zephyr-7b-gemma-sft-10p-2048 / trainer_state.json
Qin Liu
Model save
19987b2 verified
{
"best_metric": null,
"best_model_checkpoint": null,
"epoch": 2.997711670480549,
"eval_steps": 500,
"global_step": 1965,
"is_hyper_param_search": false,
"is_local_process_zero": true,
"is_world_process_zero": true,
"log_history": [
{
"epoch": 0.0,
"grad_norm": 74.09103341509594,
"learning_rate": 1.015228426395939e-06,
"loss": 17.1746,
"step": 1
},
{
"epoch": 0.01,
"grad_norm": 72.3290702180604,
"learning_rate": 5.076142131979695e-06,
"loss": 16.9572,
"step": 5
},
{
"epoch": 0.02,
"grad_norm": 68.83490251260932,
"learning_rate": 1.015228426395939e-05,
"loss": 16.4638,
"step": 10
},
{
"epoch": 0.02,
"grad_norm": 66.14177674907667,
"learning_rate": 1.5228426395939088e-05,
"loss": 15.6072,
"step": 15
},
{
"epoch": 0.03,
"grad_norm": 54.81375243549984,
"learning_rate": 2.030456852791878e-05,
"loss": 13.4345,
"step": 20
},
{
"epoch": 0.04,
"grad_norm": 50.488727866909294,
"learning_rate": 2.5380710659898476e-05,
"loss": 10.913,
"step": 25
},
{
"epoch": 0.05,
"grad_norm": 31.023628043144985,
"learning_rate": 3.0456852791878175e-05,
"loss": 7.8116,
"step": 30
},
{
"epoch": 0.05,
"grad_norm": 23.598786792296284,
"learning_rate": 3.553299492385787e-05,
"loss": 5.6786,
"step": 35
},
{
"epoch": 0.06,
"grad_norm": 14.877927520936698,
"learning_rate": 4.060913705583756e-05,
"loss": 4.5198,
"step": 40
},
{
"epoch": 0.07,
"grad_norm": 10.064755961584666,
"learning_rate": 4.568527918781726e-05,
"loss": 3.6186,
"step": 45
},
{
"epoch": 0.08,
"grad_norm": 9.060836838854962,
"learning_rate": 5.076142131979695e-05,
"loss": 2.7201,
"step": 50
},
{
"epoch": 0.08,
"grad_norm": 7.929927938462856,
"learning_rate": 5.583756345177665e-05,
"loss": 2.0934,
"step": 55
},
{
"epoch": 0.09,
"grad_norm": 3.6854754029311034,
"learning_rate": 6.091370558375635e-05,
"loss": 1.7053,
"step": 60
},
{
"epoch": 0.1,
"grad_norm": 3.113155800813489,
"learning_rate": 6.598984771573604e-05,
"loss": 1.5882,
"step": 65
},
{
"epoch": 0.11,
"grad_norm": 2.516100336604323,
"learning_rate": 7.106598984771574e-05,
"loss": 1.4144,
"step": 70
},
{
"epoch": 0.11,
"grad_norm": 2.301065817041555,
"learning_rate": 7.614213197969543e-05,
"loss": 1.2941,
"step": 75
},
{
"epoch": 0.12,
"grad_norm": 1.9632078441101082,
"learning_rate": 8.121827411167512e-05,
"loss": 1.3128,
"step": 80
},
{
"epoch": 0.13,
"grad_norm": 2.0984053055137286,
"learning_rate": 8.629441624365483e-05,
"loss": 1.3046,
"step": 85
},
{
"epoch": 0.14,
"grad_norm": 1.931380727499494,
"learning_rate": 9.137055837563452e-05,
"loss": 1.3041,
"step": 90
},
{
"epoch": 0.14,
"grad_norm": 1.6180374677897205,
"learning_rate": 9.644670050761421e-05,
"loss": 1.21,
"step": 95
},
{
"epoch": 0.15,
"grad_norm": 2.0950245146438564,
"learning_rate": 0.0001015228426395939,
"loss": 1.21,
"step": 100
},
{
"epoch": 0.16,
"grad_norm": 2.1189031385931925,
"learning_rate": 0.00010659898477157362,
"loss": 1.1054,
"step": 105
},
{
"epoch": 0.17,
"grad_norm": 2.015635255767197,
"learning_rate": 0.0001116751269035533,
"loss": 1.1536,
"step": 110
},
{
"epoch": 0.18,
"grad_norm": 1.3137399170764255,
"learning_rate": 0.000116751269035533,
"loss": 1.0306,
"step": 115
},
{
"epoch": 0.18,
"grad_norm": 1.4459264638000027,
"learning_rate": 0.0001218274111675127,
"loss": 1.0956,
"step": 120
},
{
"epoch": 0.19,
"grad_norm": 1.8464804044878194,
"learning_rate": 0.0001269035532994924,
"loss": 1.1259,
"step": 125
},
{
"epoch": 0.2,
"grad_norm": 2.1383334768591142,
"learning_rate": 0.00013197969543147207,
"loss": 1.0779,
"step": 130
},
{
"epoch": 0.21,
"grad_norm": 2.816865870724176,
"learning_rate": 0.00013705583756345178,
"loss": 1.0787,
"step": 135
},
{
"epoch": 0.21,
"grad_norm": 1.514643987387779,
"learning_rate": 0.00014213197969543148,
"loss": 1.1153,
"step": 140
},
{
"epoch": 0.22,
"grad_norm": 1.9840936459404523,
"learning_rate": 0.00014720812182741116,
"loss": 1.0968,
"step": 145
},
{
"epoch": 0.23,
"grad_norm": 2.1892173343482852,
"learning_rate": 0.00015228426395939087,
"loss": 1.0693,
"step": 150
},
{
"epoch": 0.24,
"grad_norm": 2.0955352886552725,
"learning_rate": 0.00015736040609137057,
"loss": 1.0467,
"step": 155
},
{
"epoch": 0.24,
"grad_norm": 1.989968599726472,
"learning_rate": 0.00016243654822335025,
"loss": 0.9779,
"step": 160
},
{
"epoch": 0.25,
"grad_norm": 1.4084167700372459,
"learning_rate": 0.00016751269035532995,
"loss": 0.9782,
"step": 165
},
{
"epoch": 0.26,
"grad_norm": 1.850084427217725,
"learning_rate": 0.00017258883248730966,
"loss": 0.9894,
"step": 170
},
{
"epoch": 0.27,
"grad_norm": 2.035510825701115,
"learning_rate": 0.00017766497461928934,
"loss": 1.0363,
"step": 175
},
{
"epoch": 0.27,
"grad_norm": 1.833391476127078,
"learning_rate": 0.00018274111675126904,
"loss": 1.0404,
"step": 180
},
{
"epoch": 0.28,
"grad_norm": 2.166606883573874,
"learning_rate": 0.00018781725888324875,
"loss": 1.0367,
"step": 185
},
{
"epoch": 0.29,
"grad_norm": 1.803084621710003,
"learning_rate": 0.00019289340101522843,
"loss": 1.0255,
"step": 190
},
{
"epoch": 0.3,
"grad_norm": 1.624927933213703,
"learning_rate": 0.00019796954314720813,
"loss": 0.9999,
"step": 195
},
{
"epoch": 0.31,
"grad_norm": 1.6910332351905768,
"learning_rate": 0.00019999857915498025,
"loss": 1.0179,
"step": 200
},
{
"epoch": 0.31,
"grad_norm": 1.7237344176394134,
"learning_rate": 0.00019998989635940995,
"loss": 1.073,
"step": 205
},
{
"epoch": 0.32,
"grad_norm": 1.5125995515843347,
"learning_rate": 0.00019997332081116373,
"loss": 0.9436,
"step": 210
},
{
"epoch": 0.33,
"grad_norm": 1.4259560737656336,
"learning_rate": 0.00019994885381864085,
"loss": 1.0299,
"step": 215
},
{
"epoch": 0.34,
"grad_norm": 1.745342365482267,
"learning_rate": 0.00019991649731315554,
"loss": 1.0236,
"step": 220
},
{
"epoch": 0.34,
"grad_norm": 1.347141268066664,
"learning_rate": 0.0001998762538487849,
"loss": 1.0469,
"step": 225
},
{
"epoch": 0.35,
"grad_norm": 1.6280396542343814,
"learning_rate": 0.00019982812660216702,
"loss": 1.0283,
"step": 230
},
{
"epoch": 0.36,
"grad_norm": 1.8616024137182106,
"learning_rate": 0.0001997721193722502,
"loss": 1.018,
"step": 235
},
{
"epoch": 0.37,
