{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9946332737030411,
  "eval_steps": 500,
  "global_step": 2780,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0035778175313059034,
      "grad_norm": 0.5825825929641724,
      "learning_rate": 0.00019928443649373882,
      "loss": 1.248,
      "num_input_tokens_seen": 6646,
      "step": 10
    },
    {
      "epoch": 0.007155635062611807,
      "grad_norm": 0.5380188822746277,
      "learning_rate": 0.00019856887298747765,
      "loss": 0.5478,
      "num_input_tokens_seen": 13063,
      "step": 20
    },
    {
      "epoch": 0.01073345259391771,
      "grad_norm": 0.3872911036014557,
      "learning_rate": 0.00019785330948121648,
      "loss": 0.5135,
      "num_input_tokens_seen": 19512,
      "step": 30
    },
    {
      "epoch": 0.014311270125223614,
      "grad_norm": 0.4991438686847687,
      "learning_rate": 0.0001971377459749553,
      "loss": 0.5092,
      "num_input_tokens_seen": 26884,
      "step": 40
    },
    {
      "epoch": 0.017889087656529516,
      "grad_norm": 0.6744784116744995,
      "learning_rate": 0.0001964221824686941,
      "loss": 0.4799,
      "num_input_tokens_seen": 34831,
      "step": 50
    },
    {
      "epoch": 0.02146690518783542,
      "grad_norm": 0.5413841009140015,
      "learning_rate": 0.00019570661896243293,
      "loss": 0.4738,
      "num_input_tokens_seen": 40074,
      "step": 60
    },
    {
      "epoch": 0.025044722719141325,
      "grad_norm": 0.33517029881477356,
      "learning_rate": 0.00019499105545617174,
      "loss": 0.4907,
      "num_input_tokens_seen": 47194,
      "step": 70
    },
    {
      "epoch": 0.028622540250447227,
      "grad_norm": 0.34275758266448975,
      "learning_rate": 0.00019427549194991057,
      "loss": 0.4642,
      "num_input_tokens_seen": 53439,
      "step": 80
    },
    {
      "epoch": 0.03220035778175313,
      "grad_norm": 0.4074145257472992,
      "learning_rate": 0.00019355992844364938,
      "loss": 0.4431,
      "num_input_tokens_seen": 59366,
      "step": 90
    },
    {
      "epoch": 0.03577817531305903,
      "grad_norm": 0.372760534286499,
      "learning_rate": 0.0001928443649373882,
      "loss": 0.4824,
      "num_input_tokens_seen": 66414,
      "step": 100
    },
    {
      "epoch": 0.03935599284436494,
      "grad_norm": 0.35169002413749695,
      "learning_rate": 0.00019212880143112702,
      "loss": 0.4863,
      "num_input_tokens_seen": 73451,
      "step": 110
    },
    {
      "epoch": 0.04293381037567084,
      "grad_norm": 0.4088020324707031,
      "learning_rate": 0.00019141323792486585,
      "loss": 0.4792,
      "num_input_tokens_seen": 81934,
      "step": 120
    },
    {
      "epoch": 0.046511627906976744,
      "grad_norm": 0.40062326192855835,
      "learning_rate": 0.00019069767441860466,
      "loss": 0.4607,
      "num_input_tokens_seen": 88335,
      "step": 130
    },
    {
      "epoch": 0.05008944543828265,
      "grad_norm": 0.5044320225715637,
      "learning_rate": 0.0001899821109123435,
      "loss": 0.456,
      "num_input_tokens_seen": 96192,
      "step": 140
    },
    {
      "epoch": 0.05366726296958855,
      "grad_norm": 0.4566495418548584,
      "learning_rate": 0.0001892665474060823,
      "loss": 0.429,
      "num_input_tokens_seen": 101609,
      "step": 150
    },
    {
      "epoch": 0.057245080500894455,
      "grad_norm": 0.4657338559627533,
      "learning_rate": 0.0001885509838998211,
      "loss": 0.4445,
      "num_input_tokens_seen": 107467,
      "step": 160
    },
    {
      "epoch": 0.06082289803220036,
      "grad_norm": 0.5721924304962158,
      "learning_rate": 0.00018783542039355994,
      "loss": 0.4304,
      "num_input_tokens_seen": 113612,
      "step": 170
    },
    {
      "epoch": 0.06440071556350627,
      "grad_norm": 0.2883516848087311,
      "learning_rate": 0.00018711985688729877,
      "loss": 0.4525,
      "num_input_tokens_seen": 121416,
      "step": 180
    },
    {
      "epoch": 0.06797853309481217,
      "grad_norm": 0.5061659216880798,
      "learning_rate": 0.00018640429338103758,
      "loss": 0.4439,
      "num_input_tokens_seen": 128284,
      "step": 190
    },
    {
      "epoch": 0.07155635062611806,
      "grad_norm": 0.3323754072189331,
      "learning_rate": 0.00018568872987477638,
      "loss": 0.4489,
      "num_input_tokens_seen": 135571,
      "step": 200
    },
    {
      "epoch": 0.07513416815742398,
      "grad_norm": 0.5354058742523193,
      "learning_rate": 0.00018497316636851522,
      "loss": 0.4634,
      "num_input_tokens_seen": 141479,
      "step": 210
    },
    {
      "epoch": 0.07871198568872988,
      "grad_norm": 0.4328760504722595,
      "learning_rate": 0.00018425760286225405,
      "loss": 0.4545,
      "num_input_tokens_seen": 147677,
      "step": 220
    },
    {
      "epoch": 0.08228980322003578,
      "grad_norm": 0.28675127029418945,
      "learning_rate": 0.00018354203935599286,
      "loss": 0.4814,
      "num_input_tokens_seen": 154847,
      "step": 230
    },
    {
      "epoch": 0.08586762075134168,
      "grad_norm": 0.31572216749191284,
      "learning_rate": 0.00018282647584973166,
      "loss": 0.446,
      "num_input_tokens_seen": 162267,
      "step": 240
    },
    {
      "epoch": 0.08944543828264759,
      "grad_norm": 0.360166996717453,
      "learning_rate": 0.0001821109123434705,
      "loss": 0.4549,
      "num_input_tokens_seen": 168817,
      "step": 250
    },
    {
      "epoch": 0.09302325581395349,
      "grad_norm": 0.342385470867157,
      "learning_rate": 0.0001813953488372093,
      "loss": 0.4297,
      "num_input_tokens_seen": 174828,
      "step": 260
    },
    {
      "epoch": 0.09660107334525939,
      "grad_norm": 0.37481924891471863,
      "learning_rate": 0.00018067978533094814,
      "loss": 0.4314,
      "num_input_tokens_seen": 181578,
      "step": 270
    },
    {
      "epoch": 0.1001788908765653,
      "grad_norm": 0.28545519709587097,
      "learning_rate": 0.00017996422182468694,
      "loss": 0.4332,
      "num_input_tokens_seen": 187842,
      "step": 280
    },
    {
      "epoch": 0.1037567084078712,
      "grad_norm": 0.38877248764038086,
      "learning_rate": 0.00017924865831842578,
      "loss": 0.4248,
      "num_input_tokens_seen": 194651,
      "step": 290
    },
    {
      "epoch": 0.1073345259391771,
      "grad_norm": 0.30087631940841675,
      "learning_rate": 0.00017853309481216458,
      "loss": 0.4405,
      "num_input_tokens_seen": 202203,
      "step": 300
    },
    {
      "epoch": 0.11091234347048301,
      "grad_norm": 0.33470776677131653,
      "learning_rate": 0.00017781753130590342,
      "loss": 0.4485,
      "num_input_tokens_seen": 210364,
      "step": 310
    },
    {
      "epoch": 0.11449016100178891,
      "grad_norm": 0.44205668568611145,
      "learning_rate": 0.00017710196779964222,
      "loss": 0.4431,
      "num_input_tokens_seen": 217965,
      "step": 320
    },
    {
      "epoch": 0.11806797853309481,
      "grad_norm": 0.39270082116127014,
      "learning_rate": 0.00017638640429338106,
      "loss": 0.4475,
      "num_input_tokens_seen": 224249,
      "step": 330
    },
    {
      "epoch": 0.12164579606440072,
      "grad_norm": 0.37138304114341736,
      "learning_rate": 0.00017567084078711986,
      "loss": 0.4139,
      "num_input_tokens_seen": 230513,
      "step": 340
    },
    {
      "epoch": 0.1252236135957066,
      "grad_norm": 0.32019296288490295,
