{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9998574709003089,
  "eval_steps": 1000,
  "global_step": 3946,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.002533850661176657,
      "grad_norm": 0.2532438337802887,
      "learning_rate": 3.1645569620253167e-06,
      "loss": 1.4823,
      "num_input_tokens_seen": 930032,
      "step": 10
    },
    {
      "epoch": 0.005067701322353314,
      "grad_norm": 0.2576071619987488,
      "learning_rate": 6.329113924050633e-06,
      "loss": 1.45,
      "num_input_tokens_seen": 1834624,
      "step": 20
    },
    {
      "epoch": 0.007601551983529971,
      "grad_norm": 0.27260521054267883,
      "learning_rate": 9.49367088607595e-06,
      "loss": 1.44,
      "num_input_tokens_seen": 2731480,
      "step": 30
    },
    {
      "epoch": 0.010135402644706628,
      "grad_norm": 0.2555162012577057,
      "learning_rate": 1.2658227848101267e-05,
      "loss": 1.4811,
      "num_input_tokens_seen": 3620696,
      "step": 40
    },
    {
      "epoch": 0.012669253305883284,
      "grad_norm": 0.24727804958820343,
      "learning_rate": 1.5822784810126583e-05,
      "loss": 1.4547,
      "num_input_tokens_seen": 4537164,
      "step": 50
    },
    {
      "epoch": 0.015203103967059942,
      "grad_norm": 0.26695573329925537,
      "learning_rate": 1.89873417721519e-05,
      "loss": 1.4288,
      "num_input_tokens_seen": 5457344,
      "step": 60
    },
    {
      "epoch": 0.017736954628236597,
      "grad_norm": 0.2801561653614044,
      "learning_rate": 2.2151898734177217e-05,
      "loss": 1.4569,
      "num_input_tokens_seen": 6349292,
      "step": 70
    },
    {
      "epoch": 0.020270805289413257,
      "grad_norm": 0.22158554196357727,
      "learning_rate": 2.5e-05,
      "loss": 1.4942,
      "num_input_tokens_seen": 7245840,
      "step": 80
    },
    {
      "epoch": 0.022804655950589912,
      "grad_norm": 0.26374679803848267,
      "learning_rate": 2.5e-05,
      "loss": 1.4492,
      "num_input_tokens_seen": 8171092,
      "step": 90
    },
    {
      "epoch": 0.025338506611766568,
      "grad_norm": 0.23668645322322845,
      "learning_rate": 2.5e-05,
      "loss": 1.4677,
      "num_input_tokens_seen": 9093156,
      "step": 100
    },
    {
      "epoch": 0.027872357272943227,
      "grad_norm": 0.25576356053352356,
      "learning_rate": 2.5e-05,
      "loss": 1.4109,
      "num_input_tokens_seen": 9976384,
      "step": 110
    },
    {
      "epoch": 0.030406207934119883,
      "grad_norm": 0.2770518660545349,
      "learning_rate": 2.5e-05,
      "loss": 1.4304,
      "num_input_tokens_seen": 10906092,
      "step": 120
    },
    {
      "epoch": 0.03294005859529654,
      "grad_norm": 0.2333258092403412,
      "learning_rate": 2.5e-05,
      "loss": 1.4142,
      "num_input_tokens_seen": 11818744,
      "step": 130
    },
    {
      "epoch": 0.035473909256473195,
      "grad_norm": 0.24696557223796844,
      "learning_rate": 2.5e-05,
      "loss": 1.4246,
      "num_input_tokens_seen": 12743780,
      "step": 140
    },
    {
      "epoch": 0.038007759917649854,
      "grad_norm": 0.2408542037010193,
      "learning_rate": 2.5e-05,
      "loss": 1.4475,
      "num_input_tokens_seen": 13674048,
      "step": 150
    },
    {
      "epoch": 0.04054161057882651,
      "grad_norm": 0.2496064305305481,
      "learning_rate": 2.5e-05,
      "loss": 1.4529,
      "num_input_tokens_seen": 14538524,
      "step": 160
    },
    {
      "epoch": 0.043075461240003166,
      "grad_norm": 0.2827187478542328,
      "learning_rate": 2.5e-05,
      "loss": 1.4542,
      "num_input_tokens_seen": 15470540,
      "step": 170
    },
    {
      "epoch": 0.045609311901179825,
      "grad_norm": 0.25148963928222656,
      "learning_rate": 2.5e-05,
      "loss": 1.4361,
      "num_input_tokens_seen": 16422976,
      "step": 180
    },
    {
      "epoch": 0.048143162562356484,
      "grad_norm": 0.24195212125778198,
      "learning_rate": 2.5e-05,
      "loss": 1.4775,
      "num_input_tokens_seen": 17344648,
      "step": 190
    },
    {
      "epoch": 0.050677013223533136,
      "grad_norm": 0.3068198263645172,
      "learning_rate": 2.5e-05,
      "loss": 1.4231,
      "num_input_tokens_seen": 18245048,
      "step": 200
    },
    {
      "epoch": 0.053210863884709796,
      "grad_norm": 0.24267973005771637,
      "learning_rate": 2.5e-05,
      "loss": 1.4354,
      "num_input_tokens_seen": 19169384,
      "step": 210
    },
    {
      "epoch": 0.055744714545886455,
      "grad_norm": 0.21026775240898132,
      "learning_rate": 2.5e-05,
      "loss": 1.4153,
      "num_input_tokens_seen": 20096992,
      "step": 220
    },
    {
      "epoch": 0.05827856520706311,
      "grad_norm": 0.21877512335777283,
      "learning_rate": 2.5e-05,
      "loss": 1.3941,
      "num_input_tokens_seen": 21025604,
      "step": 230
    },
    {
      "epoch": 0.060812415868239766,
      "grad_norm": 0.24055704474449158,
      "learning_rate": 2.5e-05,
      "loss": 1.4348,
      "num_input_tokens_seen": 21935180,
      "step": 240
    },
    {
      "epoch": 0.06334626652941643,
      "grad_norm": 0.24673806130886078,
      "learning_rate": 2.5e-05,
      "loss": 1.3719,
      "num_input_tokens_seen": 22857776,
      "step": 250
    },
    {
      "epoch": 0.06588011719059308,
      "grad_norm": 0.21661491692066193,
      "learning_rate": 2.5e-05,
      "loss": 1.4707,
      "num_input_tokens_seen": 23805840,
      "step": 260
    },
    {
      "epoch": 0.06841396785176973,
      "grad_norm": 0.2766810357570648,
      "learning_rate": 2.5e-05,
      "loss": 1.4558,
      "num_input_tokens_seen": 24694772,
      "step": 270
    },
    {
      "epoch": 0.07094781851294639,
      "grad_norm": 0.2665688097476959,
      "learning_rate": 2.5e-05,
      "loss": 1.4026,
      "num_input_tokens_seen": 25637024,
      "step": 280
    },
    {
      "epoch": 0.07348166917412305,
      "grad_norm": 0.2424854040145874,
      "learning_rate": 2.5e-05,
      "loss": 1.3998,
      "num_input_tokens_seen": 26530332,
      "step": 290
    },
    {
      "epoch": 0.07601551983529971,
      "grad_norm": 0.23512804508209229,
      "learning_rate": 2.5e-05,
      "loss": 1.4188,
      "num_input_tokens_seen": 27449020,
      "step": 300
    },
    {
      "epoch": 0.07854937049647637,
      "grad_norm": 0.23620112240314484,
      "learning_rate": 2.5e-05,
      "loss": 1.39,
      "num_input_tokens_seen": 28367404,
      "step": 310
    },
    {
      "epoch": 0.08108322115765303,
      "grad_norm": 0.2523897588253021,
      "learning_rate": 2.5e-05,
      "loss": 1.4487,
      "num_input_tokens_seen": 29277896,
      "step": 320
    },
    {
      "epoch": 0.08361707181882967,
      "grad_norm": 0.24064438045024872,
      "learning_rate": 2.5e-05,
      "loss": 1.4214,
      "num_input_tokens_seen": 30200812,
      "step": 330
    },
    {
      "epoch": 0.08615092248000633,
      "grad_norm": 0.2440669983625412,
      "learning_rate": 2.5e-05,
      "loss": 1.444,
      "num_input_tokens_seen": 31158760,
      "step": 340
    },
    {
      "epoch": 0.08868477314118299,
      "grad_norm": 0.22009992599487305,
      "learning_rate": 2.5e-05,
      "loss": 1.4127,
      "num_input_tokens_seen": 32069424,
      "step": 350
    },
    {
      "epoch": 0.09121862380235965,
      "grad_norm": 0.29601845145225525,
      "learning_rate": 2.5e-05,
      "loss": 1.4437,
      "num_input_tokens_seen": 33009436,
      "step": 360
    },
    {
      "epoch": 0.09375247446353631,
      "grad_norm": 0.2240906059741974,
      "learning_rate": 2.5e-05,
      "loss": 1.3871,
      "num_input_tokens_seen": 33933612,
      "step": 370
    },
    {
      "epoch": 0.09628632512471297,
      "grad_norm": 0.23164159059524536,
      "learning_rate": 2.5e-05,
      "loss": 1.4149,
      "num_input_tokens_seen": 34839560,
      "step": 380
    },
    {
      "epoch": 0.09882017578588961,
      "grad_norm": 0.335622638463974,
      "learning_rate": 2.5e-05,
      "loss": 1.427,
      "num_input_tokens_seen": 35748032,
      "step": 390
    },
    {
      "epoch": 0.10135402644706627,
      "grad_norm": 0.22885636985301971,