"grad_norm": 1.6748163706768688,
"learning_rate": 0.00019970823657999327,
"loss": 0.9641,
"step": 240
},
{
"epoch": 0.37,
"grad_norm": 1.8217157997237183,
"learning_rate": 0.00019963648326801653,
"loss": 0.9978,
"step": 245
},
{
"epoch": 0.38,
"grad_norm": 1.7192183617890824,
"learning_rate": 0.00019955686510020372,
"loss": 1.0003,
"step": 250
},
{
"epoch": 0.39,
"grad_norm": 1.6670101592399749,
"learning_rate": 0.00019946938836125483,
"loss": 1.0061,
"step": 255
},
{
"epoch": 0.4,
"grad_norm": 2.0401527430040987,
"learning_rate": 0.00019937405995619023,
"loss": 1.0114,
"step": 260
},
{
"epoch": 0.4,
"grad_norm": 1.5731808739124065,
"learning_rate": 0.0001992708874098054,
"loss": 0.9585,
"step": 265
},
{
"epoch": 0.41,
"grad_norm": 1.2522516219731639,
"learning_rate": 0.00019915987886607713,
"loss": 0.9405,
"step": 270
},
{
"epoch": 0.42,
"grad_norm": 1.5809359876057816,
"learning_rate": 0.0001990410430875205,
"loss": 0.9561,
"step": 275
},
{
"epoch": 0.43,
"grad_norm": 1.4808669010565572,
"learning_rate": 0.00019891438945449745,
"loss": 0.9935,
"step": 280
},
{
"epoch": 0.43,
"grad_norm": 1.4718483710588592,
"learning_rate": 0.00019877992796447601,
"loss": 0.951,
"step": 285
},
{
"epoch": 0.44,
"grad_norm": 1.2492130220680642,
"learning_rate": 0.00019863766923124148,
"loss": 0.9418,
"step": 290
},
{
"epoch": 0.45,
"grad_norm": 1.3890861204364473,
"learning_rate": 0.00019848762448405834,
"loss": 0.9585,
"step": 295
},
{
"epoch": 0.46,
"grad_norm": 2.1030414314285437,
"learning_rate": 0.00019832980556678406,
"loss": 0.9363,
"step": 300
},
{
"epoch": 0.47,
"grad_norm": 1.3041154321040325,
"learning_rate": 0.00019816422493693417,
"loss": 0.9555,
"step": 305
},
{
"epoch": 0.47,
"grad_norm": 1.4561950481471824,
"learning_rate": 0.0001979908956646988,
"loss": 1.006,
"step": 310
},
{
"epoch": 0.48,
"grad_norm": 1.3447506745274114,
"learning_rate": 0.00019780983143191102,
"loss": 0.938,
"step": 315
},
{
"epoch": 0.49,
"grad_norm": 1.7691372827404117,
"learning_rate": 0.00019762104653096708,
"loss": 0.9217,
"step": 320
},
{
"epoch": 0.5,
"grad_norm": 1.6326336607001684,
"learning_rate": 0.00019742455586369784,
"loss": 0.9642,
"step": 325
},
{
"epoch": 0.5,
"grad_norm": 1.6469761536298397,
"learning_rate": 0.0001972203749401928,
"loss": 0.9049,
"step": 330
},
{
"epoch": 0.51,
"grad_norm": 1.4156285399996218,
"learning_rate": 0.0001970085198775757,
"loss": 0.9832,
"step": 335
},
{
"epoch": 0.52,
"grad_norm": 1.2369154811780847,
"learning_rate": 0.00019678900739873226,
"loss": 0.9586,
"step": 340
},
{
"epoch": 0.53,
"grad_norm": 1.184783040387616,
"learning_rate": 0.00019656185483099024,
"loss": 0.9512,
"step": 345
},
{
"epoch": 0.53,
"grad_norm": 1.5777666184106938,
"learning_rate": 0.00019632708010475165,
"loss": 0.9688,
"step": 350
},
{
"epoch": 0.54,
"grad_norm": 1.5444096080267231,
"learning_rate": 0.00019608470175207733,
"loss": 0.926,
"step": 355
},
{
"epoch": 0.55,
"grad_norm": 1.5314192534663842,
"learning_rate": 0.00019583473890522432,
"loss": 0.9758,
"step": 360
},
{
"epoch": 0.56,
"grad_norm": 1.7369618751197407,
"learning_rate": 0.0001955772112951354,
"loss": 0.8986,
"step": 365
},
{
"epoch": 0.56,
"grad_norm": 1.9574089323769333,
"learning_rate": 0.00019531213924988176,
"loss": 0.9164,
"step": 370
},
{
"epoch": 0.57,
"grad_norm": 1.2296445171044836,
"learning_rate": 0.00019503954369305847,
"loss": 0.9108,
"step": 375
},
{
"epoch": 0.58,
"grad_norm": 1.4908276926290502,
"learning_rate": 0.00019475944614213262,
"loss": 0.9318,
"step": 380
},
{
"epoch": 0.59,
"grad_norm": 1.3086171099702917,
"learning_rate": 0.00019447186870674506,
"loss": 0.9266,
"step": 385
},
{
"epoch": 0.59,
"grad_norm": 1.1788811143001057,
"learning_rate": 0.00019417683408696503,
"loss": 0.9388,
"step": 390
},
{
"epoch": 0.6,
"grad_norm": 1.1833646675888452,
"learning_rate": 0.00019387436557149841,
"loss": 0.982,
"step": 395
},
{
"epoch": 0.61,
"grad_norm": 1.4649925294006232,
"learning_rate": 0.00019356448703584933,
"loss": 0.9697,
"step": 400
},
{
"epoch": 0.62,
"grad_norm": 1.3041908495917867,
"learning_rate": 0.00019324722294043558,
"loss": 0.9623,
"step": 405
},
{
"epoch": 0.63,
"grad_norm": 1.3330225209891455,
"learning_rate": 0.00019292259832865785,
"loss": 0.9365,
"step": 410
},
{
"epoch": 0.63,
"grad_norm": 1.2450996327905286,
"learning_rate": 0.00019259063882492292,
"loss": 1.0171,
"step": 415
},
{
"epoch": 0.64,
"grad_norm": 1.3505588278305898,
"learning_rate": 0.00019225137063262084,
"loss": 0.9522,
"step": 420
},
{
"epoch": 0.65,
"grad_norm": 1.884935596805578,
"learning_rate": 0.00019190482053205673,
"loss": 0.922,
"step": 425
},
{
"epoch": 0.66,
"grad_norm": 1.549801647217274,
"learning_rate": 0.00019155101587833677,
"loss": 0.9534,
"step": 430
},
{
"epoch": 0.66,
"grad_norm": 1.0786713746828498,
"learning_rate": 0.00019118998459920902,
"loss": 0.9176,
"step": 435
},
{
"epoch": 0.67,
"grad_norm": 1.1098145056921047,
"learning_rate": 0.0001908217551928587,
"loss": 0.9351,
"step": 440
},
{
"epoch": 0.68,
"grad_norm": 1.478688514055702,
"learning_rate": 0.00019044635672565895,
"loss": 0.9469,
"step": 445
},
{
"epoch": 0.69,
"grad_norm": 1.015346045441744,
"learning_rate": 0.00019006381882987627,
"loss": 0.9703,
"step": 450
},
{
"epoch": 0.69,
"grad_norm": 1.1820782296841172,
"learning_rate": 0.0001896741717013315,
"loss": 0.9601,
"step": 455
},
{
"epoch": 0.7,
"grad_norm": 1.2966754894510633,
"learning_rate": 0.00018927744609701643,
"loss": 0.9605,
"step": 460
},
{
"epoch": 0.71,
"grad_norm": 1.0403285337969717,
"learning_rate": 0.00018887367333266582,
"loss": 0.9418,
"step": 465
},
{
"epoch": 0.72,
"grad_norm": 1.1722720307755863,
"learning_rate": 0.00018846288528028555,
"loss": 0.9741,
"step": 470
},
{
"epoch": 0.72,
"grad_norm": 1.2117042399971847,
"learning_rate": 0.0001880451143656368,
"loss": 0.9435,
"step": 475
},
{
"epoch": 0.73,
"grad_norm": 1.0509023160172017,
"learning_rate": 0.00018762039356567655,
"loss": 0.8706,
"step": 480
},
{
"epoch": 0.74,
"grad_norm": 1.2265158396763423,
"learning_rate": 0.00018718875640595434,
"loss": 0.9051,
"step": 485
},
{
"epoch": 0.75,
"grad_norm": 1.2935461295547515,
"learning_rate": 0.00018675023695796607,