      "learning_rate": 0.00017495527728085867,
      "loss": 0.4488,
      "num_input_tokens_seen": 237103,
      "step": 350
    },
    {
      "epoch": 0.12880143112701253,
      "grad_norm": 0.42748796939849854,
      "learning_rate": 0.0001742397137745975,
      "loss": 0.4059,
      "num_input_tokens_seen": 242530,
      "step": 360
    },
    {
      "epoch": 0.13237924865831843,
      "grad_norm": 0.39997342228889465,
      "learning_rate": 0.00017352415026833634,
      "loss": 0.4069,
      "num_input_tokens_seen": 247524,
      "step": 370
    },
    {
      "epoch": 0.13595706618962433,
      "grad_norm": 0.45402902364730835,
      "learning_rate": 0.00017280858676207514,
      "loss": 0.4354,
      "num_input_tokens_seen": 252526,
      "step": 380
    },
    {
      "epoch": 0.13953488372093023,
      "grad_norm": 0.500701904296875,
      "learning_rate": 0.00017209302325581395,
      "loss": 0.4421,
      "num_input_tokens_seen": 258974,
      "step": 390
    },
    {
      "epoch": 0.14311270125223613,
      "grad_norm": 0.2604714035987854,
      "learning_rate": 0.00017137745974955278,
      "loss": 0.4338,
      "num_input_tokens_seen": 265411,
      "step": 400
    },
    {
      "epoch": 0.14669051878354203,
      "grad_norm": 0.5467566251754761,
      "learning_rate": 0.00017066189624329162,
      "loss": 0.4336,
      "num_input_tokens_seen": 271069,
      "step": 410
    },
    {
      "epoch": 0.15026833631484796,
      "grad_norm": 0.4278429448604584,
      "learning_rate": 0.00016994633273703042,
      "loss": 0.4384,
      "num_input_tokens_seen": 277913,
      "step": 420
    },
    {
      "epoch": 0.15384615384615385,
      "grad_norm": 0.38349634408950806,
      "learning_rate": 0.00016923076923076923,
      "loss": 0.445,
      "num_input_tokens_seen": 285726,
      "step": 430
    },
    {
      "epoch": 0.15742397137745975,
      "grad_norm": 0.298753559589386,
      "learning_rate": 0.00016851520572450806,
      "loss": 0.411,
      "num_input_tokens_seen": 293886,
      "step": 440
    },
    {
      "epoch": 0.16100178890876565,
      "grad_norm": 0.4590730667114258,
      "learning_rate": 0.00016779964221824687,
      "loss": 0.4441,
      "num_input_tokens_seen": 300933,
      "step": 450
    },
    {
      "epoch": 0.16457960644007155,
      "grad_norm": 0.25614652037620544,
      "learning_rate": 0.0001670840787119857,
      "loss": 0.4201,
      "num_input_tokens_seen": 306782,
      "step": 460
    },
    {
      "epoch": 0.16815742397137745,
      "grad_norm": 0.3007524013519287,
      "learning_rate": 0.0001663685152057245,
      "loss": 0.444,
      "num_input_tokens_seen": 314243,
      "step": 470
    },
    {
      "epoch": 0.17173524150268335,
      "grad_norm": 0.4292968809604645,
      "learning_rate": 0.00016565295169946334,
      "loss": 0.4089,
      "num_input_tokens_seen": 319432,
      "step": 480
    },
    {
      "epoch": 0.17531305903398928,
      "grad_norm": 0.31139102578163147,
      "learning_rate": 0.00016493738819320215,
      "loss": 0.4244,
      "num_input_tokens_seen": 325211,
      "step": 490
    },
    {
      "epoch": 0.17889087656529518,
      "grad_norm": 0.34276363253593445,
      "learning_rate": 0.00016422182468694098,
      "loss": 0.4133,
      "num_input_tokens_seen": 332100,
      "step": 500
    },
    {
      "epoch": 0.18246869409660108,
      "grad_norm": 0.39435234665870667,
      "learning_rate": 0.0001635062611806798,
      "loss": 0.4428,
      "num_input_tokens_seen": 337848,
      "step": 510
    },
    {
      "epoch": 0.18604651162790697,
      "grad_norm": 0.2763209640979767,
      "learning_rate": 0.00016279069767441862,
      "loss": 0.4464,
      "num_input_tokens_seen": 345759,
      "step": 520
    },
    {
      "epoch": 0.18962432915921287,
      "grad_norm": 0.2753828763961792,
      "learning_rate": 0.00016207513416815743,
      "loss": 0.438,
      "num_input_tokens_seen": 354078,
      "step": 530
    },
    {
      "epoch": 0.19320214669051877,
      "grad_norm": 0.3228139579296112,
      "learning_rate": 0.00016135957066189623,
      "loss": 0.4314,
      "num_input_tokens_seen": 360961,
      "step": 540
    },
    {
      "epoch": 0.1967799642218247,
      "grad_norm": 0.30050089955329895,
      "learning_rate": 0.00016064400715563507,
      "loss": 0.4364,
      "num_input_tokens_seen": 367083,
      "step": 550
    },
    {
      "epoch": 0.2003577817531306,
      "grad_norm": 0.3418981432914734,
      "learning_rate": 0.0001599284436493739,
      "loss": 0.4329,
      "num_input_tokens_seen": 373196,
      "step": 560
    },
    {
      "epoch": 0.2039355992844365,
      "grad_norm": 0.36333030462265015,
      "learning_rate": 0.0001592128801431127,
      "loss": 0.4482,
      "num_input_tokens_seen": 380449,
      "step": 570
    },
    {
      "epoch": 0.2075134168157424,
      "grad_norm": 0.2979726195335388,
      "learning_rate": 0.00015849731663685151,
      "loss": 0.4258,
      "num_input_tokens_seen": 386577,
      "step": 580
    },
    {
      "epoch": 0.2110912343470483,
      "grad_norm": 0.2969113886356354,
      "learning_rate": 0.00015778175313059035,
      "loss": 0.433,
      "num_input_tokens_seen": 392953,
      "step": 590
    },
    {
      "epoch": 0.2146690518783542,
      "grad_norm": 0.4132014811038971,
      "learning_rate": 0.00015706618962432918,
      "loss": 0.4148,
      "num_input_tokens_seen": 399368,
      "step": 600
    },
    {
      "epoch": 0.2182468694096601,
      "grad_norm": 0.35863760113716125,
      "learning_rate": 0.000156350626118068,
      "loss": 0.4105,
      "num_input_tokens_seen": 407343,
      "step": 610
    },
    {
      "epoch": 0.22182468694096602,
      "grad_norm": 0.287056028842926,
      "learning_rate": 0.0001556350626118068,
      "loss": 0.4495,
      "num_input_tokens_seen": 413867,
      "step": 620
    },
    {
      "epoch": 0.22540250447227192,
      "grad_norm": 0.41710999608039856,
      "learning_rate": 0.00015491949910554563,
      "loss": 0.426,
      "num_input_tokens_seen": 422712,
      "step": 630
    },
    {
      "epoch": 0.22898032200357782,
      "grad_norm": 0.42847341299057007,
      "learning_rate": 0.00015420393559928446,
      "loss": 0.4163,
      "num_input_tokens_seen": 428523,
      "step": 640
    },
    {
      "epoch": 0.23255813953488372,
      "grad_norm": 0.35523882508277893,
      "learning_rate": 0.00015348837209302327,
      "loss": 0.4237,
      "num_input_tokens_seen": 436283,
      "step": 650
    },
    {
      "epoch": 0.23613595706618962,
      "grad_norm": 0.32238948345184326,
      "learning_rate": 0.00015277280858676207,
      "loss": 0.4527,
      "num_input_tokens_seen": 444368,
      "step": 660
    },
    {
      "epoch": 0.23971377459749552,
      "grad_norm": 0.3075689375400543,
      "learning_rate": 0.0001520572450805009,
      "loss": 0.4057,
      "num_input_tokens_seen": 449919,
      "step": 670
    },
    {
      "epoch": 0.24329159212880144,
      "grad_norm": 0.3819845914840698,
      "learning_rate": 0.00015134168157423971,
      "loss": 0.3951,
      "num_input_tokens_seen": 455612,
      "step": 680
    },
    {
      "epoch": 0.24686940966010734,
      "grad_norm": 0.3687816560268402,
      "learning_rate": 0.00015062611806797855,
      "loss": 0.4176,
      "num_input_tokens_seen": 461846,
      "step": 690
    },
    {
      "epoch": 0.2504472271914132,
      "grad_norm": 0.32070156931877136,