      "learning_rate": 2.5e-05,
      "loss": 1.4521,
      "num_input_tokens_seen": 36672280,
      "step": 400
    },
    {
      "epoch": 0.10388787710824293,
      "grad_norm": 0.2555045783519745,
      "learning_rate": 2.5e-05,
      "loss": 1.4175,
      "num_input_tokens_seen": 37599516,
      "step": 410
    },
    {
      "epoch": 0.10642172776941959,
      "grad_norm": 0.24946229159832,
      "learning_rate": 2.5e-05,
      "loss": 1.4276,
      "num_input_tokens_seen": 38529556,
      "step": 420
    },
    {
      "epoch": 0.10895557843059625,
      "grad_norm": 0.24785666167736053,
      "learning_rate": 2.5e-05,
      "loss": 1.4007,
      "num_input_tokens_seen": 39460044,
      "step": 430
    },
    {
      "epoch": 0.11148942909177291,
      "grad_norm": 0.22006012499332428,
      "learning_rate": 2.5e-05,
      "loss": 1.4238,
      "num_input_tokens_seen": 40369364,
      "step": 440
    },
    {
      "epoch": 0.11402327975294956,
      "grad_norm": 0.26216018199920654,
      "learning_rate": 2.5e-05,
      "loss": 1.4318,
      "num_input_tokens_seen": 41307640,
      "step": 450
    },
    {
      "epoch": 0.11655713041412621,
      "grad_norm": 0.23494452238082886,
      "learning_rate": 2.5e-05,
      "loss": 1.417,
      "num_input_tokens_seen": 42200280,
      "step": 460
    },
    {
      "epoch": 0.11909098107530287,
      "grad_norm": 0.23429952561855316,
      "learning_rate": 2.5e-05,
      "loss": 1.4277,
      "num_input_tokens_seen": 43112444,
      "step": 470
    },
    {
      "epoch": 0.12162483173647953,
      "grad_norm": 0.2510409355163574,
      "learning_rate": 2.5e-05,
      "loss": 1.3853,
      "num_input_tokens_seen": 44021860,
      "step": 480
    },
    {
      "epoch": 0.12415868239765619,
      "grad_norm": 0.2570734918117523,
      "learning_rate": 2.5e-05,
      "loss": 1.4056,
      "num_input_tokens_seen": 44938384,
      "step": 490
    },
    {
      "epoch": 0.12669253305883285,
      "grad_norm": 0.23910905420780182,
      "learning_rate": 2.5e-05,
      "loss": 1.4303,
      "num_input_tokens_seen": 45871544,
      "step": 500
    },
    {
      "epoch": 0.1292263837200095,
      "grad_norm": 0.2258525788784027,
      "learning_rate": 2.5e-05,
      "loss": 1.4243,
      "num_input_tokens_seen": 46798524,
      "step": 510
    },
    {
      "epoch": 0.13176023438118617,
      "grad_norm": 0.21156556904315948,
      "learning_rate": 2.5e-05,
      "loss": 1.3965,
      "num_input_tokens_seen": 47696192,
      "step": 520
    },
    {
      "epoch": 0.13429408504236282,
      "grad_norm": 0.2665134370326996,
      "learning_rate": 2.5e-05,
      "loss": 1.393,
      "num_input_tokens_seen": 48669228,
      "step": 530
    },
    {
      "epoch": 0.13682793570353946,
      "grad_norm": 0.2551543414592743,
      "learning_rate": 2.5e-05,
      "loss": 1.4184,
      "num_input_tokens_seen": 49616616,
      "step": 540
    },
    {
      "epoch": 0.13936178636471613,
      "grad_norm": 0.2285103052854538,
      "learning_rate": 2.5e-05,
      "loss": 1.3964,
      "num_input_tokens_seen": 50540636,
      "step": 550
    },
    {
      "epoch": 0.14189563702589278,
      "grad_norm": 0.23576393723487854,
      "learning_rate": 2.5e-05,
      "loss": 1.4187,
      "num_input_tokens_seen": 51440464,
      "step": 560
    },
    {
      "epoch": 0.14442948768706945,
      "grad_norm": 0.22209148108959198,
      "learning_rate": 2.5e-05,
      "loss": 1.403,
      "num_input_tokens_seen": 52315124,
      "step": 570
    },
    {
      "epoch": 0.1469633383482461,
      "grad_norm": 0.23545274138450623,
      "learning_rate": 2.5e-05,
      "loss": 1.4313,
      "num_input_tokens_seen": 53261804,
      "step": 580
    },
    {
      "epoch": 0.14949718900942277,
      "grad_norm": 0.25153088569641113,
      "learning_rate": 2.5e-05,
      "loss": 1.3798,
      "num_input_tokens_seen": 54106436,
      "step": 590
    },
    {
      "epoch": 0.15203103967059942,
      "grad_norm": 0.23856191337108612,
      "learning_rate": 2.5e-05,
      "loss": 1.3679,
      "num_input_tokens_seen": 55035052,
      "step": 600
    },
    {
      "epoch": 0.15456489033177606,
      "grad_norm": 0.23667120933532715,
      "learning_rate": 2.5e-05,
      "loss": 1.4191,
      "num_input_tokens_seen": 55935200,
      "step": 610
    },
    {
      "epoch": 0.15709874099295273,
      "grad_norm": 0.26784512400627136,
      "learning_rate": 2.5e-05,
      "loss": 1.3684,
      "num_input_tokens_seen": 56843340,
      "step": 620
    },
    {
      "epoch": 0.15963259165412938,
      "grad_norm": 0.22612795233726501,
      "learning_rate": 2.5e-05,
      "loss": 1.436,
      "num_input_tokens_seen": 57720808,
      "step": 630
    },
    {
      "epoch": 0.16216644231530605,
      "grad_norm": 0.24946410953998566,
      "learning_rate": 2.5e-05,
      "loss": 1.426,
      "num_input_tokens_seen": 58575924,
      "step": 640
    },
    {
      "epoch": 0.1647002929764827,
      "grad_norm": 0.2528791129589081,
      "learning_rate": 2.5e-05,
      "loss": 1.4191,
      "num_input_tokens_seen": 59484056,
      "step": 650
    },
    {
      "epoch": 0.16723414363765934,
      "grad_norm": 0.21960842609405518,
      "learning_rate": 2.5e-05,
      "loss": 1.443,
      "num_input_tokens_seen": 60382860,
      "step": 660
    },
    {
      "epoch": 0.16976799429883602,
      "grad_norm": 0.2500540018081665,
      "learning_rate": 2.5e-05,
      "loss": 1.4284,
      "num_input_tokens_seen": 61291764,
      "step": 670
    },
    {
      "epoch": 0.17230184496001266,
      "grad_norm": 0.27140355110168457,
      "learning_rate": 2.5e-05,
      "loss": 1.3909,
      "num_input_tokens_seen": 62183556,
      "step": 680
    },
    {
      "epoch": 0.17483569562118934,
      "grad_norm": 0.22307205200195312,
      "learning_rate": 2.5e-05,
      "loss": 1.3682,
      "num_input_tokens_seen": 63098340,
      "step": 690
    },
    {
      "epoch": 0.17736954628236598,
      "grad_norm": 0.24494685232639313,
      "learning_rate": 2.5e-05,
      "loss": 1.3903,
      "num_input_tokens_seen": 64000524,
      "step": 700
    },
    {
      "epoch": 0.17990339694354263,
      "grad_norm": 0.2667907476425171,
      "learning_rate": 2.5e-05,
      "loss": 1.4569,
      "num_input_tokens_seen": 64937424,
      "step": 710
    },
    {
      "epoch": 0.1824372476047193,
      "grad_norm": 0.22164462506771088,
      "learning_rate": 2.5e-05,
      "loss": 1.3806,
      "num_input_tokens_seen": 65822472,
      "step": 720
    },
    {
      "epoch": 0.18497109826589594,
      "grad_norm": 0.23859019577503204,
      "learning_rate": 2.5e-05,
      "loss": 1.4003,
      "num_input_tokens_seen": 66691752,
      "step": 730
    },
    {
      "epoch": 0.18750494892707262,
      "grad_norm": 0.28847405314445496,
      "learning_rate": 2.5e-05,
      "loss": 1.4076,
      "num_input_tokens_seen": 67658948,
      "step": 740
    },
    {
      "epoch": 0.19003879958824926,
      "grad_norm": 0.2571374177932739,
      "learning_rate": 2.5e-05,
      "loss": 1.3924,
      "num_input_tokens_seen": 68572048,
      "step": 750
    },
    {
      "epoch": 0.19257265024942594,
      "grad_norm": 0.24991680681705475,
      "learning_rate": 2.5e-05,
      "loss": 1.4164,
      "num_input_tokens_seen": 69502808,
      "step": 760
    },
    {
      "epoch": 0.19510650091060258,
      "grad_norm": 0.23006725311279297,
      "learning_rate": 2.5e-05,
      "loss": 1.4019,
      "num_input_tokens_seen": 70423124,
      "step": 770
    },
    {
      "epoch": 0.19764035157177923,
      "grad_norm": 0.2484099566936493,
      "learning_rate": 2.5e-05,
      "loss": 1.3565,
      "num_input_tokens_seen": 71271484,
      "step": 780
    },
    {
      "epoch": 0.2001742022329559,
      "grad_norm": 0.2604601979255676,
      "learning_rate": 2.5e-05,
      "loss": 1.4098,
      "num_input_tokens_seen": 72179604,
      "step": 790
    },
    {
      "epoch": 0.20270805289413255,
      "grad_norm": 0.2681257724761963,
      "learning_rate": 2.5e-05,
      "loss": 1.4152,
      "num_input_tokens_seen": 73085296,
      "step": 800
    },
    {
      "epoch": 0.20524190355530922,
