chansung's picture
Model save
6cbf5f7 verified
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 0.9991416309012876,
"eval_steps": 500,
"global_step": 582,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0017167381974248926,
"grad_norm": 3.986985683441162,
"learning_rate": 3.3898305084745763e-06,
"loss": 1.497,
"step": 1
},
{
"epoch": 0.008583690987124463,
"grad_norm": 4.083123683929443,
"learning_rate": 1.694915254237288e-05,
"loss": 1.5029,
"step": 5
},
{
"epoch": 0.017167381974248927,
"grad_norm": 2.801696300506592,
"learning_rate": 3.389830508474576e-05,
"loss": 1.4007,
"step": 10
},
{
"epoch": 0.02575107296137339,
"grad_norm": 2.138871908187866,
"learning_rate": 5.0847457627118643e-05,
"loss": 1.2542,
"step": 15
},
{
"epoch": 0.034334763948497854,
"grad_norm": 1.4012372493743896,
"learning_rate": 6.779661016949152e-05,
"loss": 1.1505,
"step": 20
},
{
"epoch": 0.04291845493562232,
"grad_norm": 0.8826782703399658,
"learning_rate": 8.474576271186441e-05,
"loss": 1.0653,
"step": 25
},
{
"epoch": 0.05150214592274678,
"grad_norm": 0.7491444945335388,
"learning_rate": 0.00010169491525423729,
"loss": 1.0047,
"step": 30
},
{
"epoch": 0.060085836909871244,
"grad_norm": 0.691436767578125,
"learning_rate": 0.00011864406779661017,
"loss": 0.9515,
"step": 35
},
{
"epoch": 0.06866952789699571,
"grad_norm": 0.5920073390007019,
"learning_rate": 0.00013559322033898305,
"loss": 0.9006,
"step": 40
},
{
"epoch": 0.07725321888412018,
"grad_norm": 0.5658953785896301,
"learning_rate": 0.00015254237288135592,
"loss": 0.864,
"step": 45
},
{
"epoch": 0.08583690987124463,
"grad_norm": 0.4744928181171417,
"learning_rate": 0.00016949152542372882,
"loss": 0.8326,
"step": 50
},
{
"epoch": 0.0944206008583691,
"grad_norm": 1.3415422439575195,
"learning_rate": 0.0001864406779661017,
"loss": 0.8227,
"step": 55
},
{
"epoch": 0.10300429184549356,
"grad_norm": 0.6074433922767639,
"learning_rate": 0.0001999981958814743,
"loss": 0.8157,
"step": 60
},
{
"epoch": 0.11158798283261803,
"grad_norm": 0.5061601996421814,
"learning_rate": 0.00019993505856798243,
"loss": 0.806,
"step": 65
},
{
"epoch": 0.12017167381974249,
"grad_norm": 0.48751935362815857,
"learning_rate": 0.00019978178041451998,
"loss": 0.7938,
"step": 70
},
{
"epoch": 0.12875536480686695,
"grad_norm": 0.4422796666622162,
"learning_rate": 0.0001995384996770874,
"loss": 0.8038,
"step": 75
},
{
"epoch": 0.13733905579399142,
"grad_norm": 0.4579014182090759,
"learning_rate": 0.000199205435793491,
"loss": 0.7865,
"step": 80
},
{
"epoch": 0.1459227467811159,
"grad_norm": 0.43843501806259155,
"learning_rate": 0.00019878288918541114,
"loss": 0.7769,
"step": 85
},
{
"epoch": 0.15450643776824036,
"grad_norm": 0.4283618927001953,
"learning_rate": 0.00019827124098742394,
"loss": 0.7815,
"step": 90
},
{
"epoch": 0.1630901287553648,
"grad_norm": 0.4591907858848572,
"learning_rate": 0.0001976709527032198,
"loss": 0.7852,
"step": 95
},
{
"epoch": 0.17167381974248927,
"grad_norm": 0.3763423264026642,
"learning_rate": 0.00019698256578933005,
"loss": 0.7712,
"step": 100
},
{
"epoch": 0.18025751072961374,
"grad_norm": 0.46547284722328186,
"learning_rate": 0.00019620670116673604,
"loss": 0.759,
"step": 105
},
{
"epoch": 0.1888412017167382,
"grad_norm": 0.4377448260784149,
"learning_rate": 0.0001953440586608022,
"loss": 0.7664,
"step": 110
},
{
"epoch": 0.19742489270386265,