      "learning_rate": 0.00014991055456171735,
      "loss": 0.4111,
      "num_input_tokens_seen": 468333,
      "step": 700
    },
    {
      "epoch": 0.25402504472271914,
      "grad_norm": 0.35491397976875305,
      "learning_rate": 0.0001491949910554562,
      "loss": 0.4215,
      "num_input_tokens_seen": 476220,
      "step": 710
    },
    {
      "epoch": 0.25760286225402507,
      "grad_norm": 0.3560994863510132,
      "learning_rate": 0.000148479427549195,
      "loss": 0.4011,
      "num_input_tokens_seen": 481867,
      "step": 720
    },
    {
      "epoch": 0.26118067978533094,
      "grad_norm": 0.3188275992870331,
      "learning_rate": 0.0001477638640429338,
      "loss": 0.4372,
      "num_input_tokens_seen": 489088,
      "step": 730
    },
    {
      "epoch": 0.26475849731663686,
      "grad_norm": 0.3422275483608246,
      "learning_rate": 0.00014704830053667263,
      "loss": 0.4178,
      "num_input_tokens_seen": 494747,
      "step": 740
    },
    {
      "epoch": 0.26833631484794274,
      "grad_norm": 0.29682761430740356,
      "learning_rate": 0.00014633273703041147,
      "loss": 0.4309,
      "num_input_tokens_seen": 501076,
      "step": 750
    },
    {
      "epoch": 0.27191413237924866,
      "grad_norm": 0.44356122612953186,
      "learning_rate": 0.00014561717352415027,
      "loss": 0.4113,
      "num_input_tokens_seen": 508236,
      "step": 760
    },
    {
      "epoch": 0.27549194991055453,
      "grad_norm": 0.4110952317714691,
      "learning_rate": 0.00014490161001788908,
      "loss": 0.4232,
      "num_input_tokens_seen": 514982,
      "step": 770
    },
    {
      "epoch": 0.27906976744186046,
      "grad_norm": 0.3558269441127777,
      "learning_rate": 0.00014418604651162791,
      "loss": 0.4076,
      "num_input_tokens_seen": 521248,
      "step": 780
    },
    {
      "epoch": 0.2826475849731664,
      "grad_norm": 0.3435330092906952,
      "learning_rate": 0.00014347048300536675,
      "loss": 0.4109,
      "num_input_tokens_seen": 527954,
      "step": 790
    },
    {
      "epoch": 0.28622540250447226,
      "grad_norm": 0.2566317021846771,
      "learning_rate": 0.00014275491949910555,
      "loss": 0.4144,
      "num_input_tokens_seen": 535308,
      "step": 800
    },
    {
      "epoch": 0.2898032200357782,
      "grad_norm": 0.24462071061134338,
      "learning_rate": 0.00014203935599284436,
      "loss": 0.3984,
      "num_input_tokens_seen": 542297,
      "step": 810
    },
    {
      "epoch": 0.29338103756708406,
      "grad_norm": 0.4308456480503082,
      "learning_rate": 0.0001413237924865832,
      "loss": 0.4142,
      "num_input_tokens_seen": 548183,
      "step": 820
    },
    {
      "epoch": 0.29695885509839,
      "grad_norm": 0.4515162408351898,
      "learning_rate": 0.00014060822898032203,
      "loss": 0.404,
      "num_input_tokens_seen": 553861,
      "step": 830
    },
    {
      "epoch": 0.3005366726296959,
      "grad_norm": 0.3883764147758484,
      "learning_rate": 0.00013989266547406083,
      "loss": 0.4159,
      "num_input_tokens_seen": 559943,
      "step": 840
    },
    {
      "epoch": 0.3041144901610018,
      "grad_norm": 0.2981751263141632,
      "learning_rate": 0.00013917710196779964,
      "loss": 0.4099,
      "num_input_tokens_seen": 566038,
      "step": 850
    },
    {
      "epoch": 0.3076923076923077,
      "grad_norm": 0.42181873321533203,
      "learning_rate": 0.00013846153846153847,
      "loss": 0.4365,
      "num_input_tokens_seen": 574126,
      "step": 860
    },
    {
      "epoch": 0.3112701252236136,
      "grad_norm": 0.31020838022232056,
      "learning_rate": 0.00013774597495527728,
      "loss": 0.4076,
      "num_input_tokens_seen": 581039,
      "step": 870
    },
    {
      "epoch": 0.3148479427549195,
      "grad_norm": 0.2839311361312866,
      "learning_rate": 0.00013703041144901611,
      "loss": 0.4139,
      "num_input_tokens_seen": 587384,
      "step": 880
    },
    {
      "epoch": 0.3184257602862254,
      "grad_norm": 0.30183765292167664,
      "learning_rate": 0.00013631484794275492,
      "loss": 0.4412,
      "num_input_tokens_seen": 594626,
      "step": 890
    },
    {
      "epoch": 0.3220035778175313,
      "grad_norm": 0.24405309557914734,
      "learning_rate": 0.00013559928443649375,
      "loss": 0.4217,
      "num_input_tokens_seen": 601425,
      "step": 900
    },
    {
      "epoch": 0.32558139534883723,
      "grad_norm": 0.3325401544570923,
      "learning_rate": 0.00013488372093023256,
      "loss": 0.399,
      "num_input_tokens_seen": 607033,
      "step": 910
    },
    {
      "epoch": 0.3291592128801431,
      "grad_norm": 0.4284108877182007,
      "learning_rate": 0.00013416815742397137,
      "loss": 0.4329,
      "num_input_tokens_seen": 612959,
      "step": 920
    },
    {
      "epoch": 0.33273703041144903,
      "grad_norm": 0.39697644114494324,
      "learning_rate": 0.0001334525939177102,
      "loss": 0.4062,
      "num_input_tokens_seen": 620181,
      "step": 930
    },
    {
      "epoch": 0.3363148479427549,
      "grad_norm": 0.4615568220615387,
      "learning_rate": 0.00013273703041144903,
      "loss": 0.3929,
      "num_input_tokens_seen": 625029,
      "step": 940
    },
    {
      "epoch": 0.33989266547406083,
      "grad_norm": 0.3364561200141907,
      "learning_rate": 0.00013202146690518784,
      "loss": 0.4217,
      "num_input_tokens_seen": 630571,
      "step": 950
    },
    {
      "epoch": 0.3434704830053667,
      "grad_norm": 0.3077852129936218,
      "learning_rate": 0.00013130590339892665,
      "loss": 0.4223,
      "num_input_tokens_seen": 636343,
      "step": 960
    },
    {
      "epoch": 0.3470483005366726,
      "grad_norm": 0.4563674032688141,
      "learning_rate": 0.00013059033989266548,
      "loss": 0.4093,
      "num_input_tokens_seen": 644448,
      "step": 970
    },
    {
      "epoch": 0.35062611806797855,
      "grad_norm": 0.3391217291355133,
      "learning_rate": 0.00012987477638640431,
      "loss": 0.4146,
      "num_input_tokens_seen": 651211,
      "step": 980
    },
    {
      "epoch": 0.3542039355992844,
      "grad_norm": 0.2930082082748413,
      "learning_rate": 0.00012915921288014312,
      "loss": 0.4248,
      "num_input_tokens_seen": 658984,
      "step": 990
    },
    {
      "epoch": 0.35778175313059035,
      "grad_norm": 0.3553549647331238,
      "learning_rate": 0.00012844364937388193,
      "loss": 0.417,
      "num_input_tokens_seen": 666184,
      "step": 1000
    },
    {
      "epoch": 0.3613595706618962,
      "grad_norm": 0.3744185268878937,
      "learning_rate": 0.00012772808586762076,
      "loss": 0.3988,
      "num_input_tokens_seen": 673125,
      "step": 1010
    },
    {
      "epoch": 0.36493738819320215,
      "grad_norm": 0.4000609219074249,
      "learning_rate": 0.0001270125223613596,
      "loss": 0.4301,
      "num_input_tokens_seen": 680863,
      "step": 1020
    },
    {
      "epoch": 0.3685152057245081,
      "grad_norm": 0.27482128143310547,
      "learning_rate": 0.0001262969588550984,
      "loss": 0.4122,
      "num_input_tokens_seen": 688582,
      "step": 1030
    },
    {
      "epoch": 0.37209302325581395,
      "grad_norm": 0.438516765832901,
      "learning_rate": 0.0001255813953488372,
      "loss": 0.4253,
      "num_input_tokens_seen": 695760,
      "step": 1040
    },
    {
      "epoch": 0.3756708407871199,
      "grad_norm": 0.26929447054862976,