      "grad_norm": 0.20966367423534393,
      "learning_rate": 2.5e-05,
      "loss": 1.4006,
      "num_input_tokens_seen": 74003640,
      "step": 810
    },
    {
      "epoch": 0.20777575421648586,
      "grad_norm": 0.2371470183134079,
      "learning_rate": 2.5e-05,
      "loss": 1.3651,
      "num_input_tokens_seen": 74957748,
      "step": 820
    },
    {
      "epoch": 0.2103096048776625,
      "grad_norm": 0.24214884638786316,
      "learning_rate": 2.5e-05,
      "loss": 1.3969,
      "num_input_tokens_seen": 75841664,
      "step": 830
    },
    {
      "epoch": 0.21284345553883918,
      "grad_norm": 0.24258075654506683,
      "learning_rate": 2.5e-05,
      "loss": 1.4356,
      "num_input_tokens_seen": 76765412,
      "step": 840
    },
    {
      "epoch": 0.21537730620001583,
      "grad_norm": 0.25199827551841736,
      "learning_rate": 2.5e-05,
      "loss": 1.4171,
      "num_input_tokens_seen": 77675892,
      "step": 850
    },
    {
      "epoch": 0.2179111568611925,
      "grad_norm": 0.219390869140625,
      "learning_rate": 2.5e-05,
      "loss": 1.3713,
      "num_input_tokens_seen": 78646236,
      "step": 860
    },
    {
      "epoch": 0.22044500752236915,
      "grad_norm": 0.2546541690826416,
      "learning_rate": 2.5e-05,
      "loss": 1.4154,
      "num_input_tokens_seen": 79594216,
      "step": 870
    },
    {
      "epoch": 0.22297885818354582,
      "grad_norm": 0.28596746921539307,
      "learning_rate": 2.5e-05,
      "loss": 1.3981,
      "num_input_tokens_seen": 80523804,
      "step": 880
    },
    {
      "epoch": 0.22551270884472246,
      "grad_norm": 0.21436405181884766,
      "learning_rate": 2.5e-05,
      "loss": 1.3889,
      "num_input_tokens_seen": 81405376,
      "step": 890
    },
    {
      "epoch": 0.2280465595058991,
      "grad_norm": 0.2508715093135834,
      "learning_rate": 2.5e-05,
      "loss": 1.3682,
      "num_input_tokens_seen": 82260336,
      "step": 900
    },
    {
      "epoch": 0.23058041016707578,
      "grad_norm": 0.24959874153137207,
      "learning_rate": 2.5e-05,
      "loss": 1.3651,
      "num_input_tokens_seen": 83190224,
      "step": 910
    },
    {
      "epoch": 0.23311426082825243,
      "grad_norm": 0.27335524559020996,
      "learning_rate": 2.5e-05,
      "loss": 1.4221,
      "num_input_tokens_seen": 84107372,
      "step": 920
    },
    {
      "epoch": 0.2356481114894291,
      "grad_norm": 0.2550046443939209,
      "learning_rate": 2.5e-05,
      "loss": 1.4029,
      "num_input_tokens_seen": 85024192,
      "step": 930
    },
    {
      "epoch": 0.23818196215060575,
      "grad_norm": 0.23554718494415283,
      "learning_rate": 2.5e-05,
      "loss": 1.4045,
      "num_input_tokens_seen": 85956220,
      "step": 940
    },
    {
      "epoch": 0.2407158128117824,
      "grad_norm": 0.21662922203540802,
      "learning_rate": 2.5e-05,
      "loss": 1.3908,
      "num_input_tokens_seen": 86858100,
      "step": 950
    },
    {
      "epoch": 0.24324966347295907,
      "grad_norm": 0.22381572425365448,
      "learning_rate": 2.5e-05,
      "loss": 1.4337,
      "num_input_tokens_seen": 87771400,
      "step": 960
    },
    {
      "epoch": 0.2457835141341357,
      "grad_norm": 0.2680582106113434,
      "learning_rate": 2.5e-05,
      "loss": 1.4325,
      "num_input_tokens_seen": 88675708,
      "step": 970
    },
    {
      "epoch": 0.24831736479531238,
      "grad_norm": 0.22555038332939148,
      "learning_rate": 2.5e-05,
      "loss": 1.3741,
      "num_input_tokens_seen": 89561964,
      "step": 980
    },
    {
      "epoch": 0.25085121545648903,
      "grad_norm": 0.2812931537628174,
      "learning_rate": 2.5e-05,
      "loss": 1.4104,
      "num_input_tokens_seen": 90488048,
      "step": 990
    },
    {
      "epoch": 0.2533850661176657,
      "grad_norm": 0.23613446950912476,
      "learning_rate": 2.5e-05,
      "loss": 1.4008,
      "num_input_tokens_seen": 91375832,
      "step": 1000
    },
    {
      "epoch": 0.2533850661176657,
      "eval_loss": 1.4020060300827026,
      "eval_runtime": 2.9465,
      "eval_samples_per_second": 50.908,
      "eval_steps_per_second": 6.448,
      "num_input_tokens_seen": 91375832,
      "step": 1000
    },
    {
      "epoch": 0.2559189167788423,
      "grad_norm": 0.2325298935174942,
      "learning_rate": 2.5e-05,
      "loss": 1.3544,
      "num_input_tokens_seen": 92347240,
      "step": 1010
    },
    {
      "epoch": 0.258452767440019,
      "grad_norm": 0.24142597615718842,
      "learning_rate": 2.5e-05,
      "loss": 1.3706,
      "num_input_tokens_seen": 93237456,
      "step": 1020
    },
    {
      "epoch": 0.26098661810119567,
      "grad_norm": 0.2356724739074707,
      "learning_rate": 2.5e-05,
      "loss": 1.3998,
      "num_input_tokens_seen": 94145764,
      "step": 1030
    },
    {
      "epoch": 0.26352046876237234,
      "grad_norm": 0.243470698595047,
      "learning_rate": 2.5e-05,
      "loss": 1.4013,
      "num_input_tokens_seen": 95055692,
      "step": 1040
    },
    {
      "epoch": 0.26605431942354896,
      "grad_norm": 0.2412971556186676,
      "learning_rate": 2.5e-05,
      "loss": 1.373,
      "num_input_tokens_seen": 95921656,
      "step": 1050
    },
    {
      "epoch": 0.26858817008472563,
      "grad_norm": 0.2889567017555237,
      "learning_rate": 2.5e-05,
      "loss": 1.3898,
      "num_input_tokens_seen": 96821452,
      "step": 1060
    },
    {
      "epoch": 0.2711220207459023,
      "grad_norm": 0.23939931392669678,
      "learning_rate": 2.5e-05,
      "loss": 1.4088,
      "num_input_tokens_seen": 97727612,
      "step": 1070
    },
    {
      "epoch": 0.2736558714070789,
      "grad_norm": 0.25132742524147034,
      "learning_rate": 2.5e-05,
      "loss": 1.3853,
      "num_input_tokens_seen": 98677952,
      "step": 1080
    },
    {
      "epoch": 0.2761897220682556,
      "grad_norm": 0.2225540727376938,
      "learning_rate": 2.5e-05,
      "loss": 1.4041,
      "num_input_tokens_seen": 99640748,
      "step": 1090
    },
    {
      "epoch": 0.27872357272943227,
      "grad_norm": 0.24503560364246368,
      "learning_rate": 2.5e-05,
      "loss": 1.3719,
      "num_input_tokens_seen": 100557008,
      "step": 1100
    },
    {
      "epoch": 0.28125742339060894,
      "grad_norm": 0.2348717302083969,
      "learning_rate": 2.5e-05,
      "loss": 1.3937,
      "num_input_tokens_seen": 101442164,
      "step": 1110
    },
    {
      "epoch": 0.28379127405178556,
      "grad_norm": 0.24240590631961823,
      "learning_rate": 2.5e-05,
      "loss": 1.3641,
      "num_input_tokens_seen": 102366056,
      "step": 1120
    },
    {
      "epoch": 0.28632512471296223,
      "grad_norm": 0.2246118187904358,
      "learning_rate": 2.5e-05,
      "loss": 1.3631,
      "num_input_tokens_seen": 103261480,
      "step": 1130
    },
    {
      "epoch": 0.2888589753741389,
      "grad_norm": 0.2967662513256073,
      "learning_rate": 2.5e-05,
      "loss": 1.3883,
      "num_input_tokens_seen": 104163484,
      "step": 1140
    },
    {
      "epoch": 0.2913928260353155,
      "grad_norm": 0.24722802639007568,
      "learning_rate": 2.5e-05,
      "loss": 1.4444,
      "num_input_tokens_seen": 105077064,
      "step": 1150
    },
    {
      "epoch": 0.2939266766964922,
      "grad_norm": 0.2221587598323822,
      "learning_rate": 2.5e-05,
      "loss": 1.3809,
      "num_input_tokens_seen": 105968728,
      "step": 1160
    },
    {
      "epoch": 0.29646052735766887,
      "grad_norm": 0.23813994228839874,
      "learning_rate": 2.5e-05,
      "loss": 1.3941,
      "num_input_tokens_seen": 106838388,
      "step": 1170
    },
    {
      "epoch": 0.29899437801884554,
      "grad_norm": 0.24747894704341888,
      "learning_rate": 2.5e-05,
      "loss": 1.3812,
      "num_input_tokens_seen": 107764200,
      "step": 1180
    },
    {
      "epoch": 0.30152822868002216,
      "grad_norm": 0.26802727580070496,
      "learning_rate": 2.5e-05,
      "loss": 1.3717,
      "num_input_tokens_seen": 108683176,
      "step": 1190
    },
    {
      "epoch": 0.30406207934119883,
      "grad_norm": 0.27138280868530273,