"grad_norm": 0.4102751910686493,
"learning_rate": 0.00019439541637003772,
"loss": 0.7705,
"step": 115
},
{
"epoch": 0.20600858369098712,
"grad_norm": 0.5243790745735168,
"learning_rate": 0.00019336162996425597,
"loss": 0.7639,
"step": 120
},
{
"epoch": 0.2145922746781116,
"grad_norm": 0.4595816731452942,
"learning_rate": 0.00019224363191276559,
"loss": 0.7593,
"step": 125
},
{
"epoch": 0.22317596566523606,
"grad_norm": 0.42034828662872314,
"learning_rate": 0.00019104243064328838,
"loss": 0.7549,
"step": 130
},
{
"epoch": 0.2317596566523605,
"grad_norm": 0.443181574344635,
"learning_rate": 0.00018975910963236366,
"loss": 0.7561,
"step": 135
},
{
"epoch": 0.24034334763948498,
"grad_norm": 0.45033738017082214,
"learning_rate": 0.00018839482642805843,
"loss": 0.7543,
"step": 140
},
{
"epoch": 0.24892703862660945,
"grad_norm": 0.6711452603340149,
"learning_rate": 0.000186950811605866,
"loss": 0.7435,
"step": 145
},
{
"epoch": 0.2575107296137339,
"grad_norm": 0.45346158742904663,
"learning_rate": 0.00018542836765873396,
"loss": 0.7569,
"step": 150
},
{
"epoch": 0.26609442060085836,
"grad_norm": 0.39888471364974976,
"learning_rate": 0.00018382886782222313,
"loss": 0.7464,
"step": 155
},
{
"epoch": 0.27467811158798283,
"grad_norm": 0.59100741147995,
"learning_rate": 0.00018215375483585726,
"loss": 0.7452,
"step": 160
},
{
"epoch": 0.2832618025751073,
"grad_norm": 0.4381592273712158,
"learning_rate": 0.0001804045396417801,
"loss": 0.7324,
"step": 165
},
{
"epoch": 0.2918454935622318,
"grad_norm": 0.5130012631416321,
"learning_rate": 0.0001785828000218947,
"loss": 0.7354,
"step": 170
},
{
"epoch": 0.30042918454935624,
"grad_norm": 0.4690544307231903,
"learning_rate": 0.00017669017917471314,
"loss": 0.743,
"step": 175
},
{
"epoch": 0.3090128755364807,
"grad_norm": 0.5336308479309082,
"learning_rate": 0.00017472838423320108,
"loss": 0.7337,
"step": 180
},
{
"epoch": 0.31759656652360513,
"grad_norm": 0.46447548270225525,
"learning_rate": 0.0001726991847249539,
"loss": 0.7451,
"step": 185
},
{
"epoch": 0.3261802575107296,
"grad_norm": 0.39880847930908203,
"learning_rate": 0.000170604410976093,
"loss": 0.7303,
"step": 190
},
{
"epoch": 0.33476394849785407,
"grad_norm": 0.42278820276260376,
"learning_rate": 0.00016844595246032234,
"loss": 0.7467,
"step": 195
},
{
"epoch": 0.34334763948497854,
"grad_norm": 0.4219389855861664,
"learning_rate": 0.00016622575609463432,
"loss": 0.7454,
"step": 200
},
{
"epoch": 0.351931330472103,
"grad_norm": 0.49944549798965454,
"learning_rate": 0.00016394582448320163,
"loss": 0.7372,
"step": 205
},
{
"epoch": 0.3605150214592275,
"grad_norm": 0.45013898611068726,
"learning_rate": 0.00016160821411104057,
"loss": 0.7143,
"step": 210
},
{
"epoch": 0.36909871244635195,
"grad_norm": 0.3965418040752411,
"learning_rate": 0.00015921503348907288,
"loss": 0.7294,
"step": 215
},
{
"epoch": 0.3776824034334764,
"grad_norm": 0.3835819363594055,
"learning_rate": 0.00015676844125226178,
"loss": 0.7137,
"step": 220
},
{
"epoch": 0.38626609442060084,
"grad_norm": 0.40587151050567627,
"learning_rate": 0.00015427064421253542,
"loss": 0.7301,
"step": 225
},
{
"epoch": 0.3948497854077253,
"grad_norm": 0.40463292598724365,
"learning_rate": 0.00015172389536825554,
"loss": 0.73,
"step": 230
},
{
"epoch": 0.4034334763948498,
"grad_norm": 0.4850827157497406,
"learning_rate": 0.00014913049187202613,
