
{
"epoch": 0.8193123628383321,
"grad_norm": 0.3469100901523423,
"learning_rate": 4.350084489745601e-06,
"loss": 0.6491,
"step": 280
},
{
"epoch": 0.8485735186539868,
"grad_norm": 0.3308234776132991,
"learning_rate": 4.294724603680461e-06,
"loss": 0.6413,
"step": 290
},
{
"epoch": 0.8778346744696416,
"grad_norm": 0.3435672110063973,
"learning_rate": 4.23748617428443e-06,
"loss": 0.6331,
"step": 300
},
{
"epoch": 0.9070958302852963,
"grad_norm": 0.35464153574051116,
"learning_rate": 4.178429113183835e-06,
"loss": 0.6463,
"step": 310
},
{
"epoch": 0.936356986100951,
"grad_norm": 0.3250073753317316,
"learning_rate": 4.117615235571962e-06,
"loss": 0.6381,
"step": 320
},
{
"epoch": 0.9656181419166057,
"grad_norm": 0.37939565000110126,
"learning_rate": 4.05510819550692e-06,
"loss": 0.6414,
"step": 330
},
{
"epoch": 0.9948792977322605,
"grad_norm": 0.3580142785620277,
"learning_rate": 3.990973419284758e-06,
"loss": 0.6374,
"step": 340
},
{
"epoch": 0.9978054133138259,
"eval_loss": 0.641979992389679,
"eval_runtime": 346.288,
"eval_samples_per_second": 26.593,
"eval_steps_per_second": 0.416,
"step": 341
},
{
"epoch": 1.025237746891002,
"grad_norm": 0.3644926759716025,
"learning_rate": 3.925278036957578e-06,
"loss": 0.6553,
"step": 350
},
{
"epoch": 1.054498902706657,
"grad_norm": 0.38301527457304524,
"learning_rate": 3.858090812068314e-06,
"loss": 0.6099,
"step": 360
},
{
"epoch": 1.0837600585223117,
"grad_norm": 0.3546790345224804,
"learning_rate": 3.789482069675738e-06,
"loss": 0.6133,
"step": 370
},
{
"epoch": 1.1130212143379663,
"grad_norm": 0.3146730032039869,
"learning_rate": 3.719523622745026e-06,
"loss": 0.6105,
"step": 380
},
{
"epoch": 1.142282370153621,
"grad_norm": 0.38139306525912997,
"learning_rate": 3.6482886969809204e-06,
"loss": 0.6158,
"step": 390
},
{
"epoch": 1.1715435259692757,
"grad_norm": 0.3468751817352307,
"learning_rate": 3.575851854182175e-06,
"loss": 0.6172,
"step": 400
},
{
"epoch": 1.2008046817849305,
"grad_norm": 0.36354294198162557,
"learning_rate": 3.502288914197513e-06,
"loss": 0.6169,
"step": 410
},
{
"epoch": 1.2300658376005853,
"grad_norm": 0.32988905808120284,
"learning_rate": 3.4276768755647748e-06,
"loss": 0.6049,
"step": 420
},
{
"epoch": 1.2593269934162399,
"grad_norm": 0.3173866656888502,
"learning_rate": 3.352093834916329e-06,
"loss": 0.615,
"step": 430
},
{
"epoch": 1.2885881492318947,
"grad_norm": 0.39499650219748045,
"learning_rate": 3.275618905235109e-06,
"loss": 0.6075,
"step": 440
},
{
"epoch": 1.3178493050475493,
"grad_norm": 0.34196576749131313,
"learning_rate": 3.1983321330468235e-06,
"loss": 0.6114,
"step": 450
},
{
"epoch": 1.347110460863204,
"grad_norm": 0.323009005146513,
"learning_rate": 3.1203144146350396e-06,
"loss": 0.6089,
"step": 460
},
{
"epoch": 1.3763716166788589,
"grad_norm": 0.31817976264323716,
"learning_rate": 3.0416474113668055e-06,
"loss": 0.6062,
"step": 470
},
{
"epoch": 1.4056327724945135,
"grad_norm": 0.3262356503952236,
"learning_rate": 2.9624134642174706e-06,
"loss": 0.6068,
"step": 480
},
{
"epoch": 1.4348939283101683,
"grad_norm": 0.3457424267563388,
"learning_rate": 2.8826955075841527e-06,
"loss": 0.6131,
"step": 490
},
{
"epoch": 1.464155084125823,
"grad_norm": 0.3307935925728294,
"learning_rate": 2.802576982478068e-06,
"loss": 0.6016,
"step": 500
},
{
"epoch": 1.4934162399414777,
"grad_norm": 0.3311781576210982,
"learning_rate": 2.7221417491865885e-06,
"loss": 0.6096,
"step": 510
},
{
"epoch": 1.5226773957571325,
"grad_norm": 0.34216041994362073,
"learning_rate": 2.6414739994964416e-06,
"loss": 0.6119,
"step": 520
},
{
"epoch": 1.5519385515727873,
"grad_norm": 0.33367655978352256,