      "learning_rate": 2.5e-05,
      "loss": 1.367,
      "num_input_tokens_seen": 109606364,
      "step": 1200
    },
    {
      "epoch": 0.3065959300023755,
      "grad_norm": 0.24378275871276855,
      "learning_rate": 2.5e-05,
      "loss": 1.3762,
      "num_input_tokens_seen": 110518296,
      "step": 1210
    },
    {
      "epoch": 0.3091297806635521,
      "grad_norm": 0.261106938123703,
      "learning_rate": 2.5e-05,
      "loss": 1.4227,
      "num_input_tokens_seen": 111436828,
      "step": 1220
    },
    {
      "epoch": 0.3116636313247288,
      "grad_norm": 0.2597008943557739,
      "learning_rate": 2.5e-05,
      "loss": 1.3847,
      "num_input_tokens_seen": 112334112,
      "step": 1230
    },
    {
      "epoch": 0.31419748198590547,
      "grad_norm": 0.24535202980041504,
      "learning_rate": 2.5e-05,
      "loss": 1.3706,
      "num_input_tokens_seen": 113211652,
      "step": 1240
    },
    {
      "epoch": 0.3167313326470821,
      "grad_norm": 0.2770673632621765,
      "learning_rate": 2.5e-05,
      "loss": 1.3975,
      "num_input_tokens_seen": 114117744,
      "step": 1250
    },
    {
      "epoch": 0.31926518330825876,
      "grad_norm": 0.21976234018802643,
      "learning_rate": 2.5e-05,
      "loss": 1.4036,
      "num_input_tokens_seen": 115002568,
      "step": 1260
    },
    {
      "epoch": 0.32179903396943543,
      "grad_norm": 0.22749099135398865,
      "learning_rate": 2.5e-05,
      "loss": 1.3625,
      "num_input_tokens_seen": 115904964,
      "step": 1270
    },
    {
      "epoch": 0.3243328846306121,
      "grad_norm": 0.22470030188560486,
      "learning_rate": 2.5e-05,
      "loss": 1.3905,
      "num_input_tokens_seen": 116843732,
      "step": 1280
    },
    {
      "epoch": 0.3268667352917887,
      "grad_norm": 0.2671917974948883,
      "learning_rate": 2.5e-05,
      "loss": 1.3839,
      "num_input_tokens_seen": 117752200,
      "step": 1290
    },
    {
      "epoch": 0.3294005859529654,
      "grad_norm": 0.24347306787967682,
      "learning_rate": 2.5e-05,
      "loss": 1.36,
      "num_input_tokens_seen": 118656912,
      "step": 1300
    },
    {
      "epoch": 0.33193443661414207,
      "grad_norm": 0.22786876559257507,
      "learning_rate": 2.5e-05,
      "loss": 1.361,
      "num_input_tokens_seen": 119561700,
      "step": 1310
    },
    {
      "epoch": 0.3344682872753187,
      "grad_norm": 0.22891202569007874,
      "learning_rate": 2.5e-05,
      "loss": 1.3916,
      "num_input_tokens_seen": 120537120,
      "step": 1320
    },
    {
      "epoch": 0.33700213793649536,
      "grad_norm": 0.2579503357410431,
      "learning_rate": 2.5e-05,
      "loss": 1.4077,
      "num_input_tokens_seen": 121473416,
      "step": 1330
    },
    {
      "epoch": 0.33953598859767203,
      "grad_norm": 0.24670307338237762,
      "learning_rate": 2.5e-05,
      "loss": 1.4055,
      "num_input_tokens_seen": 122383356,
      "step": 1340
    },
    {
      "epoch": 0.3420698392588487,
      "grad_norm": 0.2923058569431305,
      "learning_rate": 2.5e-05,
      "loss": 1.3875,
      "num_input_tokens_seen": 123309020,
      "step": 1350
    },
    {
      "epoch": 0.3446036899200253,
      "grad_norm": 0.2256019562482834,
      "learning_rate": 2.5e-05,
      "loss": 1.3872,
      "num_input_tokens_seen": 124234924,
      "step": 1360
    },
    {
      "epoch": 0.347137540581202,
      "grad_norm": 0.2368822544813156,
      "learning_rate": 2.5e-05,
      "loss": 1.3969,
      "num_input_tokens_seen": 125162100,
      "step": 1370
    },
    {
      "epoch": 0.34967139124237867,
      "grad_norm": 0.2430727332830429,
      "learning_rate": 2.5e-05,
      "loss": 1.3638,
      "num_input_tokens_seen": 126113704,
      "step": 1380
    },
    {
      "epoch": 0.3522052419035553,
      "grad_norm": 0.23543952405452728,
      "learning_rate": 2.5e-05,
      "loss": 1.3642,
      "num_input_tokens_seen": 127052976,
      "step": 1390
    },
    {
      "epoch": 0.35473909256473196,
      "grad_norm": 0.24988651275634766,
      "learning_rate": 2.5e-05,
      "loss": 1.3784,
      "num_input_tokens_seen": 127996892,
      "step": 1400
    },
    {
      "epoch": 0.35727294322590863,
      "grad_norm": 0.2787221670150757,
      "learning_rate": 2.5e-05,
      "loss": 1.4052,
      "num_input_tokens_seen": 128935380,
      "step": 1410
    },
    {
      "epoch": 0.35980679388708525,
      "grad_norm": 0.24997858703136444,
      "learning_rate": 2.5e-05,
      "loss": 1.3791,
      "num_input_tokens_seen": 129871964,
      "step": 1420
    },
    {
      "epoch": 0.3623406445482619,
      "grad_norm": 0.24547652900218964,
      "learning_rate": 2.5e-05,
      "loss": 1.395,
      "num_input_tokens_seen": 130767084,
      "step": 1430
    },
    {
      "epoch": 0.3648744952094386,
      "grad_norm": 0.23068061470985413,
      "learning_rate": 2.5e-05,
      "loss": 1.3677,
      "num_input_tokens_seen": 131674508,
      "step": 1440
    },
    {
      "epoch": 0.36740834587061527,
      "grad_norm": 0.23524820804595947,
      "learning_rate": 2.5e-05,
      "loss": 1.4161,
      "num_input_tokens_seen": 132602416,
      "step": 1450
    },
    {
      "epoch": 0.3699421965317919,
      "grad_norm": 0.23469901084899902,
      "learning_rate": 2.5e-05,
      "loss": 1.3721,
      "num_input_tokens_seen": 133506196,
      "step": 1460
    },
    {
      "epoch": 0.37247604719296856,
      "grad_norm": 0.24987129867076874,
      "learning_rate": 2.5e-05,
      "loss": 1.4049,
      "num_input_tokens_seen": 134427152,
      "step": 1470
    },
    {
      "epoch": 0.37500989785414524,
      "grad_norm": 0.24462181329727173,
      "learning_rate": 2.5e-05,
      "loss": 1.3797,
      "num_input_tokens_seen": 135314244,
      "step": 1480
    },
    {
      "epoch": 0.37754374851532185,
      "grad_norm": 0.2653500437736511,
      "learning_rate": 2.5e-05,
      "loss": 1.3503,
      "num_input_tokens_seen": 136230948,
      "step": 1490
    },
    {
      "epoch": 0.3800775991764985,
      "grad_norm": 0.2400883287191391,
      "learning_rate": 2.5e-05,
      "loss": 1.3957,
      "num_input_tokens_seen": 137179452,
      "step": 1500
    },
    {
      "epoch": 0.3826114498376752,
      "grad_norm": 0.2289241999387741,
      "learning_rate": 2.5e-05,
      "loss": 1.3529,
      "num_input_tokens_seen": 138078404,
      "step": 1510
    },
    {
      "epoch": 0.3851453004988519,
      "grad_norm": 0.26289331912994385,
      "learning_rate": 2.5e-05,
      "loss": 1.4398,
      "num_input_tokens_seen": 138991724,
      "step": 1520
    },
    {
      "epoch": 0.3876791511600285,
      "grad_norm": 0.2165287286043167,
      "learning_rate": 2.5e-05,
      "loss": 1.41,
      "num_input_tokens_seen": 139933240,
      "step": 1530
    },
    {
      "epoch": 0.39021300182120516,
      "grad_norm": 0.29837462306022644,
      "learning_rate": 2.5e-05,
      "loss": 1.3662,
      "num_input_tokens_seen": 140836772,
      "step": 1540
    },
    {
      "epoch": 0.39274685248238184,
      "grad_norm": 0.24651922285556793,
      "learning_rate": 2.5e-05,
      "loss": 1.3412,
      "num_input_tokens_seen": 141744576,
      "step": 1550
    },
    {
      "epoch": 0.39528070314355845,
      "grad_norm": 0.29952993988990784,
      "learning_rate": 2.5e-05,
      "loss": 1.3907,
      "num_input_tokens_seen": 142624188,
      "step": 1560
    },
    {
      "epoch": 0.3978145538047351,
      "grad_norm": 0.2563650608062744,
      "learning_rate": 2.5e-05,
      "loss": 1.3858,
      "num_input_tokens_seen": 143554872,
      "step": 1570
    },
    {
      "epoch": 0.4003484044659118,
      "grad_norm": 0.2565977871417999,
      "learning_rate": 2.5e-05,
      "loss": 1.3732,
      "num_input_tokens_seen": 144477588,
      "step": 1580
    },
    {
      "epoch": 0.4028822551270885,
      "grad_norm": 0.2879079282283783,
      "learning_rate": 2.5e-05,
      "loss": 1.3692,
      "num_input_tokens_seen": 145354620,
      "step": 1590
    },
    {
      "epoch": 0.4054161057882651,
      "grad_norm": 0.2640700936317444,
      "learning_rate": 2.5e-05,
      "loss": 1.3909,
      "num_input_tokens_seen": 146266280,