"loss": 0.7143,
"step": 235
},
{
"epoch": 0.41201716738197425,
"grad_norm": 0.4655815064907074,
"learning_rate": 0.0001464927729586751,
"loss": 0.7228,
"step": 240
},
{
"epoch": 0.4206008583690987,
"grad_norm": 0.42954930663108826,
"learning_rate": 0.0001438131178352782,
"loss": 0.7424,
"step": 245
},
{
"epoch": 0.4291845493562232,
"grad_norm": 0.47542840242385864,
"learning_rate": 0.000141093943535128,
"loss": 0.7196,
"step": 250
},
{
"epoch": 0.43776824034334766,
"grad_norm": 0.4334765672683716,
"learning_rate": 0.00013833770273758396,
"loss": 0.7407,
"step": 255
},
{
"epoch": 0.44635193133047213,
"grad_norm": 0.43749311566352844,
"learning_rate": 0.00013554688155577012,
"loss": 0.7174,
"step": 260
},
{
"epoch": 0.45493562231759654,
"grad_norm": 0.4014078378677368,
"learning_rate": 0.0001327239972941153,
"loss": 0.7243,
"step": 265
},
{
"epoch": 0.463519313304721,
"grad_norm": 0.47700533270835876,
"learning_rate": 0.00012987159617775952,
"loss": 0.7318,
"step": 270
},
{
"epoch": 0.4721030042918455,
"grad_norm": 0.40848225355148315,
"learning_rate": 0.0001269922510558735,
"loss": 0.7213,
"step": 275
},
{
"epoch": 0.48068669527896996,
"grad_norm": 0.4185446798801422,
"learning_rate": 0.00012408855908096372,
"loss": 0.7251,
"step": 280
},
{
"epoch": 0.4892703862660944,
"grad_norm": 0.41637638211250305,
"learning_rate": 0.00012116313936625616,
"loss": 0.7304,
"step": 285
},
{
"epoch": 0.4978540772532189,
"grad_norm": 0.3961445391178131,
"learning_rate": 0.00011821863062327142,
"loss": 0.7176,
"step": 290
},
{
"epoch": 0.5064377682403434,
"grad_norm": 0.4589502811431885,
"learning_rate": 0.00011525768878172217,
"loss": 0.7091,
"step": 295
},
{
"epoch": 0.5150214592274678,
"grad_norm": 0.40595656633377075,
"learning_rate": 0.0001122829845938802,
"loss": 0.7343,
"step": 300
},
{
"epoch": 0.5236051502145923,
"grad_norm": 0.5253564119338989,
"learning_rate": 0.00010929720122557346,
"loss": 0.7162,
"step": 305
},
{
"epoch": 0.5321888412017167,
"grad_norm": 0.4770199954509735,
"learning_rate": 0.00010630303183598586,
"loss": 0.7117,
"step": 310
},
{
"epoch": 0.5407725321888412,
"grad_norm": 0.4303250312805176,
"learning_rate": 0.00010330317714844382,
"loss": 0.7075,
"step": 315
},
{
"epoch": 0.5493562231759657,
"grad_norm": 0.39325711131095886,
"learning_rate": 0.0001003003430143792,
"loss": 0.7138,
"step": 320
},
{
"epoch": 0.5579399141630901,
"grad_norm": 0.45550739765167236,
"learning_rate": 9.729723797266733e-05,
"loss": 0.72,
"step": 325
},
{
"epoch": 0.5665236051502146,
"grad_norm": 0.45296674966812134,
"learning_rate": 9.429657080654066e-05,
"loss": 0.7131,
"step": 330
},
{
"epoch": 0.575107296137339,
"grad_norm": 0.4620877504348755,
"learning_rate": 9.130104810028196e-05,
"loss": 0.7212,
"step": 335
},
{
"epoch": 0.5836909871244635,
"grad_norm": 0.3927091658115387,
"learning_rate": 8.831337179790135e-05,
"loss": 0.7156,
"step": 340
},
{
"epoch": 0.592274678111588,
"grad_norm": 0.3895888924598694,
"learning_rate": 8.533623676599832e-05,
"loss": 0.7166,
"step": 345
},
{
"epoch": 0.6008583690987125,
"grad_norm": 0.4301272928714752,
"learning_rate": 8.237232836300797e-05,
"loss": 0.7287,
"step": 350
},
{
"epoch": 0.6094420600858369,
"grad_norm": 0.4707149565219879,
"learning_rate": 7.94243200170232e-05,
"loss": 0.7021,
"step": 355
},
{
"epoch": 0.6180257510729614,