"learning_rate": 2.56065816856993e-06,
"loss": 0.6105,
"step": 530
},
{
"epoch": 1.5811997073884418,
"grad_norm": 0.33131489612390447,
"learning_rate": 2.4797788465664047e-06,
"loss": 0.6171,
"step": 540
},
{
"epoch": 1.6104608632040964,
"grad_norm": 0.3386564765273768,
"learning_rate": 2.398920690101506e-06,
"loss": 0.6079,
"step": 550
},
{
"epoch": 1.6397220190197512,
"grad_norm": 0.3422865962147514,
"learning_rate": 2.318168333636849e-06,
"loss": 0.6097,
"step": 560
},
{
"epoch": 1.668983174835406,
"grad_norm": 0.33015186307929917,
"learning_rate": 2.2376063008928832e-06,
"loss": 0.6002,
"step": 570
},
{
"epoch": 1.6982443306510606,
"grad_norm": 0.330356253324613,
"learning_rate": 2.1573189163776857e-06,
"loss": 0.6098,
"step": 580
},
{
"epoch": 1.7275054864667154,
"grad_norm": 0.32143316840124014,
"learning_rate": 2.0773902171242485e-06,
"loss": 0.6095,
"step": 590
},
{
"epoch": 1.7567666422823702,
"grad_norm": 0.33148914076976616,
"learning_rate": 1.997903864728688e-06,
"loss": 0.6066,
"step": 600
},
{
"epoch": 1.7860277980980248,
"grad_norm": 0.3480703652832865,
"learning_rate": 1.9189430577814063e-06,
"loss": 0.609,
"step": 610
},
{
"epoch": 1.8152889539136796,
"grad_norm": 0.31334239803694514,
"learning_rate": 1.8405904447828982e-06,
"loss": 0.6093,
"step": 620
},
{
"epoch": 1.8445501097293344,
"grad_norm": 0.31827545220106257,
"learning_rate": 1.762928037635325e-06,
"loss": 0.6086,
"step": 630
},
{
"epoch": 1.873811265544989,
"grad_norm": 0.3155798256896876,
"learning_rate": 1.6860371258004293e-06,
"loss": 0.6033,
"step": 640
},
{
"epoch": 1.9030724213606436,
"grad_norm": 0.3390645406779253,
"learning_rate": 1.6099981912136216e-06,
"loss": 0.609,
"step": 650
},
{
"epoch": 1.9323335771762986,
"grad_norm": 0.30392526393365127,
"learning_rate": 1.5348908240433125e-06,
"loss": 0.6096,
"step": 660
},
{
"epoch": 1.9615947329919532,
"grad_norm": 0.3261239230385701,
"learning_rate": 1.4607936393836542e-06,
"loss": 0.5984,
"step": 670
},
{
"epoch": 1.9908558888076078,
"grad_norm": 0.3115182329109519,
"learning_rate": 1.387784194967904e-06,
"loss": 0.6129,
"step": 680
},
{
"epoch": 1.9967081199707388,
"eval_loss": 0.6351735591888428,
"eval_runtime": 347.6382,
"eval_samples_per_second": 26.49,
"eval_steps_per_second": 0.414,
"step": 682
},
{
"epoch": 2.0212143379663496,
"grad_norm": 0.32036924437042374,
"learning_rate": 1.3159389099885194e-06,
"loss": 0.633,
"step": 690
},
{
"epoch": 2.050475493782004,
"grad_norm": 0.324600013295968,
"learning_rate": 1.2453329851089787e-06,
"loss": 0.5844,
"step": 700
},
{
"epoch": 2.0797366495976592,
"grad_norm": 0.3107112923712249,
"learning_rate": 1.1760403237510255e-06,
"loss": 0.5866,
"step": 710
},
{
"epoch": 2.108997805413314,
"grad_norm": 0.3270752834403749,
"learning_rate": 1.108133454739757e-06,
"loss": 0.5876,
"step": 720
},
{
"epoch": 2.1382589612289684,
"grad_norm": 0.31301220771612853,
"learning_rate": 1.0416834563874873e-06,
"loss": 0.5892,
"step": 730
},
{
"epoch": 2.1675201170446234,
"grad_norm": 0.33747585616696024,
"learning_rate": 9.767598820958859e-07,
"loss": 0.5939,
"step": 740
},
{
"epoch": 2.196781272860278,
"grad_norm": 0.3224610267552276,
"learning_rate": 9.134306875542273e-07,
"loss": 0.5927,
"step": 750
},
{
"epoch": 2.2260424286759326,
"grad_norm": 0.3274463198805806,
"learning_rate": 8.517621596099837e-07,
"loss": 0.5898,
"step": 760
},
{
"epoch": 2.255303584491587,
"grad_norm": 0.3096086967508753,
"learning_rate": 7.918188468861892e-07,
"loss": 0.588,
"step": 770
},
{
"epoch": 2.284564740307242,
"grad_norm": 0.3134498246737174,
"learning_rate": 7.336634922182184e-07,
"loss": 0.5875,
"step": 780
},
{
"epoch": 2.313825896122897,