      "learning_rate": 0.00012486583184257604,
      "loss": 0.405,
      "num_input_tokens_seen": 702561,
      "step": 1050
    },
    {
      "epoch": 0.37924865831842575,
      "grad_norm": 0.24823708832263947,
      "learning_rate": 0.00012415026833631485,
      "loss": 0.4317,
      "num_input_tokens_seen": 710659,
      "step": 1060
    },
    {
      "epoch": 0.3828264758497317,
      "grad_norm": 0.31499966979026794,
      "learning_rate": 0.00012343470483005368,
      "loss": 0.4357,
      "num_input_tokens_seen": 717932,
      "step": 1070
    },
    {
      "epoch": 0.38640429338103754,
      "grad_norm": 0.4409402012825012,
      "learning_rate": 0.0001227191413237925,
      "loss": 0.408,
      "num_input_tokens_seen": 724716,
      "step": 1080
    },
    {
      "epoch": 0.38998211091234347,
      "grad_norm": 0.35407742857933044,
      "learning_rate": 0.00012200357781753131,
      "loss": 0.4088,
      "num_input_tokens_seen": 730694,
      "step": 1090
    },
    {
      "epoch": 0.3935599284436494,
      "grad_norm": 0.36748751997947693,
      "learning_rate": 0.00012128801431127013,
      "loss": 0.4208,
      "num_input_tokens_seen": 739112,
      "step": 1100
    },
    {
      "epoch": 0.39713774597495527,
      "grad_norm": 0.3873242437839508,
      "learning_rate": 0.00012057245080500895,
      "loss": 0.3798,
      "num_input_tokens_seen": 744478,
      "step": 1110
    },
    {
      "epoch": 0.4007155635062612,
      "grad_norm": 0.31334424018859863,
      "learning_rate": 0.00011985688729874778,
      "loss": 0.4179,
      "num_input_tokens_seen": 753315,
      "step": 1120
    },
    {
      "epoch": 0.40429338103756707,
      "grad_norm": 0.42388609051704407,
      "learning_rate": 0.00011914132379248659,
      "loss": 0.3882,
      "num_input_tokens_seen": 759832,
      "step": 1130
    },
    {
      "epoch": 0.407871198568873,
      "grad_norm": 0.3108317255973816,
      "learning_rate": 0.00011842576028622541,
      "loss": 0.4247,
      "num_input_tokens_seen": 766054,
      "step": 1140
    },
    {
      "epoch": 0.41144901610017887,
      "grad_norm": 0.322694331407547,
      "learning_rate": 0.00011771019677996423,
      "loss": 0.416,
      "num_input_tokens_seen": 772192,
      "step": 1150
    },
    {
      "epoch": 0.4150268336314848,
      "grad_norm": 0.4780764877796173,
      "learning_rate": 0.00011699463327370303,
      "loss": 0.3937,
      "num_input_tokens_seen": 778619,
      "step": 1160
    },
    {
      "epoch": 0.4186046511627907,
      "grad_norm": 0.24456287920475006,
      "learning_rate": 0.00011627906976744187,
      "loss": 0.4264,
      "num_input_tokens_seen": 787085,
      "step": 1170
    },
    {
      "epoch": 0.4221824686940966,
      "grad_norm": 0.2968457341194153,
      "learning_rate": 0.00011556350626118069,
      "loss": 0.4196,
      "num_input_tokens_seen": 794103,
      "step": 1180
    },
    {
      "epoch": 0.4257602862254025,
      "grad_norm": 0.3390592038631439,
      "learning_rate": 0.00011484794275491951,
      "loss": 0.4073,
      "num_input_tokens_seen": 800033,
      "step": 1190
    },
    {
      "epoch": 0.4293381037567084,
      "grad_norm": 0.35398080945014954,
      "learning_rate": 0.00011413237924865831,
      "loss": 0.4032,
      "num_input_tokens_seen": 807607,
      "step": 1200
    },
    {
      "epoch": 0.4329159212880143,
      "grad_norm": 0.25886887311935425,
      "learning_rate": 0.00011341681574239715,
      "loss": 0.416,
      "num_input_tokens_seen": 813956,
      "step": 1210
    },
    {
      "epoch": 0.4364937388193202,
      "grad_norm": 0.336000919342041,
      "learning_rate": 0.00011270125223613597,
      "loss": 0.3817,
      "num_input_tokens_seen": 821166,
      "step": 1220
    },
    {
      "epoch": 0.4400715563506261,
      "grad_norm": 0.36095812916755676,
      "learning_rate": 0.00011198568872987479,
      "loss": 0.4097,
      "num_input_tokens_seen": 829461,
      "step": 1230
    },
    {
      "epoch": 0.44364937388193204,
      "grad_norm": 0.34367650747299194,
      "learning_rate": 0.0001112701252236136,
      "loss": 0.4028,
      "num_input_tokens_seen": 835022,
      "step": 1240
    },
    {
      "epoch": 0.4472271914132379,
      "grad_norm": 0.3959554135799408,
      "learning_rate": 0.00011055456171735241,
      "loss": 0.402,
      "num_input_tokens_seen": 841676,
      "step": 1250
    },
    {
      "epoch": 0.45080500894454384,
      "grad_norm": 0.4300963580608368,
      "learning_rate": 0.00010983899821109125,
      "loss": 0.4169,
      "num_input_tokens_seen": 848238,
      "step": 1260
    },
    {
      "epoch": 0.4543828264758497,
      "grad_norm": 0.36439695954322815,
      "learning_rate": 0.00010912343470483007,
      "loss": 0.4131,
      "num_input_tokens_seen": 854482,
      "step": 1270
    },
    {
      "epoch": 0.45796064400715564,
      "grad_norm": 0.4523600935935974,
      "learning_rate": 0.00010840787119856887,
      "loss": 0.3947,
      "num_input_tokens_seen": 860605,
      "step": 1280
    },
    {
      "epoch": 0.46153846153846156,
      "grad_norm": 0.43886885046958923,
      "learning_rate": 0.0001076923076923077,
      "loss": 0.4053,
      "num_input_tokens_seen": 867693,
      "step": 1290
    },
    {
      "epoch": 0.46511627906976744,
      "grad_norm": 0.33564773201942444,
      "learning_rate": 0.00010697674418604651,
      "loss": 0.3984,
      "num_input_tokens_seen": 873855,
      "step": 1300
    },
    {
      "epoch": 0.46869409660107336,
      "grad_norm": 0.31822851300239563,
      "learning_rate": 0.00010626118067978535,
      "loss": 0.4162,
      "num_input_tokens_seen": 881261,
      "step": 1310
    },
    {
      "epoch": 0.47227191413237923,
      "grad_norm": 0.2756568193435669,
      "learning_rate": 0.00010554561717352415,
      "loss": 0.4092,
      "num_input_tokens_seen": 888281,
      "step": 1320
    },
    {
      "epoch": 0.47584973166368516,
      "grad_norm": 0.31455397605895996,
      "learning_rate": 0.00010483005366726297,
      "loss": 0.3943,
      "num_input_tokens_seen": 895446,
      "step": 1330
    },
    {
      "epoch": 0.47942754919499103,
      "grad_norm": 0.25797346234321594,
      "learning_rate": 0.0001041144901610018,
      "loss": 0.4353,
      "num_input_tokens_seen": 902631,
      "step": 1340
    },
    {
      "epoch": 0.48300536672629696,
      "grad_norm": 0.2746681571006775,
      "learning_rate": 0.0001033989266547406,
      "loss": 0.3863,
      "num_input_tokens_seen": 908599,
      "step": 1350
    },
    {
      "epoch": 0.4865831842576029,
      "grad_norm": 0.3950134813785553,
      "learning_rate": 0.00010268336314847943,
      "loss": 0.4049,
      "num_input_tokens_seen": 915272,
      "step": 1360
    },
    {
      "epoch": 0.49016100178890876,
      "grad_norm": 0.249292254447937,
      "learning_rate": 0.00010196779964221825,
      "loss": 0.4064,
      "num_input_tokens_seen": 921729,
      "step": 1370
    },
    {
      "epoch": 0.4937388193202147,
      "grad_norm": 0.2855275869369507,
      "learning_rate": 0.00010125223613595707,
      "loss": 0.4092,
      "num_input_tokens_seen": 928645,
      "step": 1380
    },
    {
      "epoch": 0.49731663685152055,
      "grad_norm": 0.3198321759700775,
      "learning_rate": 0.00010053667262969588,
      "loss": 0.3814,
      "num_input_tokens_seen": 933801,
      "step": 1390
    },
    {
      "epoch": 0.5008944543828264,
      "grad_norm": 0.3592207133769989,