"grad_norm": 0.40633898973464966,
"learning_rate": 7.649487081437818e-05,
"loss": 0.6989,
"step": 360
},
{
"epoch": 0.6266094420600858,
"grad_norm": 0.4026099741458893,
"learning_rate": 7.358662310116808e-05,
"loss": 0.716,
"step": 365
},
{
"epoch": 0.6351931330472103,
"grad_norm": 0.397532194852829,
"learning_rate": 7.070220009986828e-05,
"loss": 0.6927,
"step": 370
},
{
"epoch": 0.6437768240343348,
"grad_norm": 0.4312935173511505,
"learning_rate": 6.784420354320295e-05,
"loss": 0.6962,
"step": 375
},
{
"epoch": 0.6523605150214592,
"grad_norm": 0.42235448956489563,
"learning_rate": 6.501521132739727e-05,
"loss": 0.7219,
"step": 380
},
{
"epoch": 0.6609442060085837,
"grad_norm": 0.4233233630657196,
"learning_rate": 6.221777518693023e-05,
"loss": 0.7139,
"step": 385
},
{
"epoch": 0.6695278969957081,
"grad_norm": 0.3970154821872711,
"learning_rate": 5.945441839288487e-05,
"loss": 0.7088,
"step": 390
},
{
"epoch": 0.6781115879828327,
"grad_norm": 0.39059823751449585,
"learning_rate": 5.672763347697252e-05,
"loss": 0.7122,
"step": 395
},
{
"epoch": 0.6866952789699571,
"grad_norm": 0.42738205194473267,
"learning_rate": 5.403987998328364e-05,
"loss": 0.6922,
"step": 400
},
{
"epoch": 0.6952789699570815,
"grad_norm": 0.45146945118904114,
"learning_rate": 5.139358224979368e-05,
"loss": 0.7019,
"step": 405
},
{
"epoch": 0.703862660944206,
"grad_norm": 0.4161079227924347,
"learning_rate": 4.8791127221624e-05,
"loss": 0.7089,
"step": 410
},
{
"epoch": 0.7124463519313304,
"grad_norm": 0.45124712586402893,
"learning_rate": 4.62348622980316e-05,
"loss": 0.7066,
"step": 415
},
{
"epoch": 0.721030042918455,
"grad_norm": 0.45476675033569336,
"learning_rate": 4.372709321506848e-05,
"loss": 0.7043,
"step": 420
},
{
"epoch": 0.7296137339055794,
"grad_norm": 0.424629271030426,
"learning_rate": 4.12700819658212e-05,
"loss": 0.7059,
"step": 425
},
{
"epoch": 0.7381974248927039,
"grad_norm": 0.4276103675365448,
"learning_rate": 3.8866044760106446e-05,
"loss": 0.7065,
"step": 430
},
{
"epoch": 0.7467811158798283,
"grad_norm": 0.43286246061325073,
"learning_rate": 3.651715002546251e-05,
"loss": 0.6944,
"step": 435
},
{
"epoch": 0.7553648068669528,
"grad_norm": 0.4260043799877167,
"learning_rate": 3.422551645124048e-05,
"loss": 0.7009,
"step": 440
},
{
"epoch": 0.7639484978540773,
"grad_norm": 0.39454787969589233,
"learning_rate": 3.199321107755863e-05,
"loss": 0.6939,
"step": 445
},
{
"epoch": 0.7725321888412017,
"grad_norm": 0.45329296588897705,
"learning_rate": 2.9822247430844375e-05,
"loss": 0.7146,
"step": 450
},
{
"epoch": 0.7811158798283262,
"grad_norm": 0.3966417610645294,
"learning_rate": 2.7714583707645247e-05,
"loss": 0.6974,
"step": 455
},
{
"epoch": 0.7896995708154506,
"grad_norm": 0.44266363978385925,
"learning_rate": 2.5672121008346895e-05,
"loss": 0.7029,
"step": 460
},
{
"epoch": 0.7982832618025751,
"grad_norm": 0.4370868504047394,
"learning_rate": 2.369670162239166e-05,
"loss": 0.6993,
"step": 465
},
{
"epoch": 0.8068669527896996,
"grad_norm": 0.413068562746048,
"learning_rate": 2.1790107366544366e-05,
"loss": 0.7017,
"step": 470
},
{
"epoch": 0.8154506437768241,
"grad_norm": 0.40551063418388367,
"learning_rate": 1.9954057977703776e-05,
"loss": 0.7055,
"step": 475
},
{
"epoch": 0.8240343347639485,
"grad_norm": 0.42292067408561707,
"learning_rate": 1.819020956171017e-05,
"loss": 0.6981,