"loss": 0.9758,
"step": 490
},
{
"epoch": 0.76,
"grad_norm": 1.5267873012858113,
"learning_rate": 0.00018630486983646467,
"loss": 0.9244,
"step": 495
},
{
"epoch": 0.76,
"grad_norm": 1.2742929372558043,
"learning_rate": 0.00018585269019672743,
"loss": 0.928,
"step": 500
},
{
"epoch": 0.77,
"grad_norm": 1.2358049387637984,
"learning_rate": 0.00018539373373178123,
"loss": 0.966,
"step": 505
},
{
"epoch": 0.78,
"grad_norm": 1.0460726575553243,
"learning_rate": 0.00018492803666958515,
"loss": 0.9259,
"step": 510
},
{
"epoch": 0.79,
"grad_norm": 1.361928704446451,
"learning_rate": 0.00018445563577017062,
"loss": 0.9742,
"step": 515
},
{
"epoch": 0.79,
"grad_norm": 1.139407531386836,
"learning_rate": 0.0001839765683227398,
"loss": 0.9318,
"step": 520
},
{
"epoch": 0.8,
"grad_norm": 1.242505633372625,
"learning_rate": 0.00018349087214272224,
"loss": 0.905,
"step": 525
},
{
"epoch": 0.81,
"grad_norm": 1.2336955558098357,
"learning_rate": 0.00018299858556878972,
"loss": 0.9233,
"step": 530
},
{
"epoch": 0.82,
"grad_norm": 1.1643115008804164,
"learning_rate": 0.00018249974745983023,
"loss": 0.8991,
"step": 535
},
{
"epoch": 0.82,
"grad_norm": 1.2435271188335837,
"learning_rate": 0.00018199439719188034,
"loss": 0.8918,
"step": 540
},
{
"epoch": 0.83,
"grad_norm": 1.3025449616558689,
"learning_rate": 0.00018148257465501716,
"loss": 0.9272,
"step": 545
},
{
"epoch": 0.84,
"grad_norm": 1.639151201490911,
"learning_rate": 0.0001809643202502097,
"loss": 0.9188,
"step": 550
},
{
"epoch": 0.85,
"grad_norm": 1.1046258635499404,
"learning_rate": 0.00018043967488612955,
"loss": 0.846,
"step": 555
},
{
"epoch": 0.85,
"grad_norm": 1.1404910374403605,
"learning_rate": 0.00017990867997592203,
"loss": 0.9187,
"step": 560
},
{
"epoch": 0.86,
"grad_norm": 1.1924053040242124,
"learning_rate": 0.00017937137743393694,
"loss": 0.9046,
"step": 565
},
{
"epoch": 0.87,
"grad_norm": 1.0135630618037776,
"learning_rate": 0.0001788278096724203,
"loss": 0.9146,
"step": 570
},
{
"epoch": 0.88,
"grad_norm": 0.9391119950422172,
"learning_rate": 0.00017827801959816625,
"loss": 0.9343,
"step": 575
},
{
"epoch": 0.88,
"grad_norm": 1.1918282771197495,
"learning_rate": 0.0001777220506091303,
"loss": 0.912,
"step": 580
},
{
"epoch": 0.89,
"grad_norm": 1.3729834554449354,
"learning_rate": 0.00017715994659100388,
"loss": 0.9561,
"step": 585
},
{
"epoch": 0.9,
"grad_norm": 1.4315867303078007,
"learning_rate": 0.00017659175191374983,
"loss": 0.9503,
"step": 590
},
{
"epoch": 0.91,
"grad_norm": 1.1673639446932986,
"learning_rate": 0.0001760175114281003,
"loss": 0.8902,
"step": 595
},
{
"epoch": 0.92,
"grad_norm": 1.230158811259546,
"learning_rate": 0.0001754372704620164,
"loss": 0.8828,
"step": 600
},
{
"epoch": 0.92,
"grad_norm": 1.1728800163063369,
"learning_rate": 0.00017485107481711012,
"loss": 0.9118,
"step": 605
},
{
"epoch": 0.93,
"grad_norm": 1.0131739031610576,
"learning_rate": 0.00017425897076502902,
"loss": 0.9403,
"step": 610
},
{
"epoch": 0.94,
"grad_norm": 1.351934812093798,
"learning_rate": 0.00017366100504380378,
"loss": 0.8876,
"step": 615
},
{
"epoch": 0.95,
"grad_norm": 1.0384035475562403,
"learning_rate": 0.00017305722485415882,
"loss": 0.9159,
"step": 620
},
{
"epoch": 0.95,
"grad_norm": 1.1416340076156324,
"learning_rate": 0.0001724476778557866,
"loss": 0.9358,
"step": 625
},
{
"epoch": 0.96,
"grad_norm": 1.3419613876196461,
"learning_rate": 0.00017183241216358553,
"loss": 0.9729,
"step": 630
},
{
"epoch": 0.97,
"grad_norm": 1.6092824682901703,
"learning_rate": 0.00017121147634386192,
"loss": 0.8965,
"step": 635
},
{
"epoch": 0.98,
"grad_norm": 0.8533351436444618,
"learning_rate": 0.00017058491941049649,
"loss": 0.8917,
"step": 640
},
{
"epoch": 0.98,
"grad_norm": 1.1738959779265217,
"learning_rate": 0.0001699527908210754,
"loss": 0.9366,
"step": 645
},
{
"epoch": 0.99,
"grad_norm": 1.0847553199067752,
"learning_rate": 0.0001693151404729862,
"loss": 0.9053,
"step": 650
},
{
"epoch": 1.0,
"grad_norm": 1.1087892770915242,
"learning_rate": 0.00016867201869947938,
"loss": 0.8945,
"step": 655
},
{
"epoch": 1.0,
"eval_loss": 1.2306770086288452,
"eval_runtime": 248.745,
"eval_samples_per_second": 9.287,
"eval_steps_per_second": 0.583,
"step": 655
},
{
"epoch": 1.01,
"grad_norm": 0.98970356590023,
"learning_rate": 0.00016802347626569501,
"loss": 0.8931,
"step": 660
},
{
"epoch": 1.01,
"grad_norm": 1.1038579119403917,
"learning_rate": 0.00016736956436465573,
"loss": 0.934,
"step": 665
},
{
"epoch": 1.02,
"grad_norm": 1.143467118458221,
"learning_rate": 0.00016671033461322583,
"loss": 0.8668,
"step": 670
},
{
"epoch": 1.03,
"grad_norm": 0.8924280436486196,
"learning_rate": 0.00016604583904803672,
"loss": 0.8548,
"step": 675
},
{
"epoch": 1.04,
"grad_norm": 1.1115437180292647,
"learning_rate": 0.0001653761301213795,
"loss": 0.9385,
"step": 680
},
{
"epoch": 1.05,
"grad_norm": 1.6438519669074836,
"learning_rate": 0.00016470126069706456,
"loss": 0.9406,
"step": 685
},
{
"epoch": 1.05,
"grad_norm": 1.0490682410760337,
"learning_rate": 0.00016402128404624882,
"loss": 0.9381,
"step": 690
},
{
"epoch": 1.06,
"grad_norm": 1.0451537501943204,
"learning_rate": 0.0001633362538432306,
"loss": 0.9113,
"step": 695
},
{
"epoch": 1.07,
"grad_norm": 0.9747283917774375,
"learning_rate": 0.0001626462241612129,
"loss": 0.9121,
"step": 700
},
{
"epoch": 1.08,
"grad_norm": 1.084078539640254,
"learning_rate": 0.00016195124946803525,
"loss": 0.8831,
"step": 705
},
{
"epoch": 1.08,
"grad_norm": 0.9719667527581831,
"learning_rate": 0.00016125138462187396,
"loss": 0.8436,
"step": 710
},
{
"epoch": 1.09,
"grad_norm": 0.9935638920294466,
"learning_rate": 0.00016054668486691203,
"loss": 0.9196,
"step": 715
},
{
"epoch": 1.1,
"grad_norm": 0.9417823234407828,
"learning_rate": 0.00015983720582897853,
"loss": 0.8577,
"step": 720
},
{
"epoch": 1.11,
"grad_norm": 0.8915998902762848,
"learning_rate": 0.0001591230035111576,
"loss": 0.853,
"step": 725
},
{
"epoch": 1.11,
"grad_norm": 1.01108160646832,
"learning_rate": 0.00015840413428936767,
"loss": 0.9148,
"step": 730
},
{
"epoch": 1.12,
"grad_norm": 0.9944405742170609,
"learning_rate": 0.00015768065490791183,
"loss": 0.9151,
"step": 735
},