      "learning_rate": 9.982110912343471e-05,
      "loss": 0.3776,
      "num_input_tokens_seen": 938905,
      "step": 1400
    },
    {
      "epoch": 0.5044722719141324,
      "grad_norm": 0.2043834775686264,
      "learning_rate": 9.910554561717353e-05,
      "loss": 0.4206,
      "num_input_tokens_seen": 947665,
      "step": 1410
    },
    {
      "epoch": 0.5080500894454383,
      "grad_norm": 0.3463163673877716,
      "learning_rate": 9.838998211091235e-05,
      "loss": 0.3953,
      "num_input_tokens_seen": 955261,
      "step": 1420
    },
    {
      "epoch": 0.5116279069767442,
      "grad_norm": 0.3044353723526001,
      "learning_rate": 9.767441860465116e-05,
      "loss": 0.4226,
      "num_input_tokens_seen": 961797,
      "step": 1430
    },
    {
      "epoch": 0.5152057245080501,
      "grad_norm": 0.4201965928077698,
      "learning_rate": 9.695885509838999e-05,
      "loss": 0.4031,
      "num_input_tokens_seen": 967800,
      "step": 1440
    },
    {
      "epoch": 0.518783542039356,
      "grad_norm": 0.4523704946041107,
      "learning_rate": 9.62432915921288e-05,
      "loss": 0.3753,
      "num_input_tokens_seen": 973223,
      "step": 1450
    },
    {
      "epoch": 0.5223613595706619,
      "grad_norm": 0.3966517448425293,
      "learning_rate": 9.552772808586763e-05,
      "loss": 0.4212,
      "num_input_tokens_seen": 981582,
      "step": 1460
    },
    {
      "epoch": 0.5259391771019678,
      "grad_norm": 0.3248649835586548,
      "learning_rate": 9.481216457960644e-05,
      "loss": 0.4052,
      "num_input_tokens_seen": 988838,
      "step": 1470
    },
    {
      "epoch": 0.5295169946332737,
      "grad_norm": 0.23017796874046326,
      "learning_rate": 9.409660107334526e-05,
      "loss": 0.4019,
      "num_input_tokens_seen": 995054,
      "step": 1480
    },
    {
      "epoch": 0.5330948121645797,
      "grad_norm": 0.3645225465297699,
      "learning_rate": 9.338103756708408e-05,
      "loss": 0.4038,
      "num_input_tokens_seen": 1001348,
      "step": 1490
    },
    {
      "epoch": 0.5366726296958855,
      "grad_norm": 0.25708308815956116,
      "learning_rate": 9.26654740608229e-05,
      "loss": 0.4016,
      "num_input_tokens_seen": 1007269,
      "step": 1500
    },
    {
      "epoch": 0.5402504472271914,
      "grad_norm": 0.2978523373603821,
      "learning_rate": 9.194991055456172e-05,
      "loss": 0.4019,
      "num_input_tokens_seen": 1014734,
      "step": 1510
    },
    {
      "epoch": 0.5438282647584973,
      "grad_norm": 0.32369324564933777,
      "learning_rate": 9.123434704830054e-05,
      "loss": 0.3867,
      "num_input_tokens_seen": 1022130,
      "step": 1520
    },
    {
      "epoch": 0.5474060822898033,
      "grad_norm": 0.34458237886428833,
      "learning_rate": 9.051878354203936e-05,
      "loss": 0.4247,
      "num_input_tokens_seen": 1029090,
      "step": 1530
    },
    {
      "epoch": 0.5509838998211091,
      "grad_norm": 0.34909483790397644,
      "learning_rate": 8.980322003577818e-05,
      "loss": 0.4164,
      "num_input_tokens_seen": 1035252,
      "step": 1540
    },
    {
      "epoch": 0.554561717352415,
      "grad_norm": 0.27858543395996094,
      "learning_rate": 8.9087656529517e-05,
      "loss": 0.3976,
      "num_input_tokens_seen": 1041187,
      "step": 1550
    },
    {
      "epoch": 0.5581395348837209,
      "grad_norm": 0.34940874576568604,
      "learning_rate": 8.837209302325582e-05,
      "loss": 0.3889,
      "num_input_tokens_seen": 1047255,
      "step": 1560
    },
    {
      "epoch": 0.5617173524150268,
      "grad_norm": 0.3687174916267395,
      "learning_rate": 8.765652951699464e-05,
      "loss": 0.3794,
      "num_input_tokens_seen": 1054159,
      "step": 1570
    },
    {
      "epoch": 0.5652951699463328,
      "grad_norm": 0.3228056728839874,
      "learning_rate": 8.694096601073346e-05,
      "loss": 0.4095,
      "num_input_tokens_seen": 1060584,
      "step": 1580
    },
    {
      "epoch": 0.5688729874776386,
      "grad_norm": 0.41065359115600586,
      "learning_rate": 8.622540250447228e-05,
      "loss": 0.3741,
      "num_input_tokens_seen": 1066384,
      "step": 1590
    },
    {
      "epoch": 0.5724508050089445,
      "grad_norm": 0.3346722722053528,
      "learning_rate": 8.55098389982111e-05,
      "loss": 0.4068,
      "num_input_tokens_seen": 1072956,
      "step": 1600
    },
    {
      "epoch": 0.5760286225402504,
      "grad_norm": 0.4220598042011261,
      "learning_rate": 8.479427549194992e-05,
      "loss": 0.4272,
      "num_input_tokens_seen": 1079895,
      "step": 1610
    },
    {
      "epoch": 0.5796064400715564,
      "grad_norm": 0.2898394763469696,
      "learning_rate": 8.407871198568873e-05,
      "loss": 0.4022,
      "num_input_tokens_seen": 1087604,
      "step": 1620
    },
    {
      "epoch": 0.5831842576028623,
      "grad_norm": 0.36473724246025085,
      "learning_rate": 8.336314847942756e-05,
      "loss": 0.3857,
      "num_input_tokens_seen": 1094532,
      "step": 1630
    },
    {
      "epoch": 0.5867620751341681,
      "grad_norm": 0.29369673132896423,
      "learning_rate": 8.264758497316637e-05,
      "loss": 0.405,
      "num_input_tokens_seen": 1100973,
      "step": 1640
    },
    {
      "epoch": 0.590339892665474,
      "grad_norm": 0.29171282052993774,
      "learning_rate": 8.19320214669052e-05,
      "loss": 0.4103,
      "num_input_tokens_seen": 1107805,
      "step": 1650
    },
    {
      "epoch": 0.59391771019678,
      "grad_norm": 0.24154767394065857,
      "learning_rate": 8.1216457960644e-05,
      "loss": 0.3903,
      "num_input_tokens_seen": 1115112,
      "step": 1660
    },
    {
      "epoch": 0.5974955277280859,
      "grad_norm": 0.39852896332740784,
      "learning_rate": 8.050089445438284e-05,
      "loss": 0.4128,
      "num_input_tokens_seen": 1121451,
      "step": 1670
    },
    {
      "epoch": 0.6010733452593918,
      "grad_norm": 0.37065914273262024,
      "learning_rate": 7.978533094812165e-05,
      "loss": 0.4015,
      "num_input_tokens_seen": 1129374,
      "step": 1680
    },
    {
      "epoch": 0.6046511627906976,
      "grad_norm": 0.25734543800354004,
      "learning_rate": 7.906976744186047e-05,
      "loss": 0.3982,
      "num_input_tokens_seen": 1136582,
      "step": 1690
    },
    {
      "epoch": 0.6082289803220036,
      "grad_norm": 0.263864666223526,
      "learning_rate": 7.835420393559929e-05,
      "loss": 0.414,
      "num_input_tokens_seen": 1144688,
      "step": 1700
    },
    {
      "epoch": 0.6118067978533095,
      "grad_norm": 0.4987502098083496,
      "learning_rate": 7.76386404293381e-05,
      "loss": 0.4002,
      "num_input_tokens_seen": 1151513,
      "step": 1710
    },
    {
      "epoch": 0.6153846153846154,
      "grad_norm": 0.35515296459198,
      "learning_rate": 7.692307692307693e-05,
      "loss": 0.399,
      "num_input_tokens_seen": 1159983,
      "step": 1720
    },
    {
      "epoch": 0.6189624329159212,
      "grad_norm": 0.2761338949203491,
      "learning_rate": 7.620751341681575e-05,
      "loss": 0.3954,
      "num_input_tokens_seen": 1168489,
      "step": 1730
    },
    {
      "epoch": 0.6225402504472272,
      "grad_norm": 0.4277167022228241,
      "learning_rate": 7.549194991055457e-05,
      "loss": 0.4031,
      "num_input_tokens_seen": 1175665,
      "step": 1740
    },
    {
      "epoch": 0.6261180679785331,
      "grad_norm": 0.33866697549819946,