      "step": 1600
    },
    {
      "epoch": 0.40794995644944176,
      "grad_norm": 0.26872700452804565,
      "learning_rate": 2.5e-05,
      "loss": 1.4051,
      "num_input_tokens_seen": 147165620,
      "step": 1610
    },
    {
      "epoch": 0.41048380711061844,
      "grad_norm": 0.2187357246875763,
      "learning_rate": 2.5e-05,
      "loss": 1.38,
      "num_input_tokens_seen": 148098344,
      "step": 1620
    },
    {
      "epoch": 0.41301765777179505,
      "grad_norm": 0.24293020367622375,
      "learning_rate": 2.5e-05,
      "loss": 1.3915,
      "num_input_tokens_seen": 149043924,
      "step": 1630
    },
    {
      "epoch": 0.41555150843297173,
      "grad_norm": 0.23092688620090485,
      "learning_rate": 2.5e-05,
      "loss": 1.4019,
      "num_input_tokens_seen": 149996036,
      "step": 1640
    },
    {
      "epoch": 0.4180853590941484,
      "grad_norm": 0.27063265442848206,
      "learning_rate": 2.5e-05,
      "loss": 1.3723,
      "num_input_tokens_seen": 150869152,
      "step": 1650
    },
    {
      "epoch": 0.420619209755325,
      "grad_norm": 0.25822359323501587,
      "learning_rate": 2.5e-05,
      "loss": 1.3682,
      "num_input_tokens_seen": 151783488,
      "step": 1660
    },
    {
      "epoch": 0.4231530604165017,
      "grad_norm": 0.269724041223526,
      "learning_rate": 2.5e-05,
      "loss": 1.3592,
      "num_input_tokens_seen": 152700960,
      "step": 1670
    },
    {
      "epoch": 0.42568691107767836,
      "grad_norm": 0.23563367128372192,
      "learning_rate": 2.5e-05,
      "loss": 1.3679,
      "num_input_tokens_seen": 153634040,
      "step": 1680
    },
    {
      "epoch": 0.42822076173885504,
      "grad_norm": 0.23306426405906677,
      "learning_rate": 2.5e-05,
      "loss": 1.3585,
      "num_input_tokens_seen": 154569656,
      "step": 1690
    },
    {
      "epoch": 0.43075461240003166,
      "grad_norm": 0.23761169612407684,
      "learning_rate": 2.5e-05,
      "loss": 1.3413,
      "num_input_tokens_seen": 155491724,
      "step": 1700
    },
    {
      "epoch": 0.43328846306120833,
      "grad_norm": 0.23138809204101562,
      "learning_rate": 2.5e-05,
      "loss": 1.3892,
      "num_input_tokens_seen": 156437340,
      "step": 1710
    },
    {
      "epoch": 0.435822313722385,
      "grad_norm": 0.24864792823791504,
      "learning_rate": 2.5e-05,
      "loss": 1.387,
      "num_input_tokens_seen": 157343056,
      "step": 1720
    },
    {
      "epoch": 0.4383561643835616,
      "grad_norm": 0.24503816664218903,
      "learning_rate": 2.5e-05,
      "loss": 1.3544,
      "num_input_tokens_seen": 158211084,
      "step": 1730
    },
    {
      "epoch": 0.4408900150447383,
      "grad_norm": 0.23860155045986176,
      "learning_rate": 2.5e-05,
      "loss": 1.3947,
      "num_input_tokens_seen": 159127644,
      "step": 1740
    },
    {
      "epoch": 0.44342386570591497,
      "grad_norm": 0.23359131813049316,
      "learning_rate": 2.5e-05,
      "loss": 1.3333,
      "num_input_tokens_seen": 160056144,
      "step": 1750
    },
    {
      "epoch": 0.44595771636709164,
      "grad_norm": 0.23289762437343597,
      "learning_rate": 2.5e-05,
      "loss": 1.4039,
      "num_input_tokens_seen": 161001352,
      "step": 1760
    },
    {
      "epoch": 0.44849156702826826,
      "grad_norm": 0.23038776218891144,
      "learning_rate": 2.5e-05,
      "loss": 1.3872,
      "num_input_tokens_seen": 161931048,
      "step": 1770
    },
    {
      "epoch": 0.45102541768944493,
      "grad_norm": 0.26440566778182983,
      "learning_rate": 2.5e-05,
      "loss": 1.372,
      "num_input_tokens_seen": 162861292,
      "step": 1780
    },
    {
      "epoch": 0.4535592683506216,
      "grad_norm": 0.2498098909854889,
      "learning_rate": 2.5e-05,
      "loss": 1.3287,
      "num_input_tokens_seen": 163797388,
      "step": 1790
    },
    {
      "epoch": 0.4560931190117982,
      "grad_norm": 0.2095261961221695,
      "learning_rate": 2.5e-05,
      "loss": 1.3778,
      "num_input_tokens_seen": 164671840,
      "step": 1800
    },
    {
      "epoch": 0.4586269696729749,
      "grad_norm": 0.2577464282512665,
      "learning_rate": 2.5e-05,
      "loss": 1.3821,
      "num_input_tokens_seen": 165619284,
      "step": 1810
    },
    {
      "epoch": 0.46116082033415157,
      "grad_norm": 0.23324383795261383,
      "learning_rate": 2.5e-05,
      "loss": 1.3921,
      "num_input_tokens_seen": 166521872,
      "step": 1820
    },
    {
      "epoch": 0.46369467099532824,
      "grad_norm": 0.23413369059562683,
      "learning_rate": 2.5e-05,
      "loss": 1.391,
      "num_input_tokens_seen": 167446436,
      "step": 1830
    },
    {
      "epoch": 0.46622852165650486,
      "grad_norm": 0.2720430791378021,
      "learning_rate": 2.5e-05,
      "loss": 1.359,
      "num_input_tokens_seen": 168356260,
      "step": 1840
    },
    {
      "epoch": 0.46876237231768153,
      "grad_norm": 0.2760706841945648,
      "learning_rate": 2.5e-05,
      "loss": 1.3498,
      "num_input_tokens_seen": 169262844,
      "step": 1850
    },
    {
      "epoch": 0.4712962229788582,
      "grad_norm": 0.27992355823516846,
      "learning_rate": 2.5e-05,
      "loss": 1.3984,
      "num_input_tokens_seen": 170164272,
      "step": 1860
    },
    {
      "epoch": 0.4738300736400348,
      "grad_norm": 0.23402582108974457,
      "learning_rate": 2.5e-05,
      "loss": 1.3667,
      "num_input_tokens_seen": 171067864,
      "step": 1870
    },
    {
      "epoch": 0.4763639243012115,
      "grad_norm": 0.29928284883499146,
      "learning_rate": 2.5e-05,
      "loss": 1.335,
      "num_input_tokens_seen": 172005232,
      "step": 1880
    },
    {
      "epoch": 0.47889777496238817,
      "grad_norm": 0.25357866287231445,
      "learning_rate": 2.5e-05,
      "loss": 1.3802,
      "num_input_tokens_seen": 172915708,
      "step": 1890
    },
    {
      "epoch": 0.4814316256235648,
      "grad_norm": 0.29246291518211365,
      "learning_rate": 2.5e-05,
      "loss": 1.3513,
      "num_input_tokens_seen": 173820476,
      "step": 1900
    },
    {
      "epoch": 0.48396547628474146,
      "grad_norm": 0.2792080342769623,
      "learning_rate": 2.5e-05,
      "loss": 1.3939,
      "num_input_tokens_seen": 174740920,
      "step": 1910
    },
    {
      "epoch": 0.48649932694591813,
      "grad_norm": 0.3099055588245392,
      "learning_rate": 2.5e-05,
      "loss": 1.3693,
      "num_input_tokens_seen": 175635720,
      "step": 1920
    },
    {
      "epoch": 0.4890331776070948,
      "grad_norm": 0.2375776320695877,
      "learning_rate": 2.5e-05,
      "loss": 1.3829,
      "num_input_tokens_seen": 176538688,
      "step": 1930
    },
    {
      "epoch": 0.4915670282682714,
      "grad_norm": 0.2295093983411789,
      "learning_rate": 2.5e-05,
      "loss": 1.3691,
      "num_input_tokens_seen": 177468420,
      "step": 1940
    },
    {
      "epoch": 0.4941008789294481,
      "grad_norm": 0.21639369428157806,
      "learning_rate": 2.5e-05,
      "loss": 1.3509,
      "num_input_tokens_seen": 178388296,
      "step": 1950
    },
    {
      "epoch": 0.49663472959062477,
      "grad_norm": 0.26756080985069275,
      "learning_rate": 2.5e-05,
      "loss": 1.3761,
      "num_input_tokens_seen": 179341380,
      "step": 1960
    },
    {
      "epoch": 0.4991685802518014,
      "grad_norm": 0.21319729089736938,
      "learning_rate": 2.5e-05,
      "loss": 1.3803,
      "num_input_tokens_seen": 180256564,
      "step": 1970
    },
    {
      "epoch": 0.5017024309129781,
      "grad_norm": 0.2565974295139313,
      "learning_rate": 2.5e-05,
      "loss": 1.3763,
      "num_input_tokens_seen": 181117020,
      "step": 1980
    },
    {
      "epoch": 0.5042362815741547,
      "grad_norm": 0.30257830023765564,
      "learning_rate": 2.5e-05,
      "loss": 1.3671,
      "num_input_tokens_seen": 182027528,
      "step": 1990
    },
    {
      "epoch": 0.5067701322353314,
      "grad_norm": 0.23474013805389404,
      "learning_rate": 2.5e-05,
      "loss": 1.3456,
      "num_input_tokens_seen": 182939052,
      "step": 2000
    },
    {
      "epoch": 0.5067701322353314,