{
"epoch": 1.13,
"grad_norm": 0.9869449635796341,
"learning_rate": 0.0001569526224749983,
"loss": 0.9144,
"step": 740
},
{
"epoch": 1.14,
"grad_norm": 1.5578640706977607,
"learning_rate": 0.00015622009445823273,
"loss": 0.9241,
"step": 745
},
{
"epoch": 1.14,
"grad_norm": 1.0818607624486927,
"learning_rate": 0.00015548312868008196,
"loss": 0.9123,
"step": 750
},
{
"epoch": 1.15,
"grad_norm": 0.9747826539453109,
"learning_rate": 0.0001547417833133097,
"loss": 0.8666,
"step": 755
},
{
"epoch": 1.16,
"grad_norm": 0.9965768127635756,
"learning_rate": 0.00015399611687638473,
"loss": 0.8808,
"step": 760
},
{
"epoch": 1.17,
"grad_norm": 1.1911947572353123,
"learning_rate": 0.00015324618822886165,
"loss": 0.8824,
"step": 765
},
{
"epoch": 1.17,
"grad_norm": 1.1537563580786732,
"learning_rate": 0.00015249205656673483,
"loss": 0.863,
"step": 770
},
{
"epoch": 1.18,
"grad_norm": 1.0518456260292857,
"learning_rate": 0.00015173378141776568,
"loss": 0.8812,
"step": 775
},
{
"epoch": 1.19,
"grad_norm": 1.387022437551768,
"learning_rate": 0.00015097142263678382,
"loss": 0.8835,
"step": 780
},
{
"epoch": 1.2,
"grad_norm": 1.0408593407980289,
"learning_rate": 0.0001502050404009624,
"loss": 0.94,
"step": 785
},
{
"epoch": 1.21,
"grad_norm": 0.9105554748085787,
"learning_rate": 0.00014943469520506807,
"loss": 0.9024,
"step": 790
},
{
"epoch": 1.21,
"grad_norm": 1.0445125806761943,
"learning_rate": 0.00014866044785668563,
"loss": 0.8973,
"step": 795
},
{
"epoch": 1.22,
"grad_norm": 1.0453562314924396,
"learning_rate": 0.00014788235947141832,
"loss": 0.9402,
"step": 800
},
{
"epoch": 1.23,
"grad_norm": 0.9691509611927153,
"learning_rate": 0.00014710049146806348,
"loss": 0.8924,
"step": 805
},
{
"epoch": 1.24,
"grad_norm": 1.1002288592485436,
"learning_rate": 0.00014631490556376458,
"loss": 0.8535,
"step": 810
},
{
"epoch": 1.24,
"grad_norm": 1.0588060898797662,
"learning_rate": 0.00014552566376913943,
"loss": 0.9215,
"step": 815
},
{
"epoch": 1.25,
"grad_norm": 1.1064064596494758,
"learning_rate": 0.0001447328283833854,
"loss": 0.8822,
"step": 820
},
{
"epoch": 1.26,
"grad_norm": 0.9443323504920631,
"learning_rate": 0.00014393646198936167,
"loss": 0.8519,
"step": 825
},
{
"epoch": 1.27,
"grad_norm": 1.0288834578773713,
"learning_rate": 0.00014313662744864955,
"loss": 0.8127,
"step": 830
},
{
"epoch": 1.27,
"grad_norm": 0.9217642107444823,
"learning_rate": 0.00014233338789659006,
"loss": 0.9436,
"step": 835
},
{
"epoch": 1.28,
"grad_norm": 1.1636504558701406,
"learning_rate": 0.00014152680673730058,
"loss": 0.8811,
"step": 840
},
{
"epoch": 1.29,
"grad_norm": 1.0596219737156714,
"learning_rate": 0.00014071694763866986,
"loss": 0.9467,
"step": 845
},
{
"epoch": 1.3,
"grad_norm": 1.2066959377309563,
"learning_rate": 0.00013990387452733255,
"loss": 0.8134,
"step": 850
},
{
"epoch": 1.3,
"grad_norm": 1.0002794262680956,
"learning_rate": 0.00013908765158362297,
"loss": 0.8525,
"step": 855
},
{
"epoch": 1.31,
"grad_norm": 0.9370583121034902,
"learning_rate": 0.000138268343236509,
"loss": 0.9415,
"step": 860
},
{
"epoch": 1.32,
"grad_norm": 1.068438209489481,
"learning_rate": 0.00013744601415850637,
"loss": 0.8345,
"step": 865
},
{
"epoch": 1.33,
"grad_norm": 1.0013512513559877,
"learning_rate": 0.00013662072926057383,
"loss": 0.9228,
"step": 870
},
{
"epoch": 1.33,
"grad_norm": 1.1009082230050427,
"learning_rate": 0.00013579255368698914,
"loss": 0.8734,
"step": 875
},
{
"epoch": 1.34,
"grad_norm": 0.9267504277577387,
"learning_rate": 0.00013496155281020707,
"loss": 0.883,
"step": 880
},
{
"epoch": 1.35,
"grad_norm": 0.9055721244854339,
"learning_rate": 0.00013412779222569908,
"loss": 0.909,
"step": 885
},
{
"epoch": 1.36,
"grad_norm": 1.0647099863896714,
"learning_rate": 0.00013329133774677552,
"loss": 0.8752,
"step": 890
},
{
"epoch": 1.37,
"grad_norm": 1.0280795070472157,
"learning_rate": 0.00013245225539939073,
"loss": 0.846,
"step": 895
},
{
"epoch": 1.37,
"grad_norm": 0.9768719689574438,
"learning_rate": 0.0001316106114169311,
"loss": 0.8497,
"step": 900
},
{
"epoch": 1.38,
"grad_norm": 1.1309395586064876,
"learning_rate": 0.00013076647223498702,
"loss": 0.9077,
"step": 905
},
{
"epoch": 1.39,
"grad_norm": 1.2045776660696499,
"learning_rate": 0.0001299199044861086,
"loss": 0.8688,
"step": 910
},
{
"epoch": 1.4,
"grad_norm": 0.9841428220188774,
"learning_rate": 0.0001290709749945463,
"loss": 0.8629,
"step": 915
},
{
"epoch": 1.4,
"grad_norm": 1.325435278640866,
"learning_rate": 0.0001282197507709757,
"loss": 0.8845,
"step": 920
},
{
"epoch": 1.41,
"grad_norm": 1.001414834995417,
"learning_rate": 0.0001273662990072083,
"loss": 0.8362,
"step": 925
},
{
"epoch": 1.42,
"grad_norm": 0.9850158774989259,
"learning_rate": 0.00012651068707088766,
"loss": 0.9111,
"step": 930
},
{
"epoch": 1.43,
"grad_norm": 1.125415537788732,
"learning_rate": 0.00012565298250017155,
"loss": 0.8847,
"step": 935
},
{
"epoch": 1.43,
"grad_norm": 0.9312415927403098,
"learning_rate": 0.00012479325299840095,
"loss": 0.8961,
"step": 940
},
{
"epoch": 1.44,
"grad_norm": 1.0186566605726637,
"learning_rate": 0.0001239315664287558,
"loss": 0.9071,
"step": 945
},
{
"epoch": 1.45,
"grad_norm": 0.9587942024148747,
"learning_rate": 0.00012306799080889807,
"loss": 0.8553,
"step": 950
},
{
"epoch": 1.46,
"grad_norm": 1.0677123133883124,
"learning_rate": 0.000122202594305603,
"loss": 0.8459,
"step": 955
},
{
"epoch": 1.46,
"grad_norm": 0.8807263375032743,
"learning_rate": 0.00012133544522937805,
"loss": 0.8544,
"step": 960
},
{
"epoch": 1.47,
"grad_norm": 1.0893135979400062,
"learning_rate": 0.00012046661202907101,
"loss": 0.8967,
"step": 965
},
{
"epoch": 1.48,
"grad_norm": 1.2654715562850305,
"learning_rate": 0.00011959616328646678,
"loss": 0.8554,
"step": 970
},
{
"epoch": 1.49,
"grad_norm": 0.8250130976731411,
"learning_rate": 0.00011872416771087395,
"loss": 0.9238,
"step": 975
},
{
"epoch": 1.5,
"grad_norm": 0.978527504635651,
"learning_rate": 0.00011785069413370111,
"loss": 0.8132,
"step": 980
},
{
"epoch": 1.5,
"grad_norm": 0.9818769307982783,
"learning_rate": 0.00011697581150302362,
"loss": 0.8541,
"step": 985
},
{
"epoch": 1.51,
"grad_norm": 0.951052968962308,