      "learning_rate": 7.477638640429339e-05,
      "loss": 0.412,
      "num_input_tokens_seen": 1181930,
      "step": 1750
    },
    {
      "epoch": 0.629695885509839,
      "grad_norm": 0.2600657045841217,
      "learning_rate": 7.40608228980322e-05,
      "loss": 0.3975,
      "num_input_tokens_seen": 1189185,
      "step": 1760
    },
    {
      "epoch": 0.6332737030411449,
      "grad_norm": 0.262329638004303,
      "learning_rate": 7.334525939177103e-05,
      "loss": 0.3841,
      "num_input_tokens_seen": 1196883,
      "step": 1770
    },
    {
      "epoch": 0.6368515205724508,
      "grad_norm": 0.35897818207740784,
      "learning_rate": 7.262969588550985e-05,
      "loss": 0.4004,
      "num_input_tokens_seen": 1205284,
      "step": 1780
    },
    {
      "epoch": 0.6404293381037567,
      "grad_norm": 0.3555593192577362,
      "learning_rate": 7.191413237924867e-05,
      "loss": 0.386,
      "num_input_tokens_seen": 1210701,
      "step": 1790
    },
    {
      "epoch": 0.6440071556350626,
      "grad_norm": 0.3081437349319458,
      "learning_rate": 7.119856887298749e-05,
      "loss": 0.3885,
      "num_input_tokens_seen": 1217841,
      "step": 1800
    },
    {
      "epoch": 0.6475849731663685,
      "grad_norm": 0.43780773878097534,
      "learning_rate": 7.048300536672629e-05,
      "loss": 0.3893,
      "num_input_tokens_seen": 1225095,
      "step": 1810
    },
    {
      "epoch": 0.6511627906976745,
      "grad_norm": 0.381865918636322,
      "learning_rate": 6.976744186046513e-05,
      "loss": 0.3884,
      "num_input_tokens_seen": 1234033,
      "step": 1820
    },
    {
      "epoch": 0.6547406082289803,
      "grad_norm": 0.4573935866355896,
      "learning_rate": 6.905187835420393e-05,
      "loss": 0.4199,
      "num_input_tokens_seen": 1241707,
      "step": 1830
    },
    {
      "epoch": 0.6583184257602862,
      "grad_norm": 0.2445441335439682,
      "learning_rate": 6.833631484794277e-05,
      "loss": 0.3934,
      "num_input_tokens_seen": 1248683,
      "step": 1840
    },
    {
      "epoch": 0.6618962432915921,
      "grad_norm": 0.36599206924438477,
      "learning_rate": 6.762075134168157e-05,
      "loss": 0.3758,
      "num_input_tokens_seen": 1254021,
      "step": 1850
    },
    {
      "epoch": 0.6654740608228981,
      "grad_norm": 0.29867666959762573,
      "learning_rate": 6.69051878354204e-05,
      "loss": 0.3846,
      "num_input_tokens_seen": 1260977,
      "step": 1860
    },
    {
      "epoch": 0.669051878354204,
      "grad_norm": 0.3520377576351166,
      "learning_rate": 6.618962432915921e-05,
      "loss": 0.406,
      "num_input_tokens_seen": 1268015,
      "step": 1870
    },
    {
      "epoch": 0.6726296958855098,
      "grad_norm": 0.48654794692993164,
      "learning_rate": 6.547406082289803e-05,
      "loss": 0.3948,
      "num_input_tokens_seen": 1273256,
      "step": 1880
    },
    {
      "epoch": 0.6762075134168157,
      "grad_norm": 0.3770650327205658,
      "learning_rate": 6.475849731663685e-05,
      "loss": 0.3991,
      "num_input_tokens_seen": 1280311,
      "step": 1890
    },
    {
      "epoch": 0.6797853309481217,
      "grad_norm": 0.3090347945690155,
      "learning_rate": 6.404293381037567e-05,
      "loss": 0.3835,
      "num_input_tokens_seen": 1286733,
      "step": 1900
    },
    {
      "epoch": 0.6833631484794276,
      "grad_norm": 0.299207478761673,
      "learning_rate": 6.332737030411449e-05,
      "loss": 0.3905,
      "num_input_tokens_seen": 1293454,
      "step": 1910
    },
    {
      "epoch": 0.6869409660107334,
      "grad_norm": 0.48610031604766846,
      "learning_rate": 6.261180679785331e-05,
      "loss": 0.374,
      "num_input_tokens_seen": 1299015,
      "step": 1920
    },
    {
      "epoch": 0.6905187835420393,
      "grad_norm": 0.3136894404888153,
      "learning_rate": 6.189624329159213e-05,
      "loss": 0.3952,
      "num_input_tokens_seen": 1305291,
      "step": 1930
    },
    {
      "epoch": 0.6940966010733453,
      "grad_norm": 0.3310141861438751,
      "learning_rate": 6.118067978533095e-05,
      "loss": 0.3968,
      "num_input_tokens_seen": 1312434,
      "step": 1940
    },
    {
      "epoch": 0.6976744186046512,
      "grad_norm": 0.27954959869384766,
      "learning_rate": 6.0465116279069765e-05,
      "loss": 0.4071,
      "num_input_tokens_seen": 1320352,
      "step": 1950
    },
    {
      "epoch": 0.7012522361359571,
      "grad_norm": 0.36177489161491394,
      "learning_rate": 5.974955277280859e-05,
      "loss": 0.3683,
      "num_input_tokens_seen": 1326200,
      "step": 1960
    },
    {
      "epoch": 0.7048300536672629,
      "grad_norm": 0.3898317217826843,
      "learning_rate": 5.9033989266547405e-05,
      "loss": 0.4181,
      "num_input_tokens_seen": 1333316,
      "step": 1970
    },
    {
      "epoch": 0.7084078711985689,
      "grad_norm": 0.2924859821796417,
      "learning_rate": 5.831842576028623e-05,
      "loss": 0.3828,
      "num_input_tokens_seen": 1339225,
      "step": 1980
    },
    {
      "epoch": 0.7119856887298748,
      "grad_norm": 0.3090393543243408,
      "learning_rate": 5.7602862254025045e-05,
      "loss": 0.3932,
      "num_input_tokens_seen": 1346104,
      "step": 1990
    },
    {
      "epoch": 0.7155635062611807,
      "grad_norm": 0.4194253981113434,
      "learning_rate": 5.6887298747763865e-05,
      "loss": 0.4041,
      "num_input_tokens_seen": 1352944,
      "step": 2000
    },
    {
      "epoch": 0.7191413237924866,
      "grad_norm": 0.3060110807418823,
      "learning_rate": 5.6171735241502685e-05,
      "loss": 0.3943,
      "num_input_tokens_seen": 1359671,
      "step": 2010
    },
    {
      "epoch": 0.7227191413237924,
      "grad_norm": 0.42584308981895447,
      "learning_rate": 5.5456171735241505e-05,
      "loss": 0.3982,
      "num_input_tokens_seen": 1366487,
      "step": 2020
    },
    {
      "epoch": 0.7262969588550984,
      "grad_norm": 0.2706156373023987,
      "learning_rate": 5.4740608228980325e-05,
      "loss": 0.3715,
      "num_input_tokens_seen": 1374437,
      "step": 2030
    },
    {
      "epoch": 0.7298747763864043,
      "grad_norm": 0.27796509861946106,
      "learning_rate": 5.4025044722719145e-05,
      "loss": 0.3981,
      "num_input_tokens_seen": 1382038,
      "step": 2040
    },
    {
      "epoch": 0.7334525939177102,
      "grad_norm": 0.37042665481567383,
      "learning_rate": 5.3309481216457965e-05,
      "loss": 0.3949,
      "num_input_tokens_seen": 1388519,
      "step": 2050
    },
    {
      "epoch": 0.7370304114490162,
      "grad_norm": 0.3565654754638672,
      "learning_rate": 5.2593917710196785e-05,
      "loss": 0.3869,
      "num_input_tokens_seen": 1395702,
      "step": 2060
    },
    {
      "epoch": 0.740608228980322,
      "grad_norm": 0.2900319993495941,
      "learning_rate": 5.18783542039356e-05,
      "loss": 0.3963,
      "num_input_tokens_seen": 1402155,
      "step": 2070
    },
    {
      "epoch": 0.7441860465116279,
      "grad_norm": 0.27532869577407837,
      "learning_rate": 5.1162790697674425e-05,
      "loss": 0.3961,
      "num_input_tokens_seen": 1409417,
      "step": 2080
    },
    {
      "epoch": 0.7477638640429338,
      "grad_norm": 0.3994668424129486,
      "learning_rate": 5.044722719141324e-05,
      "loss": 0.417,
      "num_input_tokens_seen": 1416404,
      "step": 2090
    },
    {
      "epoch": 0.7513416815742398,