      "eval_loss": 1.3669419288635254,
      "eval_runtime": 2.8409,
      "eval_samples_per_second": 52.801,
      "eval_steps_per_second": 6.688,
      "num_input_tokens_seen": 182939052,
      "step": 2000
    },
    {
      "epoch": 0.509303982896508,
      "grad_norm": 0.2144283950328827,
      "learning_rate": 2.5e-05,
      "loss": 1.37,
      "num_input_tokens_seen": 183841188,
      "step": 2010
    },
    {
      "epoch": 0.5118378335576846,
      "grad_norm": 0.2299591451883316,
      "learning_rate": 2.5e-05,
      "loss": 1.3436,
      "num_input_tokens_seen": 184804372,
      "step": 2020
    },
    {
      "epoch": 0.5143716842188614,
      "grad_norm": 0.2291470170021057,
      "learning_rate": 2.5e-05,
      "loss": 1.38,
      "num_input_tokens_seen": 185696628,
      "step": 2030
    },
    {
      "epoch": 0.516905534880038,
      "grad_norm": 0.25624164938926697,
      "learning_rate": 2.5e-05,
      "loss": 1.3741,
      "num_input_tokens_seen": 186584108,
      "step": 2040
    },
    {
      "epoch": 0.5194393855412147,
      "grad_norm": 0.2826102077960968,
      "learning_rate": 2.5e-05,
      "loss": 1.3786,
      "num_input_tokens_seen": 187491532,
      "step": 2050
    },
    {
      "epoch": 0.5219732362023913,
      "grad_norm": 0.23644354939460754,
      "learning_rate": 2.5e-05,
      "loss": 1.3119,
      "num_input_tokens_seen": 188398308,
      "step": 2060
    },
    {
      "epoch": 0.524507086863568,
      "grad_norm": 0.2631579041481018,
      "learning_rate": 2.5e-05,
      "loss": 1.3596,
      "num_input_tokens_seen": 189270772,
      "step": 2070
    },
    {
      "epoch": 0.5270409375247447,
      "grad_norm": 0.24663548171520233,
      "learning_rate": 2.5e-05,
      "loss": 1.3833,
      "num_input_tokens_seen": 190188192,
      "step": 2080
    },
    {
      "epoch": 0.5295747881859213,
      "grad_norm": 0.21753673255443573,
      "learning_rate": 2.5e-05,
      "loss": 1.3746,
      "num_input_tokens_seen": 191125784,
      "step": 2090
    },
    {
      "epoch": 0.5321086388470979,
      "grad_norm": 0.2312672883272171,
      "learning_rate": 2.5e-05,
      "loss": 1.3558,
      "num_input_tokens_seen": 192010984,
      "step": 2100
    },
    {
      "epoch": 0.5346424895082746,
      "grad_norm": 0.2641030251979828,
      "learning_rate": 2.5e-05,
      "loss": 1.3436,
      "num_input_tokens_seen": 192947832,
      "step": 2110
    },
    {
      "epoch": 0.5371763401694513,
      "grad_norm": 0.2314285784959793,
      "learning_rate": 2.5e-05,
      "loss": 1.3889,
      "num_input_tokens_seen": 193836096,
      "step": 2120
    },
    {
      "epoch": 0.5397101908306279,
      "grad_norm": 0.2117050439119339,
      "learning_rate": 2.5e-05,
      "loss": 1.3636,
      "num_input_tokens_seen": 194752188,
      "step": 2130
    },
    {
      "epoch": 0.5422440414918046,
      "grad_norm": 0.24790892004966736,
      "learning_rate": 2.5e-05,
      "loss": 1.3577,
      "num_input_tokens_seen": 195659416,
      "step": 2140
    },
    {
      "epoch": 0.5447778921529812,
      "grad_norm": 0.253757119178772,
      "learning_rate": 2.5e-05,
      "loss": 1.3767,
      "num_input_tokens_seen": 196584176,
      "step": 2150
    },
    {
      "epoch": 0.5473117428141578,
      "grad_norm": 0.2629224359989166,
      "learning_rate": 2.5e-05,
      "loss": 1.3771,
      "num_input_tokens_seen": 197456816,
      "step": 2160
    },
    {
      "epoch": 0.5498455934753346,
      "grad_norm": 0.2274072915315628,
      "learning_rate": 2.5e-05,
      "loss": 1.3633,
      "num_input_tokens_seen": 198358444,
      "step": 2170
    },
    {
      "epoch": 0.5523794441365112,
      "grad_norm": 0.2630630135536194,
      "learning_rate": 2.5e-05,
      "loss": 1.3702,
      "num_input_tokens_seen": 199246040,
      "step": 2180
    },
    {
      "epoch": 0.5549132947976878,
      "grad_norm": 0.24167053401470184,
      "learning_rate": 2.5e-05,
      "loss": 1.3785,
      "num_input_tokens_seen": 200167412,
      "step": 2190
    },
    {
      "epoch": 0.5574471454588645,
      "grad_norm": 0.2560918927192688,
      "learning_rate": 2.5e-05,
      "loss": 1.3757,
      "num_input_tokens_seen": 201090512,
      "step": 2200
    },
    {
      "epoch": 0.5599809961200412,
      "grad_norm": 0.23884332180023193,
      "learning_rate": 2.5e-05,
      "loss": 1.3642,
      "num_input_tokens_seen": 202070196,
      "step": 2210
    },
    {
      "epoch": 0.5625148467812179,
      "grad_norm": 0.25141972303390503,
      "learning_rate": 2.5e-05,
      "loss": 1.3669,
      "num_input_tokens_seen": 203015232,
      "step": 2220
    },
    {
      "epoch": 0.5650486974423945,
      "grad_norm": 0.20563028752803802,
      "learning_rate": 2.5e-05,
      "loss": 1.3622,
      "num_input_tokens_seen": 203955992,
      "step": 2230
    },
    {
      "epoch": 0.5675825481035711,
      "grad_norm": 0.26771050691604614,
      "learning_rate": 2.5e-05,
      "loss": 1.3551,
      "num_input_tokens_seen": 204867084,
      "step": 2240
    },
    {
      "epoch": 0.5701163987647478,
      "grad_norm": 0.2185191512107849,
      "learning_rate": 2.5e-05,
      "loss": 1.3574,
      "num_input_tokens_seen": 205818444,
      "step": 2250
    },
    {
      "epoch": 0.5726502494259245,
      "grad_norm": 0.23736274242401123,
      "learning_rate": 2.5e-05,
      "loss": 1.3478,
      "num_input_tokens_seen": 206727340,
      "step": 2260
    },
    {
      "epoch": 0.5751841000871011,
      "grad_norm": 0.2208438366651535,
      "learning_rate": 2.5e-05,
      "loss": 1.3576,
      "num_input_tokens_seen": 207682956,
      "step": 2270
    },
    {
      "epoch": 0.5777179507482778,
      "grad_norm": 0.215751051902771,
      "learning_rate": 2.5e-05,
      "loss": 1.3105,
      "num_input_tokens_seen": 208613224,
      "step": 2280
    },
    {
      "epoch": 0.5802518014094544,
      "grad_norm": 0.24414047598838806,
      "learning_rate": 2.5e-05,
      "loss": 1.3637,
      "num_input_tokens_seen": 209480700,
      "step": 2290
    },
    {
      "epoch": 0.582785652070631,
      "grad_norm": 0.27234476804733276,
      "learning_rate": 2.5e-05,
      "loss": 1.3648,
      "num_input_tokens_seen": 210380616,
      "step": 2300
    },
    {
      "epoch": 0.5853195027318078,
      "grad_norm": 0.23880694806575775,
      "learning_rate": 2.5e-05,
      "loss": 1.3452,
      "num_input_tokens_seen": 211323472,
      "step": 2310
    },
    {
      "epoch": 0.5878533533929844,
      "grad_norm": 0.24618738889694214,
      "learning_rate": 2.5e-05,
      "loss": 1.3357,
      "num_input_tokens_seen": 212269424,
      "step": 2320
    },
    {
      "epoch": 0.590387204054161,
      "grad_norm": 0.2280731499195099,
      "learning_rate": 2.5e-05,
      "loss": 1.3808,
      "num_input_tokens_seen": 213236052,
      "step": 2330
    },
    {
      "epoch": 0.5929210547153377,
      "grad_norm": 0.2641889452934265,
      "learning_rate": 2.5e-05,
      "loss": 1.3635,
      "num_input_tokens_seen": 214193180,
      "step": 2340
    },
    {
      "epoch": 0.5954549053765144,
      "grad_norm": 0.24398839473724365,
      "learning_rate": 2.5e-05,
      "loss": 1.3157,
      "num_input_tokens_seen": 215145888,
      "step": 2350
    },
    {
      "epoch": 0.5979887560376911,
      "grad_norm": 0.29194214940071106,
      "learning_rate": 2.5e-05,
      "loss": 1.3809,
      "num_input_tokens_seen": 216076328,
      "step": 2360
    },
    {
      "epoch": 0.6005226066988677,
      "grad_norm": 0.23668240010738373,
      "learning_rate": 2.5e-05,
      "loss": 1.3723,
      "num_input_tokens_seen": 216957792,
      "step": 2370
    },
    {
      "epoch": 0.6030564573600443,
      "grad_norm": 0.2053728848695755,
      "learning_rate": 2.5e-05,
      "loss": 1.3106,
      "num_input_tokens_seen": 217923088,
      "step": 2380
    },
    {
      "epoch": 0.605590308021221,
      "grad_norm": 0.2571648061275482,
      "learning_rate": 2.5e-05,
      "loss": 1.3636,
      "num_input_tokens_seen": 218831976,
      "step": 2390
    },
    {
      "epoch": 0.6081241586823977,
      "grad_norm": 0.25352680683135986,