"grad_norm": 0.3003387704727999,
"learning_rate": 6.773569669806765e-07,
"loss": 0.5854,
"step": 790
},
{
"epoch": 2.3430870519385514,
"grad_norm": 0.31046829316377283,
"learning_rate": 6.22958207373166e-07,
"loss": 0.5931,
"step": 800
},
{
"epoch": 2.3723482077542064,
"grad_norm": 0.3058504002323456,
"learning_rate": 5.705241527315924e-07,
"loss": 0.5881,
"step": 810
},
{
"epoch": 2.401609363569861,
"grad_norm": 0.3148001544813671,
"learning_rate": 5.201096859296056e-07,
"loss": 0.5943,
"step": 820
},
{
"epoch": 2.4308705193855156,
"grad_norm": 0.29314718822976743,
"learning_rate": 4.717675759325363e-07,
"loss": 0.5866,
"step": 830
},
{
"epoch": 2.4601316752011706,
"grad_norm": 0.30348786991224425,
"learning_rate": 4.255484225639739e-07,
"loss": 0.5869,
"step": 840
},
{
"epoch": 2.489392831016825,
"grad_norm": 0.31048470531749484,
"learning_rate": 3.8150060354278373e-07,
"loss": 0.5886,
"step": 850
},
{
"epoch": 2.5186539868324798,
"grad_norm": 0.31385687702363135,
"learning_rate": 3.396702238460167e-07,
"loss": 0.5818,
"step": 860
},
{
"epoch": 2.547915142648135,
"grad_norm": 0.3077001922479862,
"learning_rate": 3.001010674506932e-07,
"loss": 0.5842,
"step": 870
},
{
"epoch": 2.5771762984637894,
"grad_norm": 0.2926705998755526,
"learning_rate": 2.628345515049954e-07,
"loss": 0.5882,
"step": 880
},
{
"epoch": 2.606437454279444,
"grad_norm": 0.2995165809811577,
"learning_rate": 2.2790968297681792e-07,
"loss": 0.5875,
"step": 890
},
{
"epoch": 2.6356986100950985,
"grad_norm": 0.30466800666878485,
"learning_rate": 1.9536301782506373e-07,
"loss": 0.5921,
"step": 900
},
{
"epoch": 2.6649597659107536,
"grad_norm": 0.3053293282467882,
"learning_rate": 1.652286227364211e-07,
"loss": 0.5924,
"step": 910
},
{
"epoch": 2.694220921726408,
"grad_norm": 0.3137494141095099,
"learning_rate": 1.375380394676623e-07,
"loss": 0.5864,
"step": 920
},
{
"epoch": 2.723482077542063,
"grad_norm": 0.31676507013297805,
"learning_rate": 1.1232025183080141e-07,
"loss": 0.5898,
"step": 930
},
{
"epoch": 2.7527432333577178,
"grad_norm": 0.30895835779314046,
"learning_rate": 8.960165535565168e-08,
"loss": 0.5888,
"step": 940
},
{
"epoch": 2.7820043891733723,
"grad_norm": 0.2949131063367962,
"learning_rate": 6.940602966155341e-08,
"loss": 0.5875,
"step": 950
},
{
"epoch": 2.811265544989027,
"grad_norm": 0.3073000958086021,
"learning_rate": 5.175451356717737e-08,
"loss": 0.5914,
"step": 960
},
{
"epoch": 2.840526700804682,
"grad_norm": 0.2952136468986014,
"learning_rate": 3.6665582964462986e-08,
"loss": 0.5871,
"step": 970
},
{
"epoch": 2.8697878566203365,
"grad_norm": 0.30029138721190923,
"learning_rate": 2.41550314798511e-08,
"loss": 0.5869,
"step": 980
},
{
"epoch": 2.899049012435991,
"grad_norm": 0.29808129794254773,
"learning_rate": 1.4235953943048942e-08,
"loss": 0.5796,
"step": 990
},
{
"epoch": 2.928310168251646,
"grad_norm": 0.3005338199882765,
"learning_rate": 6.918732680635809e-09,
"loss": 0.5876,
"step": 1000
},
{
"epoch": 2.9575713240673007,
"grad_norm": 0.3005090950716719,
"learning_rate": 2.21102664885231e-09,
"loss": 0.5789,
"step": 1010
},
{
"epoch": 2.9868324798829553,
"grad_norm": 0.30127576183009913,
"learning_rate": 1.177634169488573e-10,
"loss": 0.591,
"step": 1020
},
{
"epoch": 2.9956108266276518,
"eval_loss": 0.635776937007904,
"eval_runtime": 344.6936,
"eval_samples_per_second": 26.716,
"eval_steps_per_second": 0.418,
"step": 1023
},
{
"epoch": 2.9956108266276518,
"step": 1023,
"total_flos": 2144987064041472.0,
"train_loss": 0.6243289487685859,
"train_runtime": 55081.5376,
"train_samples_per_second": 9.529,
"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
}