"learning_rate": 0.00011609958887814129,
"loss": 0.8814,
"step": 990
},
{
"epoch": 1.52,
"grad_norm": 0.9722385611491896,
"learning_rate": 0.00011522209542412696,
"loss": 0.8814,
"step": 995
},
{
"epoch": 1.53,
"grad_norm": 1.1517052356748192,
"learning_rate": 0.00011434340040636704,
"loss": 0.8906,
"step": 1000
},
{
"epoch": 1.53,
"grad_norm": 0.9501728523383044,
"learning_rate": 0.00011346357318509395,
"loss": 0.8791,
"step": 1005
},
{
"epoch": 1.54,
"grad_norm": 0.9615870877113771,
"learning_rate": 0.00011258268320991119,
"loss": 0.8547,
"step": 1010
},
{
"epoch": 1.55,
"grad_norm": 1.1528161628232605,
"learning_rate": 0.0001117008000143112,
"loss": 0.8321,
"step": 1015
},
{
"epoch": 1.56,
"grad_norm": 1.0959125037961595,
"learning_rate": 0.00011081799321018695,
"loss": 0.8782,
"step": 1020
},
{
"epoch": 1.56,
"grad_norm": 1.1574849427468559,
"learning_rate": 0.00010993433248233671,
"loss": 0.8611,
"step": 1025
},
{
"epoch": 1.57,
"grad_norm": 0.9005752589116309,
"learning_rate": 0.0001090498875829638,
"loss": 0.9169,
"step": 1030
},
{
"epoch": 1.58,
"grad_norm": 0.9992289604264046,
"learning_rate": 0.00010816472832617037,
"loss": 0.8925,
"step": 1035
},
{
"epoch": 1.59,
"grad_norm": 1.1498675103760203,
"learning_rate": 0.00010727892458244686,
"loss": 0.8942,
"step": 1040
},
{
"epoch": 1.59,
"grad_norm": 0.9081946761117776,
"learning_rate": 0.00010639254627315657,
"loss": 0.8955,
"step": 1045
},
{
"epoch": 1.6,
"grad_norm": 1.0275540701494208,
"learning_rate": 0.00010550566336501637,
"loss": 0.8442,
"step": 1050
},
{
"epoch": 1.61,
"grad_norm": 0.9434351529699477,
"learning_rate": 0.00010461834586457398,
"loss": 0.9419,
"step": 1055
},
{
"epoch": 1.62,
"grad_norm": 0.8829537272599468,
"learning_rate": 0.0001037306638126818,
"loss": 0.9257,
"step": 1060
},
{
"epoch": 1.62,
"grad_norm": 1.1223033845867914,
"learning_rate": 0.00010284268727896833,
"loss": 0.9161,
"step": 1065
},
{
"epoch": 1.63,
"grad_norm": 1.2421465538773482,
"learning_rate": 0.00010195448635630708,
"loss": 0.8701,
"step": 1070
},
{
"epoch": 1.64,
"grad_norm": 1.1218568027877072,
"learning_rate": 0.00010106613115528398,
"loss": 0.9046,
"step": 1075
},
{
"epoch": 1.65,
"grad_norm": 1.0337432513754021,
"learning_rate": 0.00010017769179866279,
"loss": 0.8609,
"step": 1080
},
{
"epoch": 1.66,
"grad_norm": 0.8566019611964384,
"learning_rate": 9.928923841585024e-05,
"loss": 0.8153,
"step": 1085
},
{
"epoch": 1.66,
"grad_norm": 0.9670758845502253,
"learning_rate": 9.840084113736024e-05,
"loss": 0.8968,
"step": 1090
},
{
"epoch": 1.67,
"grad_norm": 0.8720532251402997,
"learning_rate": 9.751257008927803e-05,
"loss": 0.9142,
"step": 1095
},
{
"epoch": 1.68,
"grad_norm": 1.005265233752645,
"learning_rate": 9.66244953877248e-05,
"loss": 0.8797,
"step": 1100
},
{
"epoch": 1.69,
"grad_norm": 1.0788499372739098,
"learning_rate": 9.573668713332304e-05,
"loss": 0.8721,
"step": 1105
},
{
"epoch": 1.69,
"grad_norm": 0.959383830376907,
"learning_rate": 9.484921540566307e-05,
"loss": 0.9281,
"step": 1110
},
{
"epoch": 1.7,
"grad_norm": 0.9960816420866151,
"learning_rate": 9.396215025777139e-05,
"loss": 0.8946,
"step": 1115
},
{
"epoch": 1.71,
"grad_norm": 0.9183171097472164,
"learning_rate": 9.307556171058085e-05,
"loss": 0.8727,
"step": 1120
},
{
"epoch": 1.72,
"grad_norm": 0.9320733649944887,
"learning_rate": 9.218951974740354e-05,
"loss": 0.9172,
"step": 1125
},
{
"epoch": 1.72,
"grad_norm": 0.8820875642651754,
"learning_rate": 9.130409430840678e-05,
"loss": 0.903,
"step": 1130
},
{
"epoch": 1.73,
"grad_norm": 0.9886857168118836,
"learning_rate": 9.041935528509216e-05,
"loss": 0.919,
"step": 1135
},
{
"epoch": 1.74,
"grad_norm": 1.0243023631867323,
"learning_rate": 8.953537251477874e-05,
"loss": 0.8972,
"step": 1140
},
{
"epoch": 1.75,
"grad_norm": 1.0083372332679066,
"learning_rate": 8.865221577509033e-05,
"loss": 0.8922,
"step": 1145
},
{
"epoch": 1.75,
"grad_norm": 1.0260814204564148,
"learning_rate": 8.776995477844768e-05,
"loss": 0.8516,
"step": 1150
},
{
"epoch": 1.76,
"grad_norm": 0.8142375395224646,
"learning_rate": 8.688865916656547e-05,
"loss": 0.8484,
"step": 1155
},
{
"epoch": 1.77,
"grad_norm": 1.0780009890850535,
"learning_rate": 8.600839850495546e-05,
"loss": 0.8693,
"step": 1160
},
{
"epoch": 1.78,
"grad_norm": 1.34710148555113,
"learning_rate": 8.512924227743481e-05,
"loss": 0.8465,
"step": 1165
},
{
"epoch": 1.78,
"grad_norm": 0.8023152238316216,
"learning_rate": 8.425125988064197e-05,
"loss": 0.8813,
"step": 1170
},
{
"epoch": 1.79,
"grad_norm": 1.0380654418842252,
"learning_rate": 8.337452061855814e-05,
"loss": 0.9174,
"step": 1175
},
{
"epoch": 1.8,
"grad_norm": 1.0726246573015967,
"learning_rate": 8.249909369703728e-05,
"loss": 0.8662,
"step": 1180
},
{
"epoch": 1.81,
"grad_norm": 1.0278553172179552,
"learning_rate": 8.162504821834295e-05,
"loss": 0.8908,
"step": 1185
},
{
"epoch": 1.82,
"grad_norm": 1.3072760509850052,
"learning_rate": 8.075245317569395e-05,
"loss": 0.8671,
"step": 1190
},
{
"epoch": 1.82,
"grad_norm": 0.9627970898558489,
"learning_rate": 7.988137744781805e-05,
"loss": 0.8796,
"step": 1195
},
{
"epoch": 1.83,
"grad_norm": 0.9715464599367558,
"learning_rate": 7.901188979351526e-05,
"loss": 0.8625,
"step": 1200
},
{
"epoch": 1.84,
"grad_norm": 1.0848501852802424,
"learning_rate": 7.814405884623011e-05,
"loss": 0.855,
"step": 1205
},
{
"epoch": 1.85,
"grad_norm": 0.9626487674718053,
"learning_rate": 7.727795310863425e-05,
"loss": 0.9057,
"step": 1210
},
{
"epoch": 1.85,
"grad_norm": 0.8586177531466549,
"learning_rate": 7.64136409472189e-05,
"loss": 0.7599,
"step": 1215
},
{
"epoch": 1.86,
"grad_norm": 0.9382983703101969,
"learning_rate": 7.55511905868985e-05,
"loss": 0.8729,
"step": 1220
},
{
"epoch": 1.87,
"grad_norm": 1.0198546805846869,
"learning_rate": 7.469067010562539e-05,
"loss": 0.9047,
"step": 1225
},
{
"epoch": 1.88,
"grad_norm": 1.027987716065299,
"learning_rate": 7.383214742901573e-05,
"loss": 0.8926,
"step": 1230
},
{
"epoch": 1.88,
"grad_norm": 0.9910955347499838,
"learning_rate": 7.297569032498817e-05,
"loss": 0.9097,
"step": 1235
},
{
"epoch": 1.89,
"grad_norm": 1.108145730506975,
"learning_rate": 7.212136639841418e-05,