      "grad_norm": 0.372256338596344,
      "learning_rate": 4.973166368515206e-05,
      "loss": 0.3864,
      "num_input_tokens_seen": 1423004,
      "step": 2100
    },
    {
      "epoch": 0.7549194991055456,
      "grad_norm": 0.32723429799079895,
      "learning_rate": 4.901610017889088e-05,
      "loss": 0.3924,
      "num_input_tokens_seen": 1429991,
      "step": 2110
    },
    {
      "epoch": 0.7584973166368515,
      "grad_norm": 0.37468647956848145,
      "learning_rate": 4.83005366726297e-05,
      "loss": 0.3879,
      "num_input_tokens_seen": 1437151,
      "step": 2120
    },
    {
      "epoch": 0.7620751341681574,
      "grad_norm": 0.2736769914627075,
      "learning_rate": 4.758497316636852e-05,
      "loss": 0.3994,
      "num_input_tokens_seen": 1443666,
      "step": 2130
    },
    {
      "epoch": 0.7656529516994633,
      "grad_norm": 0.46681609749794006,
      "learning_rate": 4.686940966010734e-05,
      "loss": 0.3754,
      "num_input_tokens_seen": 1449724,
      "step": 2140
    },
    {
      "epoch": 0.7692307692307693,
      "grad_norm": 0.32450494170188904,
      "learning_rate": 4.615384615384616e-05,
      "loss": 0.3712,
      "num_input_tokens_seen": 1455151,
      "step": 2150
    },
    {
      "epoch": 0.7728085867620751,
      "grad_norm": 0.36587849259376526,
      "learning_rate": 4.543828264758497e-05,
      "loss": 0.399,
      "num_input_tokens_seen": 1461336,
      "step": 2160
    },
    {
      "epoch": 0.776386404293381,
      "grad_norm": 0.2556416392326355,
      "learning_rate": 4.472271914132379e-05,
      "loss": 0.3925,
      "num_input_tokens_seen": 1468427,
      "step": 2170
    },
    {
      "epoch": 0.7799642218246869,
      "grad_norm": 0.38359931111335754,
      "learning_rate": 4.400715563506261e-05,
      "loss": 0.3971,
      "num_input_tokens_seen": 1475721,
      "step": 2180
    },
    {
      "epoch": 0.7835420393559929,
      "grad_norm": 0.2836684584617615,
      "learning_rate": 4.329159212880143e-05,
      "loss": 0.3814,
      "num_input_tokens_seen": 1481691,
      "step": 2190
    },
    {
      "epoch": 0.7871198568872988,
      "grad_norm": 0.3486209809780121,
      "learning_rate": 4.257602862254025e-05,
      "loss": 0.4195,
      "num_input_tokens_seen": 1488275,
      "step": 2200
    },
    {
      "epoch": 0.7906976744186046,
      "grad_norm": 0.31744202971458435,
      "learning_rate": 4.186046511627907e-05,
      "loss": 0.3672,
      "num_input_tokens_seen": 1496031,
      "step": 2210
    },
    {
      "epoch": 0.7942754919499105,
      "grad_norm": 0.3008958697319031,
      "learning_rate": 4.114490161001789e-05,
      "loss": 0.3809,
      "num_input_tokens_seen": 1503138,
      "step": 2220
    },
    {
      "epoch": 0.7978533094812165,
      "grad_norm": 0.39187124371528625,
      "learning_rate": 4.042933810375671e-05,
      "loss": 0.3779,
      "num_input_tokens_seen": 1509279,
      "step": 2230
    },
    {
      "epoch": 0.8014311270125224,
      "grad_norm": 0.3644355833530426,
      "learning_rate": 3.971377459749553e-05,
      "loss": 0.3883,
      "num_input_tokens_seen": 1515223,
      "step": 2240
    },
    {
      "epoch": 0.8050089445438283,
      "grad_norm": 0.30987629294395447,
      "learning_rate": 3.899821109123435e-05,
      "loss": 0.3868,
      "num_input_tokens_seen": 1523801,
      "step": 2250
    },
    {
      "epoch": 0.8085867620751341,
      "grad_norm": 0.2742662727832794,
      "learning_rate": 3.828264758497317e-05,
      "loss": 0.369,
      "num_input_tokens_seen": 1530483,
      "step": 2260
    },
    {
      "epoch": 0.8121645796064401,
      "grad_norm": 0.31918245553970337,
      "learning_rate": 3.756708407871199e-05,
      "loss": 0.4161,
      "num_input_tokens_seen": 1538556,
      "step": 2270
    },
    {
      "epoch": 0.815742397137746,
      "grad_norm": 0.36853113770484924,
      "learning_rate": 3.685152057245081e-05,
      "loss": 0.3834,
      "num_input_tokens_seen": 1545040,
      "step": 2280
    },
    {
      "epoch": 0.8193202146690519,
      "grad_norm": 0.5155879855155945,
      "learning_rate": 3.6135957066189624e-05,
      "loss": 0.3814,
      "num_input_tokens_seen": 1552028,
      "step": 2290
    },
    {
      "epoch": 0.8228980322003577,
      "grad_norm": 0.2956937551498413,
      "learning_rate": 3.5420393559928444e-05,
      "loss": 0.395,
      "num_input_tokens_seen": 1559137,
      "step": 2300
    },
    {
      "epoch": 0.8264758497316637,
      "grad_norm": 0.25360018014907837,
      "learning_rate": 3.4704830053667264e-05,
      "loss": 0.3743,
      "num_input_tokens_seen": 1565661,
      "step": 2310
    },
    {
      "epoch": 0.8300536672629696,
      "grad_norm": 0.27616095542907715,
      "learning_rate": 3.3989266547406084e-05,
      "loss": 0.3971,
      "num_input_tokens_seen": 1572070,
      "step": 2320
    },
    {
      "epoch": 0.8336314847942755,
      "grad_norm": 0.3011910021305084,
      "learning_rate": 3.3273703041144904e-05,
      "loss": 0.4008,
      "num_input_tokens_seen": 1579014,
      "step": 2330
    },
    {
      "epoch": 0.8372093023255814,
      "grad_norm": 0.3195567727088928,
      "learning_rate": 3.2558139534883724e-05,
      "loss": 0.362,
      "num_input_tokens_seen": 1586863,
      "step": 2340
    },
    {
      "epoch": 0.8407871198568873,
      "grad_norm": 0.3128429353237152,
      "learning_rate": 3.184257602862254e-05,
      "loss": 0.373,
      "num_input_tokens_seen": 1593227,
      "step": 2350
    },
    {
      "epoch": 0.8443649373881932,
      "grad_norm": 0.3150210678577423,
      "learning_rate": 3.112701252236136e-05,
      "loss": 0.3722,
      "num_input_tokens_seen": 1601444,
      "step": 2360
    },
    {
      "epoch": 0.8479427549194991,
      "grad_norm": 0.44683215022087097,
      "learning_rate": 3.041144901610018e-05,
      "loss": 0.4009,
      "num_input_tokens_seen": 1607535,
      "step": 2370
    },
    {
      "epoch": 0.851520572450805,
      "grad_norm": 0.3482876121997833,
      "learning_rate": 2.9695885509839e-05,
      "loss": 0.3792,
      "num_input_tokens_seen": 1615204,
      "step": 2380
    },
    {
      "epoch": 0.855098389982111,
      "grad_norm": 0.29884225130081177,
      "learning_rate": 2.898032200357782e-05,
      "loss": 0.4037,
      "num_input_tokens_seen": 1621948,
      "step": 2390
    },
    {
      "epoch": 0.8586762075134168,
      "grad_norm": 0.36023974418640137,
      "learning_rate": 2.826475849731664e-05,
      "loss": 0.3994,
      "num_input_tokens_seen": 1628983,
      "step": 2400
    },
    {
      "epoch": 0.8622540250447227,
      "grad_norm": 0.4629397988319397,
      "learning_rate": 2.7549194991055454e-05,
      "loss": 0.3868,
      "num_input_tokens_seen": 1635950,
      "step": 2410
    },
    {
      "epoch": 0.8658318425760286,
      "grad_norm": 0.4515867829322815,
      "learning_rate": 2.6833631484794274e-05,
      "loss": 0.3816,
      "num_input_tokens_seen": 1643015,
      "step": 2420
    },
    {
      "epoch": 0.8694096601073346,
      "grad_norm": 0.33169665932655334,
      "learning_rate": 2.6118067978533094e-05,
      "loss": 0.3707,
      "num_input_tokens_seen": 1648061,
      "step": 2430
    },
    {
      "epoch": 0.8729874776386404,
      "grad_norm": 0.37958571314811707,
      "learning_rate": 2.5402504472271914e-05,
      "loss": 0.3905,
      "num_input_tokens_seen": 1655672,
      "step": 2440
    },
    {
      "epoch": 0.8765652951699463,