      "learning_rate": 2.5e-05,
      "loss": 1.3448,
      "num_input_tokens_seen": 219756636,
      "step": 2400
    },
    {
      "epoch": 0.6106580093435743,
      "grad_norm": 0.23342467844486237,
      "learning_rate": 2.5e-05,
      "loss": 1.3908,
      "num_input_tokens_seen": 220660172,
      "step": 2410
    },
    {
      "epoch": 0.613191860004751,
      "grad_norm": 0.24378784000873566,
      "learning_rate": 2.5e-05,
      "loss": 1.3631,
      "num_input_tokens_seen": 221559444,
      "step": 2420
    },
    {
      "epoch": 0.6157257106659276,
      "grad_norm": 0.23902441561222076,
      "learning_rate": 2.5e-05,
      "loss": 1.3389,
      "num_input_tokens_seen": 222484304,
      "step": 2430
    },
    {
      "epoch": 0.6182595613271042,
      "grad_norm": 0.24430356919765472,
      "learning_rate": 2.5e-05,
      "loss": 1.3741,
      "num_input_tokens_seen": 223424636,
      "step": 2440
    },
    {
      "epoch": 0.620793411988281,
      "grad_norm": 0.22024385631084442,
      "learning_rate": 2.5e-05,
      "loss": 1.3173,
      "num_input_tokens_seen": 224336328,
      "step": 2450
    },
    {
      "epoch": 0.6233272626494576,
      "grad_norm": 0.2540358304977417,
      "learning_rate": 2.5e-05,
      "loss": 1.3551,
      "num_input_tokens_seen": 225268812,
      "step": 2460
    },
    {
      "epoch": 0.6258611133106342,
      "grad_norm": 0.30823466181755066,
      "learning_rate": 2.5e-05,
      "loss": 1.3315,
      "num_input_tokens_seen": 226203392,
      "step": 2470
    },
    {
      "epoch": 0.6283949639718109,
      "grad_norm": 0.22996842861175537,
      "learning_rate": 2.5e-05,
      "loss": 1.3647,
      "num_input_tokens_seen": 227073928,
      "step": 2480
    },
    {
      "epoch": 0.6309288146329876,
      "grad_norm": 0.22297543287277222,
      "learning_rate": 2.5e-05,
      "loss": 1.3673,
      "num_input_tokens_seen": 227988144,
      "step": 2490
    },
    {
      "epoch": 0.6334626652941642,
      "grad_norm": 0.2600548267364502,
      "learning_rate": 2.5e-05,
      "loss": 1.3417,
      "num_input_tokens_seen": 228908304,
      "step": 2500
    },
    {
      "epoch": 0.6359965159553409,
      "grad_norm": 0.27056604623794556,
      "learning_rate": 2.5e-05,
      "loss": 1.2998,
      "num_input_tokens_seen": 229859596,
      "step": 2510
    },
    {
      "epoch": 0.6385303666165175,
      "grad_norm": 0.22515636682510376,
      "learning_rate": 2.5e-05,
      "loss": 1.3605,
      "num_input_tokens_seen": 230760960,
      "step": 2520
    },
    {
      "epoch": 0.6410642172776942,
      "grad_norm": 0.33911067247390747,
      "learning_rate": 2.5e-05,
      "loss": 1.3648,
      "num_input_tokens_seen": 231683832,
      "step": 2530
    },
    {
      "epoch": 0.6435980679388709,
      "grad_norm": 0.2713491916656494,
      "learning_rate": 2.5e-05,
      "loss": 1.3581,
      "num_input_tokens_seen": 232586192,
      "step": 2540
    },
    {
      "epoch": 0.6461319186000475,
      "grad_norm": 0.22554545104503632,
      "learning_rate": 2.5e-05,
      "loss": 1.3217,
      "num_input_tokens_seen": 233513620,
      "step": 2550
    },
    {
      "epoch": 0.6486657692612242,
      "grad_norm": 0.23459571599960327,
      "learning_rate": 2.5e-05,
      "loss": 1.3185,
      "num_input_tokens_seen": 234405628,
      "step": 2560
    },
    {
      "epoch": 0.6511996199224008,
      "grad_norm": 0.22022689878940582,
      "learning_rate": 2.5e-05,
      "loss": 1.3724,
      "num_input_tokens_seen": 235287208,
      "step": 2570
    },
    {
      "epoch": 0.6537334705835774,
      "grad_norm": 0.2207019031047821,
      "learning_rate": 2.5e-05,
      "loss": 1.3871,
      "num_input_tokens_seen": 236206532,
      "step": 2580
    },
    {
      "epoch": 0.6562673212447542,
      "grad_norm": 0.286006897687912,
      "learning_rate": 2.5e-05,
      "loss": 1.338,
      "num_input_tokens_seen": 237132236,
      "step": 2590
    },
    {
      "epoch": 0.6588011719059308,
      "grad_norm": 0.24479633569717407,
      "learning_rate": 2.5e-05,
      "loss": 1.3636,
      "num_input_tokens_seen": 238036544,
      "step": 2600
    },
    {
      "epoch": 0.6613350225671074,
      "grad_norm": 0.21694402396678925,
      "learning_rate": 2.5e-05,
      "loss": 1.3711,
      "num_input_tokens_seen": 238978380,
      "step": 2610
    },
    {
      "epoch": 0.6638688732282841,
      "grad_norm": 0.22491593658924103,
      "learning_rate": 2.5e-05,
      "loss": 1.3516,
      "num_input_tokens_seen": 239893524,
      "step": 2620
    },
    {
      "epoch": 0.6664027238894608,
      "grad_norm": 0.24287302792072296,
      "learning_rate": 2.5e-05,
      "loss": 1.3333,
      "num_input_tokens_seen": 240753560,
      "step": 2630
    },
    {
      "epoch": 0.6689365745506374,
      "grad_norm": 0.24059581756591797,
      "learning_rate": 2.5e-05,
      "loss": 1.3172,
      "num_input_tokens_seen": 241689616,
      "step": 2640
    },
    {
      "epoch": 0.6714704252118141,
      "grad_norm": 0.24688631296157837,
      "learning_rate": 2.5e-05,
      "loss": 1.3377,
      "num_input_tokens_seen": 242618896,
      "step": 2650
    },
    {
      "epoch": 0.6740042758729907,
      "grad_norm": 0.2412404716014862,
      "learning_rate": 2.5e-05,
      "loss": 1.3512,
      "num_input_tokens_seen": 243555264,
      "step": 2660
    },
    {
      "epoch": 0.6765381265341673,
      "grad_norm": 0.23944397270679474,
      "learning_rate": 2.5e-05,
      "loss": 1.3806,
      "num_input_tokens_seen": 244450244,
      "step": 2670
    },
    {
      "epoch": 0.6790719771953441,
      "grad_norm": 0.24713559448719025,
      "learning_rate": 2.5e-05,
      "loss": 1.3251,
      "num_input_tokens_seen": 245398672,
      "step": 2680
    },
    {
      "epoch": 0.6816058278565207,
      "grad_norm": 0.31667396426200867,
      "learning_rate": 2.5e-05,
      "loss": 1.3642,
      "num_input_tokens_seen": 246320464,
      "step": 2690
    },
    {
      "epoch": 0.6841396785176974,
      "grad_norm": 0.250383585691452,
      "learning_rate": 2.5e-05,
      "loss": 1.3329,
      "num_input_tokens_seen": 247248308,
      "step": 2700
    },
    {
      "epoch": 0.686673529178874,
      "grad_norm": 0.2263907939195633,
      "learning_rate": 2.5e-05,
      "loss": 1.3281,
      "num_input_tokens_seen": 248202884,
      "step": 2710
    },
    {
      "epoch": 0.6892073798400506,
      "grad_norm": 0.24522219598293304,
      "learning_rate": 2.5e-05,
      "loss": 1.3477,
      "num_input_tokens_seen": 249166112,
      "step": 2720
    },
    {
      "epoch": 0.6917412305012274,
      "grad_norm": 0.22159820795059204,
      "learning_rate": 2.5e-05,
      "loss": 1.3192,
      "num_input_tokens_seen": 250077904,
      "step": 2730
    },
    {
      "epoch": 0.694275081162404,
      "grad_norm": 0.2300739735364914,
      "learning_rate": 2.5e-05,
      "loss": 1.3336,
      "num_input_tokens_seen": 251012120,
      "step": 2740
    },
    {
      "epoch": 0.6968089318235806,
      "grad_norm": 0.22758354246616364,
      "learning_rate": 2.5e-05,
      "loss": 1.3964,
      "num_input_tokens_seen": 251934920,
      "step": 2750
    },
    {
      "epoch": 0.6993427824847573,
      "grad_norm": 0.2598190903663635,
      "learning_rate": 2.5e-05,
      "loss": 1.3311,
      "num_input_tokens_seen": 252877580,
      "step": 2760
    },
    {
      "epoch": 0.701876633145934,
      "grad_norm": 0.23178431391716003,
      "learning_rate": 2.5e-05,
      "loss": 1.3453,
      "num_input_tokens_seen": 253792028,
      "step": 2770
    },
    {
      "epoch": 0.7044104838071106,
      "grad_norm": 0.26508447527885437,
      "learning_rate": 2.5e-05,
      "loss": 1.3635,
      "num_input_tokens_seen": 254742856,
      "step": 2780
    },
    {
      "epoch": 0.7069443344682873,
      "grad_norm": 0.263509601354599,
      "learning_rate": 2.5e-05,
      "loss": 1.3584,
      "num_input_tokens_seen": 255676980,
      "step": 2790
    },
    {
      "epoch": 0.7094781851294639,
      "grad_norm": 0.25076207518577576,
      "learning_rate": 2.5e-05,
      "loss": 1.3654,
      "num_input_tokens_seen": 256607480,