"loss": 0.8127,
"step": 1240
},
{
"epoch": 1.9,
"grad_norm": 0.78955284168565,
"learning_rate": 7.126924308578196e-05,
"loss": 0.9052,
"step": 1245
},
{
"epoch": 1.91,
"grad_norm": 0.9172876708503183,
"learning_rate": 7.041938764987297e-05,
"loss": 0.8864,
"step": 1250
},
{
"epoch": 1.91,
"grad_norm": 0.9468127168909404,
"learning_rate": 6.957186717445283e-05,
"loss": 0.8877,
"step": 1255
},
{
"epoch": 1.92,
"grad_norm": 0.9264382383778683,
"learning_rate": 6.872674855897581e-05,
"loss": 0.8407,
"step": 1260
},
{
"epoch": 1.93,
"grad_norm": 0.9055699868252433,
"learning_rate": 6.788409851330418e-05,
"loss": 0.9067,
"step": 1265
},
{
"epoch": 1.94,
"grad_norm": 0.9776429485158797,
"learning_rate": 6.704398355244237e-05,
"loss": 0.8893,
"step": 1270
},
{
"epoch": 1.95,
"grad_norm": 1.1033183037079404,
"learning_rate": 6.620646999128668e-05,
"loss": 0.919,
"step": 1275
},
{
"epoch": 1.95,
"grad_norm": 0.8873640249636788,
"learning_rate": 6.537162393939063e-05,
"loss": 0.8226,
"step": 1280
},
{
"epoch": 1.96,
"grad_norm": 1.0005367354469088,
"learning_rate": 6.453951129574644e-05,
"loss": 0.8982,
"step": 1285
},
{
"epoch": 1.97,
"grad_norm": 0.9756199772914784,
"learning_rate": 6.371019774358351e-05,
"loss": 0.898,
"step": 1290
},
{
"epoch": 1.98,
"grad_norm": 0.8573420891606421,
"learning_rate": 6.288374874518348e-05,
"loss": 0.8824,
"step": 1295
},
{
"epoch": 1.98,
"grad_norm": 0.8314787002741244,
"learning_rate": 6.206022953671303e-05,
"loss": 0.8533,
"step": 1300
},
{
"epoch": 1.99,
"grad_norm": 0.999256628643339,
"learning_rate": 6.123970512307434e-05,
"loss": 0.8766,
"step": 1305
},
{
"epoch": 2.0,
"grad_norm": 1.020370445146433,
"learning_rate": 6.042224027277403e-05,
"loss": 0.8664,
"step": 1310
},
{
"epoch": 2.0,
"eval_loss": 1.1823043823242188,
"eval_runtime": 248.6275,
"eval_samples_per_second": 9.291,
"eval_steps_per_second": 0.583,
"step": 1311
},
{
"epoch": 2.01,
"grad_norm": 0.9076519760868025,
"learning_rate": 5.960789951281052e-05,
"loss": 0.8506,
"step": 1315
},
{
"epoch": 2.01,
"grad_norm": 0.8435077804141212,
"learning_rate": 5.879674712358061e-05,
"loss": 0.8293,
"step": 1320
},
{
"epoch": 2.02,
"grad_norm": 1.2235293049079738,
"learning_rate": 5.798884713380541e-05,
"loss": 0.8727,
"step": 1325
},
{
"epoch": 2.03,
"grad_norm": 0.982943334202588,
"learning_rate": 5.718426331547626e-05,
"loss": 0.8403,
"step": 1330
},
{
"epoch": 2.04,
"grad_norm": 1.0225809341397019,
"learning_rate": 5.638305917882093e-05,
"loss": 0.825,
"step": 1335
},
{
"epoch": 2.04,
"grad_norm": 1.0068602287969126,
"learning_rate": 5.558529796729015e-05,
"loss": 0.8253,
"step": 1340
},
{
"epoch": 2.05,
"grad_norm": 0.8942997135434371,
"learning_rate": 5.479104265256583e-05,
"loss": 0.8093,
"step": 1345
},
{
"epoch": 2.06,
"grad_norm": 0.919639915565314,
"learning_rate": 5.4000355929589985e-05,
"loss": 0.8736,
"step": 1350
},
{
"epoch": 2.07,
"grad_norm": 0.9792431007986793,
"learning_rate": 5.32133002116161e-05,
"loss": 0.8134,
"step": 1355
},
{
"epoch": 2.07,
"grad_norm": 0.9705003699468054,
"learning_rate": 5.242993762528248e-05,
"loss": 0.821,
"step": 1360
},
{
"epoch": 2.08,
"grad_norm": 1.0687461433169771,
"learning_rate": 5.165033000570825e-05,
"loss": 0.8586,
"step": 1365
},
{
"epoch": 2.09,
"grad_norm": 0.9864467254623525,
"learning_rate": 5.087453889161229e-05,
"loss": 0.7872,
"step": 1370
},
{
"epoch": 2.1,
"grad_norm": 0.9022370559155457,
"learning_rate": 5.010262552045573e-05,
"loss": 0.8145,
"step": 1375
},
{
"epoch": 2.11,
"grad_norm": 0.8356213372223533,
"learning_rate": 4.933465082360807e-05,
"loss": 0.8521,
"step": 1380
},
{
"epoch": 2.11,
"grad_norm": 1.0182132748957182,
"learning_rate": 4.857067542153768e-05,
"loss": 0.8567,
"step": 1385
},
{
"epoch": 2.12,
"grad_norm": 1.086022764305583,
"learning_rate": 4.781075961902656e-05,
"loss": 0.804,
"step": 1390
},
{
"epoch": 2.13,
"grad_norm": 1.0333147014472928,
"learning_rate": 4.705496340041009e-05,
"loss": 0.8119,
"step": 1395
},
{
"epoch": 2.14,
"grad_norm": 1.0968659156489957,
"learning_rate": 4.630334642484231e-05,
"loss": 0.8434,
"step": 1400
},
{
"epoch": 2.14,
"grad_norm": 1.1297383258429818,
"learning_rate": 4.555596802158653e-05,
"loss": 0.8388,
"step": 1405
},
{
"epoch": 2.15,
"grad_norm": 0.9894862065125212,
"learning_rate": 4.481288718533238e-05,
"loss": 0.8457,
"step": 1410
},
{
"epoch": 2.16,
"grad_norm": 1.0086593655810214,
"learning_rate": 4.407416257153872e-05,
"loss": 0.8452,
"step": 1415
},
{
"epoch": 2.17,
"grad_norm": 1.077271484795884,
"learning_rate": 4.33398524918039e-05,
"loss": 0.8307,
"step": 1420
},
{
"epoch": 2.17,
"grad_norm": 1.101473877979803,
"learning_rate": 4.261001490926272e-05,
"loss": 0.8228,
"step": 1425
},
{
"epoch": 2.18,
"grad_norm": 1.1024756868438783,
"learning_rate": 4.1884707434011295e-05,
"loss": 0.8455,
"step": 1430
},
{
"epoch": 2.19,
"grad_norm": 0.9596705868289979,
"learning_rate": 4.116398731855934e-05,
"loss": 0.8429,
"step": 1435
},
{
"epoch": 2.2,
"grad_norm": 1.0578699242751421,
"learning_rate": 4.044791145331105e-05,
"loss": 0.7775,
"step": 1440
},
{
"epoch": 2.2,
"grad_norm": 0.8551950143576533,
"learning_rate": 3.973653636207437e-05,
"loss": 0.8328,
"step": 1445
},
{
"epoch": 2.21,
"grad_norm": 1.071052415435941,
"learning_rate": 3.902991819759937e-05,
"loss": 0.8459,
"step": 1450
},
{
"epoch": 2.22,
"grad_norm": 1.0178868820951739,
"learning_rate": 3.832811273714569e-05,
"loss": 0.8003,
"step": 1455
},
{
"epoch": 2.23,
"grad_norm": 1.0388700331104954,
"learning_rate": 3.763117537807972e-05,
"loss": 0.8579,
"step": 1460
},
{
"epoch": 2.23,
"grad_norm": 0.9885214128911158,
"learning_rate": 3.693916113350182e-05,
"loss": 0.8114,
"step": 1465
},
{
"epoch": 2.24,
"grad_norm": 0.9507588869671092,
"learning_rate": 3.625212462790385e-05,
"loss": 0.7656,
"step": 1470
},
{
"epoch": 2.25,
"grad_norm": 1.033384131735441,
"learning_rate": 3.55701200928574e-05,
"loss": 0.7871,
"step": 1475
},
{
"epoch": 2.26,
"grad_norm": 0.9158523594001913,
"learning_rate": 3.489320136273288e-05,
"loss": 0.8385,
"step": 1480
},
{
"epoch": 2.27,
"grad_norm": 1.0861208797350566,
"learning_rate": 3.422142187045011e-05,
"loss": 0.8296,
"step": 1485