      "grad_norm": 0.5768277645111084,
      "learning_rate": 2.4686940966010734e-05,
      "loss": 0.3941,
      "num_input_tokens_seen": 1662028,
      "step": 2450
    },
    {
      "epoch": 0.8801431127012522,
      "grad_norm": 0.4279649257659912,
      "learning_rate": 2.3971377459749554e-05,
      "loss": 0.3874,
      "num_input_tokens_seen": 1668160,
      "step": 2460
    },
    {
      "epoch": 0.8837209302325582,
      "grad_norm": 0.30337658524513245,
      "learning_rate": 2.3255813953488374e-05,
      "loss": 0.3951,
      "num_input_tokens_seen": 1675594,
      "step": 2470
    },
    {
      "epoch": 0.8872987477638641,
      "grad_norm": 0.2993133068084717,
      "learning_rate": 2.2540250447227194e-05,
      "loss": 0.3572,
      "num_input_tokens_seen": 1682014,
      "step": 2480
    },
    {
      "epoch": 0.8908765652951699,
      "grad_norm": 0.4232490062713623,
      "learning_rate": 2.182468694096601e-05,
      "loss": 0.3664,
      "num_input_tokens_seen": 1687943,
      "step": 2490
    },
    {
      "epoch": 0.8944543828264758,
      "grad_norm": 0.4080519676208496,
      "learning_rate": 2.110912343470483e-05,
      "loss": 0.391,
      "num_input_tokens_seen": 1694486,
      "step": 2500
    },
    {
      "epoch": 0.8980322003577818,
      "grad_norm": 0.2778749167919159,
      "learning_rate": 2.039355992844365e-05,
      "loss": 0.3755,
      "num_input_tokens_seen": 1701728,
      "step": 2510
    },
    {
      "epoch": 0.9016100178890877,
      "grad_norm": 0.45739054679870605,
      "learning_rate": 1.967799642218247e-05,
      "loss": 0.3857,
      "num_input_tokens_seen": 1707744,
      "step": 2520
    },
    {
      "epoch": 0.9051878354203936,
      "grad_norm": 0.3340027332305908,
      "learning_rate": 1.8962432915921287e-05,
      "loss": 0.3598,
      "num_input_tokens_seen": 1713524,
      "step": 2530
    },
    {
      "epoch": 0.9087656529516994,
      "grad_norm": 0.29041382670402527,
      "learning_rate": 1.8246869409660107e-05,
      "loss": 0.3874,
      "num_input_tokens_seen": 1720278,
      "step": 2540
    },
    {
      "epoch": 0.9123434704830053,
      "grad_norm": 0.23165664076805115,
      "learning_rate": 1.7531305903398927e-05,
      "loss": 0.4029,
      "num_input_tokens_seen": 1729448,
      "step": 2550
    },
    {
      "epoch": 0.9159212880143113,
      "grad_norm": 0.41043820977211,
      "learning_rate": 1.6815742397137747e-05,
      "loss": 0.3661,
      "num_input_tokens_seen": 1735842,
      "step": 2560
    },
    {
      "epoch": 0.9194991055456172,
      "grad_norm": 0.25310322642326355,
      "learning_rate": 1.6100178890876567e-05,
      "loss": 0.3926,
      "num_input_tokens_seen": 1742027,
      "step": 2570
    },
    {
      "epoch": 0.9230769230769231,
      "grad_norm": 0.32354310154914856,
      "learning_rate": 1.5384615384615387e-05,
      "loss": 0.3949,
      "num_input_tokens_seen": 1748817,
      "step": 2580
    },
    {
      "epoch": 0.9266547406082289,
      "grad_norm": 0.31920483708381653,
      "learning_rate": 1.4669051878354204e-05,
      "loss": 0.3741,
      "num_input_tokens_seen": 1754777,
      "step": 2590
    },
    {
      "epoch": 0.9302325581395349,
      "grad_norm": 0.26824864745140076,
      "learning_rate": 1.3953488372093024e-05,
      "loss": 0.386,
      "num_input_tokens_seen": 1762187,
      "step": 2600
    },
    {
      "epoch": 0.9338103756708408,
      "grad_norm": 0.26331019401550293,
      "learning_rate": 1.3237924865831844e-05,
      "loss": 0.3666,
      "num_input_tokens_seen": 1769463,
      "step": 2610
    },
    {
      "epoch": 0.9373881932021467,
      "grad_norm": 0.3325729966163635,
      "learning_rate": 1.2522361359570662e-05,
      "loss": 0.4048,
      "num_input_tokens_seen": 1777203,
      "step": 2620
    },
    {
      "epoch": 0.9409660107334525,
      "grad_norm": 0.43130627274513245,
      "learning_rate": 1.1806797853309482e-05,
      "loss": 0.3858,
      "num_input_tokens_seen": 1783302,
      "step": 2630
    },
    {
      "epoch": 0.9445438282647585,
      "grad_norm": 0.3261716365814209,
      "learning_rate": 1.1091234347048302e-05,
      "loss": 0.3877,
      "num_input_tokens_seen": 1790778,
      "step": 2640
    },
    {
      "epoch": 0.9481216457960644,
      "grad_norm": 0.3405713140964508,
      "learning_rate": 1.037567084078712e-05,
      "loss": 0.3674,
      "num_input_tokens_seen": 1797992,
      "step": 2650
    },
    {
      "epoch": 0.9516994633273703,
      "grad_norm": 0.4525390565395355,
      "learning_rate": 9.660107334525939e-06,
      "loss": 0.376,
      "num_input_tokens_seen": 1804337,
      "step": 2660
    },
    {
      "epoch": 0.9552772808586762,
      "grad_norm": 0.32699060440063477,
      "learning_rate": 8.944543828264759e-06,
      "loss": 0.4,
      "num_input_tokens_seen": 1811353,
      "step": 2670
    },
    {
      "epoch": 0.9588550983899821,
      "grad_norm": 0.37904760241508484,
      "learning_rate": 8.228980322003579e-06,
      "loss": 0.396,
      "num_input_tokens_seen": 1817606,
      "step": 2680
    },
    {
      "epoch": 0.962432915921288,
      "grad_norm": 0.401185005903244,
      "learning_rate": 7.513416815742398e-06,
      "loss": 0.4044,
      "num_input_tokens_seen": 1826557,
      "step": 2690
    },
    {
      "epoch": 0.9660107334525939,
      "grad_norm": 0.3426940441131592,
      "learning_rate": 6.797853309481217e-06,
      "loss": 0.3881,
      "num_input_tokens_seen": 1832731,
      "step": 2700
    },
    {
      "epoch": 0.9695885509838998,
      "grad_norm": 0.36823177337646484,
      "learning_rate": 6.082289803220036e-06,
      "loss": 0.3836,
      "num_input_tokens_seen": 1838814,
      "step": 2710
    },
    {
      "epoch": 0.9731663685152058,
      "grad_norm": 0.40080201625823975,
      "learning_rate": 5.366726296958855e-06,
      "loss": 0.3393,
      "num_input_tokens_seen": 1844328,
      "step": 2720
    },
    {
      "epoch": 0.9767441860465116,
      "grad_norm": 0.3002487123012543,
      "learning_rate": 4.651162790697674e-06,
      "loss": 0.358,
      "num_input_tokens_seen": 1850313,
      "step": 2730
    },
    {
      "epoch": 0.9803220035778175,
      "grad_norm": 0.29295697808265686,
      "learning_rate": 3.935599284436494e-06,
      "loss": 0.3986,
      "num_input_tokens_seen": 1857019,
      "step": 2740
    },
    {
      "epoch": 0.9838998211091234,
      "grad_norm": 0.33198845386505127,
      "learning_rate": 3.2200357781753134e-06,
      "loss": 0.3699,
      "num_input_tokens_seen": 1865080,
      "step": 2750
    },
    {
      "epoch": 0.9874776386404294,
      "grad_norm": 0.3775390386581421,
      "learning_rate": 2.5044722719141326e-06,
      "loss": 0.3659,
      "num_input_tokens_seen": 1870697,
      "step": 2760
    },
    {
      "epoch": 0.9910554561717353,
      "grad_norm": 0.2900574803352356,
      "learning_rate": 1.7889087656529517e-06,
      "loss": 0.3616,
      "num_input_tokens_seen": 1876560,
      "step": 2770
    },
    {
      "epoch": 0.9946332737030411,
      "grad_norm": 0.3781711459159851,
      "learning_rate": 1.073345259391771e-06,
      "loss": 0.3501,
      "num_input_tokens_seen": 1882565,
      "step": 2780
    }
  ],
  "logging_steps": 10,
  "max_steps": 2795,
  "num_input_tokens_seen": 1882565,
  "num_train_epochs": 1,
  "save_steps": 20,
  "total_flos": 4.233223877501952e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}