"step": 480
},
{
"epoch": 0.8326180257510729,
"grad_norm": 0.41726067662239075,
"learning_rate": 1.6500153099547344e-05,
"loss": 0.6953,
"step": 485
},
{
"epoch": 0.8412017167381974,
"grad_norm": 0.4085144102573395,
"learning_rate": 1.4885413012287098e-05,
"loss": 0.7045,
"step": 490
},
{
"epoch": 0.8497854077253219,
"grad_norm": 0.40902742743492126,
"learning_rate": 1.3347445786070456e-05,
"loss": 0.7125,
"step": 495
},
{
"epoch": 0.8583690987124464,
"grad_norm": 0.408077597618103,
"learning_rate": 1.1887638658365452e-05,
"loss": 0.6806,
"step": 500
},
{
"epoch": 0.8669527896995708,
"grad_norm": 0.4097277820110321,
"learning_rate": 1.0507308366687141e-05,
"loss": 0.7041,
"step": 505
},
{
"epoch": 0.8755364806866953,
"grad_norm": 0.406400591135025,
"learning_rate": 9.207699960908012e-06,
"loss": 0.7052,
"step": 510
},
{
"epoch": 0.8841201716738197,
"grad_norm": 0.3999215364456177,
"learning_rate": 7.989985680230305e-06,
"loss": 0.7097,
"step": 515
},
{
"epoch": 0.8927038626609443,
"grad_norm": 0.4089025557041168,
"learning_rate": 6.855263895833064e-06,
"loss": 0.708,
"step": 520
},
{
"epoch": 0.9012875536480687,
"grad_norm": 0.398349791765213,
"learning_rate": 5.804558120147674e-06,
"loss": 0.7117,
"step": 525
},
{
"epoch": 0.9098712446351931,
"grad_norm": 0.40962719917297363,
"learning_rate": 4.838816083655695e-06,
"loss": 0.7083,
"step": 530
},
{
"epoch": 0.9184549356223176,
"grad_norm": 0.4219372868537903,
"learning_rate": 3.95890888004149e-06,
"loss": 0.7023,
"step": 535
},
{
"epoch": 0.927038626609442,
"grad_norm": 0.4286939799785614,
"learning_rate": 3.165630180470802e-06,
"loss": 0.7072,
"step": 540
},
{
"epoch": 0.9356223175965666,
"grad_norm": 0.4043157696723938,
"learning_rate": 2.459695517703919e-06,
"loss": 0.7078,
"step": 545
},
{
"epoch": 0.944206008583691,
"grad_norm": 0.3982146978378296,
"learning_rate": 1.8417416406894094e-06,
"loss": 0.6925,
"step": 550
},
{
"epoch": 0.9527896995708155,
"grad_norm": 0.387021005153656,
"learning_rate": 1.3123259402202048e-06,
"loss": 0.6931,
"step": 555
},
{
"epoch": 0.9613733905579399,
"grad_norm": 0.39866942167282104,
"learning_rate": 8.71925946170371e-07,
"loss": 0.7034,
"step": 560
},
{
"epoch": 0.9699570815450643,
"grad_norm": 0.4020968973636627,
"learning_rate": 5.209388967659768e-07,
"loss": 0.7247,
"step": 565
},
{
"epoch": 0.9785407725321889,
"grad_norm": 0.4172390103340149,
"learning_rate": 2.596813802785092e-07,
"loss": 0.7069,
"step": 570
},
{
"epoch": 0.9871244635193133,
"grad_norm": 0.4030327796936035,
"learning_rate": 8.838904946407711e-08,
"loss": 0.7,
"step": 575
},
{
"epoch": 0.9957081545064378,
"grad_norm": 0.4182168245315552,
"learning_rate": 7.216409006038927e-09,
"loss": 0.7218,
"step": 580
},
{
"epoch": 0.9991416309012876,
"eval_loss": 1.711365818977356,
"eval_runtime": 0.9327,
"eval_samples_per_second": 6.433,
"eval_steps_per_second": 1.072,
"step": 582
},
{
"epoch": 0.9991416309012876,
"step": 582,
"total_flos": 8.143946755646423e+17,
"train_loss": 0.7577278978636175,
"train_runtime": 4155.2284,
"train_samples_per_second": 4.484,
"train_steps_per_second": 0.14
}
],
"logging_steps": 5,
"max_steps": 582,
"num_input_tokens_seen": 0,
"num_train_epochs": 1,
"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": 8.143946755646423e+17,
"train_batch_size": 2,
"trial_name": null,
"trial_params": null
}