},
{
"epoch": 2.27,
"grad_norm": 1.0108364033236483,
"learning_rate": 3.3554834643260624e-05,
"loss": 0.8405,
"step": 1490
},
{
"epoch": 2.28,
"grad_norm": 1.017348028808686,
"learning_rate": 3.289349229856197e-05,
"loss": 0.8041,
"step": 1495
},
{
"epoch": 2.29,
"grad_norm": 0.9720934709242245,
"learning_rate": 3.223744703974413e-05,
"loss": 0.8477,
"step": 1500
},
{
"epoch": 2.3,
"grad_norm": 1.0491113172636481,
"learning_rate": 3.1586750652069074e-05,
"loss": 0.8455,
"step": 1505
},
{
"epoch": 2.3,
"grad_norm": 0.969350149657045,
"learning_rate": 3.094145449858285e-05,
"loss": 0.8373,
"step": 1510
},
{
"epoch": 2.31,
"grad_norm": 1.1292836391329433,
"learning_rate": 3.030160951606138e-05,
"loss": 0.8265,
"step": 1515
},
{
"epoch": 2.32,
"grad_norm": 1.1377550753616885,
"learning_rate": 2.966726621098963e-05,
"loss": 0.8724,
"step": 1520
},
{
"epoch": 2.33,
"grad_norm": 0.9248650902738645,
"learning_rate": 2.9038474655574897e-05,
"loss": 0.8247,
"step": 1525
},
{
"epoch": 2.33,
"grad_norm": 1.24903061509406,
"learning_rate": 2.84152844837943e-05,
"loss": 0.8407,
"step": 1530
},
{
"epoch": 2.34,
"grad_norm": 0.9300456009720597,
"learning_rate": 2.779774488747704e-05,
"loss": 0.8017,
"step": 1535
},
{
"epoch": 2.35,
"grad_norm": 0.9247283493036916,
"learning_rate": 2.7185904612421176e-05,
"loss": 0.7873,
"step": 1540
},
{
"epoch": 2.36,
"grad_norm": 0.984266641348006,
"learning_rate": 2.6579811954546052e-05,
"loss": 0.8231,
"step": 1545
},
{
"epoch": 2.36,
"grad_norm": 1.0283812513796944,
"learning_rate": 2.5979514756079914e-05,
"loss": 0.8149,
"step": 1550
},
{
"epoch": 2.37,
"grad_norm": 1.1106242321896724,
"learning_rate": 2.5385060401783566e-05,
"loss": 0.8276,
"step": 1555
},
{
"epoch": 2.38,
"grad_norm": 0.9567893029629212,
"learning_rate": 2.4796495815209996e-05,
"loss": 0.816,
"step": 1560
},
{
"epoch": 2.39,
"grad_norm": 1.0867782224252835,
"learning_rate": 2.4213867455000338e-05,
"loss": 0.892,
"step": 1565
},
{
"epoch": 2.4,
"grad_norm": 1.1389935740624955,
"learning_rate": 2.363722131121675e-05,
"loss": 0.8478,
"step": 1570
},
{
"epoch": 2.4,
"grad_norm": 0.9718824376096208,
"learning_rate": 2.3066602901712108e-05,
"loss": 0.8496,
"step": 1575
},
{
"epoch": 2.41,
"grad_norm": 1.1771964742364636,
"learning_rate": 2.2502057268537137e-05,
"loss": 0.8449,
"step": 1580
},
{
"epoch": 2.42,
"grad_norm": 1.4160024538664562,
"learning_rate": 2.1943628974384857e-05,
"loss": 0.8491,
"step": 1585
},
{
"epoch": 2.43,
"grad_norm": 1.0415796126412569,
"learning_rate": 2.1391362099073032e-05,
"loss": 0.802,
"step": 1590
},
{
"epoch": 2.43,
"grad_norm": 0.8715990916453478,
"learning_rate": 2.0845300236064757e-05,
"loss": 0.8696,
"step": 1595
},
{
"epoch": 2.44,
"grad_norm": 1.049043541881097,
"learning_rate": 2.0305486489027424e-05,
"loss": 0.8302,
"step": 1600
},
{
"epoch": 2.45,
"grad_norm": 0.9586867830799501,
"learning_rate": 1.9771963468430187e-05,
"loss": 0.8173,
"step": 1605
},
{
"epoch": 2.46,
"grad_norm": 1.0529652964302254,
"learning_rate": 1.924477328818054e-05,
"loss": 0.8617,
"step": 1610
},
{
"epoch": 2.46,
"grad_norm": 1.0449868664649247,
"learning_rate": 1.8723957562300133e-05,
"loss": 0.8291,
"step": 1615
},
{
"epoch": 2.47,
"grad_norm": 1.0564336584456713,
"learning_rate": 1.8209557401639732e-05,
"loss": 0.8175,
"step": 1620
},
{
"epoch": 2.48,
"grad_norm": 0.9643616639994842,
"learning_rate": 1.7701613410634365e-05,
"loss": 0.7645,
"step": 1625
},
{
"epoch": 2.49,
"grad_norm": 1.1870712950124547,
"learning_rate": 1.720016568409799e-05,
"loss": 0.8509,
"step": 1630
},
{
"epoch": 2.49,
"grad_norm": 1.06154190477567,
"learning_rate": 1.6705253804058675e-05,
"loss": 0.8725,
"step": 1635
},
{
"epoch": 2.5,
"grad_norm": 1.0388956977852164,
"learning_rate": 1.621691683663418e-05,
"loss": 0.8139,
"step": 1640
},
{
"epoch": 2.51,
"grad_norm": 1.0678593249410824,
"learning_rate": 1.573519332894824e-05,
"loss": 0.7882,
"step": 1645
},
{
"epoch": 2.52,
"grad_norm": 1.0504305711608912,
"learning_rate": 1.5260121306087828e-05,
"loss": 0.8289,
"step": 1650
},
{
"epoch": 2.52,
"grad_norm": 1.039744935641794,
"learning_rate": 1.4791738268101584e-05,
"loss": 0.7841,
"step": 1655
},
{
"epoch": 2.53,
"grad_norm": 1.0672936964776525,
"learning_rate": 1.433008118703979e-05,
"loss": 0.7895,
"step": 1660
},
{
"epoch": 2.54,
"grad_norm": 1.095135634651846,
"learning_rate": 1.3875186504035965e-05,
"loss": 0.8143,
"step": 1665
},
{
"epoch": 2.55,
"grad_norm": 1.0672595953625994,
"learning_rate": 1.3427090126430342e-05,
"loss": 0.8258,
"step": 1670
},
{
"epoch": 2.56,
"grad_norm": 1.0487848694270845,
"learning_rate": 1.298582742493546e-05,
"loss": 0.8433,
"step": 1675
},
{
"epoch": 2.56,
"grad_norm": 0.8678608016330454,
"learning_rate": 1.2551433230844224e-05,
"loss": 0.7898,
"step": 1680
},
{
"epoch": 2.57,
"grad_norm": 0.929677257792853,
"learning_rate": 1.2123941833280473e-05,
"loss": 0.7707,
"step": 1685
},
{
"epoch": 2.58,
"grad_norm": 1.0288546508719252,
"learning_rate": 1.1703386976492337e-05,
"loss": 0.8115,
"step": 1690
},
{
"epoch": 2.59,
"grad_norm": 0.9439251135887912,
"learning_rate": 1.1289801857188587e-05,
"loss": 0.8584,
"step": 1695
},
{
"epoch": 2.59,
"grad_norm": 0.9292081088950934,
"learning_rate": 1.0883219121918265e-05,
"loss": 0.8164,
"step": 1700
},
{
"epoch": 2.6,
"grad_norm": 0.9421757462427118,
"learning_rate": 1.0483670864493778e-05,
"loss": 0.795,
"step": 1705
},
{
"epoch": 2.61,
"grad_norm": 0.999919346225195,
"learning_rate": 1.0091188623457415e-05,
"loss": 0.8184,
"step": 1710
},
{
"epoch": 2.62,
"grad_norm": 0.9742832828955831,
"learning_rate": 9.70580337959196e-06,
"loss": 0.8299,
"step": 1715
},
{
"epoch": 2.62,
"grad_norm": 0.9089507431330445,
"learning_rate": 9.327545553475114e-06,
"loss": 0.7723,
"step": 1720
},
{
"epoch": 2.63,
"grad_norm": 0.9732825474788489,
"learning_rate": 8.95644500307835e-06,
"loss": 0.8501,
"step": 1725
},
{
"epoch": 2.64,
"grad_norm": 0.9960143529194506,
"learning_rate": 8.592531021409945e-06,
"loss": 0.9096,
"step": 1730
},
{
"epoch": 2.65,
"grad_norm": 0.9339895800344439,
"learning_rate": 8.235832334202776e-06,
"loss": 0.8828,
"step": 1735
},
{
"epoch": 2.65,
"grad_norm": 1.056692915239154,
"learning_rate": 7.886377097646792e-06,
"loss": 0.8749,
"step": 1740
},
{
"epoch": 2.66,
"grad_norm": 0.9711524416206577,
"learning_rate": 7.544192896166568e-06,
"loss": 0.815,
"step": 1745
},
{
"epoch": 2.67,
"grad_norm": 1.0985213550852182,
"learning_rate": 7.209306740243893e-06,
"loss": 0.8256,
"step": 1750
},
{
"epoch": 2.68,
"grad_norm": 1.0441483688013498,
"learning_rate": 6.881745064285617e-06,
"loss": 0.8083,
"step": 1755
},
{
"epoch": 2.68,
"grad_norm": 0.9998637235645016,
"learning_rate": 6.561533724537095e-06,
"loss": 0.8116,
"step": 1760
},
{
"epoch": 2.69,
"grad_norm": 1.1263808784059275,
"learning_rate": 6.248697997041219e-06,
"loss": 0.811,
"step": 1765
},
{
"epoch": 2.7,
"grad_norm": 0.9529944445283575,
"learning_rate": 5.943262575643238e-06,
"loss": 0.8225,
"step": 1770
},
{
"epoch": 2.71,
"grad_norm": 0.9459961315540705,
"learning_rate": 5.645251570041488e-06,
"loss": 0.7937,
"step": 1775
},
{
"epoch": 2.72,
"grad_norm": 0.9556679665444172,
"learning_rate": 5.354688503884342e-06,
"loss": 0.8109,
"step": 1780
},
{
"epoch": 2.72,
"grad_norm": 1.162754539840243,
"learning_rate": 5.071596312913329e-06,
"loss": 0.8337,
"step": 1785
},
{
"epoch": 2.73,
"grad_norm": 0.987118403127622,
"learning_rate": 4.79599734315267e-06,
"loss": 0.8208,
"step": 1790
},
{
"epoch": 2.74,
"grad_norm": 1.0185381683595645,
"learning_rate": 4.527913349145441e-06,
"loss": 0.7854,
"step": 1795
},
{
"epoch": 2.75,
"grad_norm": 1.036855323975471,
"learning_rate": 4.267365492236319e-06,
"loss": 0.8138,
"step": 1800
},
{
"epoch": 2.75,
"grad_norm": 0.906565660746568,
"learning_rate": 4.014374338901206e-06,
"loss": 0.782,
"step": 1805
},
{
"epoch": 2.76,
"grad_norm": 0.9770509541103689,
"learning_rate": 3.768959859123822e-06,
"loss": 0.8191,
"step": 1810
},
{
"epoch": 2.77,
"grad_norm": 1.0486391035448164,
"learning_rate": 3.531141424819373e-06,
"loss": 0.8603,
"step": 1815
},
{
"epoch": 2.78,
"grad_norm": 1.0773318638652334,
"learning_rate": 3.300937808305371e-06,
"loss": 0.8521,
"step": 1820
},
{
"epoch": 2.78,
"grad_norm": 0.985804357345318,
"learning_rate": 3.0783671808198635e-06,
"loss": 0.8141,
"step": 1825
},
{
"epoch": 2.79,
"grad_norm": 1.1529169013999445,
"learning_rate": 2.8634471110871032e-06,
"loss": 0.7832,
"step": 1830
},
{
"epoch": 2.8,
"grad_norm": 0.9031631771021017,
"learning_rate": 2.656194563930714e-06,
"loss": 0.8585,
"step": 1835
},
{
"epoch": 2.81,
"grad_norm": 0.8637751952649739,
"learning_rate": 2.4566258989345726e-06,
"loss": 0.7996,
"step": 1840
},
{
"epoch": 2.81,
"grad_norm": 1.0164681340381803,
"learning_rate": 2.264756869151441e-06,
"loss": 0.8044,
"step": 1845
},
{
"epoch": 2.82,
"grad_norm": 1.0556120127779176,
"learning_rate": 2.0806026198595173e-06,
"loss": 0.7943,
"step": 1850
},
{
"epoch": 2.83,
"grad_norm": 1.1048150934514633,
"learning_rate": 1.9041776873669704e-06,
"loss": 0.8225,
"step": 1855
},
{
"epoch": 2.84,
"grad_norm": 0.9507298133478834,
"learning_rate": 1.7354959978644225e-06,
"loss": 0.8107,
"step": 1860
},
{
"epoch": 2.85,
"grad_norm": 1.1226388926103825,
"learning_rate": 1.5745708663257197e-06,
"loss": 0.8282,
"step": 1865
},
{
"epoch": 2.85,
"grad_norm": 0.929914623001807,
"learning_rate": 1.4214149954569268e-06,
"loss": 0.7743,
"step": 1870
},
{
"epoch": 2.86,
"grad_norm": 0.8451615970437887,
"learning_rate": 1.2760404746936183e-06,
"loss": 0.799,
"step": 1875
},
{
"epoch": 2.87,
"grad_norm": 0.9722281685095321,
"learning_rate": 1.1384587792465872e-06,
"loss": 0.8559,
"step": 1880
},
{
"epoch": 2.88,
"grad_norm": 0.9340387136482731,
"learning_rate": 1.0086807691960244e-06,
"loss": 0.822,
"step": 1885
},
{
"epoch": 2.88,
"grad_norm": 0.9046388002266489,
"learning_rate": 8.867166886343592e-07,
"loss": 0.8384,
"step": 1890
},
{
"epoch": 2.89,
"grad_norm": 0.9601062532338479,
"learning_rate": 7.725761648574969e-07,
"loss": 0.8342,
"step": 1895
},
{
"epoch": 2.9,
"grad_norm": 0.99460636173352,
"learning_rate": 6.662682076050031e-07,
"loss": 0.8721,
"step": 1900
},
{
"epoch": 2.91,
"grad_norm": 0.9721186202775147,
"learning_rate": 5.678012083488727e-07,
"loss": 0.8559,
"step": 1905
},
{
"epoch": 2.91,
"grad_norm": 1.2579065537174423,
"learning_rate": 4.771829396311267e-07,
"loss": 0.7897,
"step": 1910
},
{
"epoch": 2.92,
"grad_norm": 1.011827783498998,
"learning_rate": 3.9442055445029214e-07,
"loss": 0.7987,
"step": 1915
},
{
"epoch": 2.93,
"grad_norm": 1.133610710351972,
"learning_rate": 3.1952058569681975e-07,
"loss": 0.8083,
"step": 1920
},
{
"epoch": 2.94,
"grad_norm": 0.9463613366312343,
"learning_rate": 2.5248894563735247e-07,
"loss": 0.848,
"step": 1925
},
{
"epoch": 2.94,
"grad_norm": 1.0867623661570065,
"learning_rate": 1.933309254480542e-07,
"loss": 0.83,
"step": 1930
},
{
"epoch": 2.95,
"grad_norm": 0.8980549292174614,
"learning_rate": 1.4205119479693274e-07,
"loss": 0.8238,
"step": 1935
},
{
"epoch": 2.96,
"grad_norm": 0.8870015504798887,
"learning_rate": 9.865380147530134e-08,
"loss": 0.8323,
"step": 1940
},
{
"epoch": 2.97,
"grad_norm": 1.1236359515597645,
"learning_rate": 6.314217107817877e-08,
"loss": 0.8563,
"step": 1945
},
{
"epoch": 2.97,
"grad_norm": 1.0317942271237952,
"learning_rate": 3.551910673394998e-08,
"loss": 0.851,
"step": 1950
},
{
"epoch": 2.98,
"grad_norm": 1.0177460190731211,
"learning_rate": 1.5786788883065394e-08,
"loss": 0.8421,
"step": 1955
},
{
"epoch": 2.99,
"grad_norm": 1.0956132894984734,
"learning_rate": 3.9467751059341085e-09,
"loss": 0.8274,
"step": 1960
},
{
"epoch": 3.0,
"grad_norm": 1.2600799171859316,
"learning_rate": 0.0,
"loss": 0.8345,
"step": 1965
},
{
"epoch": 3.0,
"eval_loss": 1.2143404483795166,
"eval_runtime": 248.6803,
"eval_samples_per_second": 9.289,
"eval_steps_per_second": 0.583,
"step": 1965
},
{
"epoch": 3.0,
"step": 1965,
"total_flos": 4085393396662272.0,
"train_loss": 1.1282114190303036,
"train_runtime": 21563.1342,
"train_samples_per_second": 2.918,
"train_steps_per_second": 0.091
}
],
"logging_steps": 5,
"max_steps": 1965,
"num_input_tokens_seen": 0,
"num_train_epochs": 3,
"save_steps": 500,
"total_flos": 4085393396662272.0,
"train_batch_size": 4,
"trial_name": null,
"trial_params": null
}