      "step": 2800
    },
    {
      "epoch": 0.7120120357906405,
      "grad_norm": 0.3114246726036072,
      "learning_rate": 2.5e-05,
      "loss": 1.3626,
      "num_input_tokens_seen": 257486156,
      "step": 2810
    },
    {
      "epoch": 0.7145458864518173,
      "grad_norm": 0.2184561789035797,
      "learning_rate": 2.5e-05,
      "loss": 1.3481,
      "num_input_tokens_seen": 258406168,
      "step": 2820
    },
    {
      "epoch": 0.7170797371129939,
      "grad_norm": 0.27279725670814514,
      "learning_rate": 2.5e-05,
      "loss": 1.3358,
      "num_input_tokens_seen": 259298936,
      "step": 2830
    },
    {
      "epoch": 0.7196135877741705,
      "grad_norm": 0.23473051190376282,
      "learning_rate": 2.5e-05,
      "loss": 1.3157,
      "num_input_tokens_seen": 260214884,
      "step": 2840
    },
    {
      "epoch": 0.7221474384353472,
      "grad_norm": 0.2273094654083252,
      "learning_rate": 2.5e-05,
      "loss": 1.3695,
      "num_input_tokens_seen": 261150656,
      "step": 2850
    },
    {
      "epoch": 0.7246812890965239,
      "grad_norm": 0.23328402638435364,
      "learning_rate": 2.5e-05,
      "loss": 1.3491,
      "num_input_tokens_seen": 262090748,
      "step": 2860
    },
    {
      "epoch": 0.7272151397577006,
      "grad_norm": 0.27058523893356323,
      "learning_rate": 2.5e-05,
      "loss": 1.3164,
      "num_input_tokens_seen": 263047956,
      "step": 2870
    },
    {
      "epoch": 0.7297489904188772,
      "grad_norm": 0.26919999718666077,
      "learning_rate": 2.5e-05,
      "loss": 1.3429,
      "num_input_tokens_seen": 263952708,
      "step": 2880
    },
    {
      "epoch": 0.7322828410800538,
      "grad_norm": 0.2629719078540802,
      "learning_rate": 2.5e-05,
      "loss": 1.3736,
      "num_input_tokens_seen": 264850904,
      "step": 2890
    },
    {
      "epoch": 0.7348166917412305,
      "grad_norm": 0.2600915729999542,
      "learning_rate": 2.5e-05,
      "loss": 1.3179,
      "num_input_tokens_seen": 265795528,
      "step": 2900
    },
    {
      "epoch": 0.7373505424024072,
      "grad_norm": 0.29251357913017273,
      "learning_rate": 2.5e-05,
      "loss": 1.3671,
      "num_input_tokens_seen": 266703240,
      "step": 2910
    },
    {
      "epoch": 0.7398843930635838,
      "grad_norm": 0.23803594708442688,
      "learning_rate": 2.5e-05,
      "loss": 1.3632,
      "num_input_tokens_seen": 267637720,
      "step": 2920
    },
    {
      "epoch": 0.7424182437247605,
      "grad_norm": 0.24492381513118744,
      "learning_rate": 2.5e-05,
      "loss": 1.3275,
      "num_input_tokens_seen": 268547588,
      "step": 2930
    },
    {
      "epoch": 0.7449520943859371,
      "grad_norm": 0.2277376800775528,
      "learning_rate": 2.5e-05,
      "loss": 1.3058,
      "num_input_tokens_seen": 269503056,
      "step": 2940
    },
    {
      "epoch": 0.7474859450471137,
      "grad_norm": 0.22645527124404907,
      "learning_rate": 2.5e-05,
      "loss": 1.3462,
      "num_input_tokens_seen": 270372524,
      "step": 2950
    },
    {
      "epoch": 0.7500197957082905,
      "grad_norm": 0.27738144993782043,
      "learning_rate": 2.5e-05,
      "loss": 1.2953,
      "num_input_tokens_seen": 271255520,
      "step": 2960
    },
    {
      "epoch": 0.7525536463694671,
      "grad_norm": 0.2460719496011734,
      "learning_rate": 2.5e-05,
      "loss": 1.3291,
      "num_input_tokens_seen": 272173512,
      "step": 2970
    },
    {
      "epoch": 0.7550874970306437,
      "grad_norm": 0.23774035274982452,
      "learning_rate": 2.5e-05,
      "loss": 1.3105,
      "num_input_tokens_seen": 273082396,
      "step": 2980
    },
    {
      "epoch": 0.7576213476918204,
      "grad_norm": 0.2344847470521927,
      "learning_rate": 2.5e-05,
      "loss": 1.3379,
      "num_input_tokens_seen": 273951600,
      "step": 2990
    },
    {
      "epoch": 0.760155198352997,
      "grad_norm": 0.2422836273908615,
      "learning_rate": 2.5e-05,
      "loss": 1.3437,
      "num_input_tokens_seen": 274855796,
      "step": 3000
    },
    {
      "epoch": 0.760155198352997,
      "eval_loss": 1.3378311395645142,
      "eval_runtime": 2.7862,
      "eval_samples_per_second": 53.837,
      "eval_steps_per_second": 6.819,
      "num_input_tokens_seen": 274855796,
      "step": 3000
    },
    {
      "epoch": 0.7626890490141738,
      "grad_norm": 0.2418714016675949,
      "learning_rate": 2.5e-05,
      "loss": 1.3683,
      "num_input_tokens_seen": 275793364,
      "step": 3010
    },
    {
      "epoch": 0.7652228996753504,
      "grad_norm": 0.2433195561170578,
      "learning_rate": 2.5e-05,
      "loss": 1.3397,
      "num_input_tokens_seen": 276766688,
      "step": 3020
    },
    {
      "epoch": 0.767756750336527,
      "grad_norm": 0.2531881034374237,
      "learning_rate": 2.5e-05,
      "loss": 1.3069,
      "num_input_tokens_seen": 277692944,
      "step": 3030
    },
    {
      "epoch": 0.7702906009977037,
      "grad_norm": 0.228854700922966,
      "learning_rate": 2.5e-05,
      "loss": 1.3467,
      "num_input_tokens_seen": 278633648,
      "step": 3040
    },
    {
      "epoch": 0.7728244516588804,
      "grad_norm": 0.21645446121692657,
      "learning_rate": 2.5e-05,
      "loss": 1.2949,
      "num_input_tokens_seen": 279542668,
      "step": 3050
    },
    {
      "epoch": 0.775358302320057,
      "grad_norm": 0.2668648362159729,
      "learning_rate": 2.5e-05,
      "loss": 1.3272,
      "num_input_tokens_seen": 280474528,
      "step": 3060
    },
    {
      "epoch": 0.7778921529812337,
      "grad_norm": 0.26199036836624146,
      "learning_rate": 2.5e-05,
      "loss": 1.3395,
      "num_input_tokens_seen": 281383776,
      "step": 3070
    },
    {
      "epoch": 0.7804260036424103,
      "grad_norm": 0.23948872089385986,
      "learning_rate": 2.5e-05,
      "loss": 1.3534,
      "num_input_tokens_seen": 282297260,
      "step": 3080
    },
    {
      "epoch": 0.7829598543035869,
      "grad_norm": 0.2561713755130768,
      "learning_rate": 2.5e-05,
      "loss": 1.3251,
      "num_input_tokens_seen": 283169516,
      "step": 3090
    },
    {
      "epoch": 0.7854937049647637,
      "grad_norm": 0.26099705696105957,
      "learning_rate": 2.5e-05,
      "loss": 1.3394,
      "num_input_tokens_seen": 284109700,
      "step": 3100
    },
    {
      "epoch": 0.7880275556259403,
      "grad_norm": 0.23930218815803528,
      "learning_rate": 2.5e-05,
      "loss": 1.3242,
      "num_input_tokens_seen": 285031264,
      "step": 3110
    },
    {
      "epoch": 0.7905614062871169,
      "grad_norm": 0.23478297889232635,
      "learning_rate": 2.5e-05,
      "loss": 1.3647,
      "num_input_tokens_seen": 285943620,
      "step": 3120
    },
    {
      "epoch": 0.7930952569482936,
      "grad_norm": 0.24018226563930511,
      "learning_rate": 2.5e-05,
      "loss": 1.3166,
      "num_input_tokens_seen": 286819840,
      "step": 3130
    },
    {
      "epoch": 0.7956291076094703,
      "grad_norm": 0.22437995672225952,
      "learning_rate": 2.5e-05,
      "loss": 1.3418,
      "num_input_tokens_seen": 287731640,
      "step": 3140
    },
    {
      "epoch": 0.7981629582706469,
      "grad_norm": 0.2912137806415558,
      "learning_rate": 2.5e-05,
      "loss": 1.3336,
      "num_input_tokens_seen": 288650768,
      "step": 3150
    },
    {
      "epoch": 0.8006968089318236,
      "grad_norm": 0.27003979682922363,
      "learning_rate": 2.5e-05,
      "loss": 1.3094,
      "num_input_tokens_seen": 289579424,
      "step": 3160
    },
    {
      "epoch": 0.8032306595930002,
      "grad_norm": 0.24906513094902039,
      "learning_rate": 2.5e-05,
      "loss": 1.3089,
      "num_input_tokens_seen": 290506080,
      "step": 3170
    },
    {
      "epoch": 0.805764510254177,
      "grad_norm": 0.2620064616203308,
      "learning_rate": 2.5e-05,
      "loss": 1.3741,
      "num_input_tokens_seen": 291447632,
      "step": 3180
    },
    {
      "epoch": 0.8082983609153536,
      "grad_norm": 0.22881096601486206,
      "learning_rate": 2.5e-05,
      "loss": 1.3601,
      "num_input_tokens_seen": 292382736,
      "step": 3190
    },
    {
      "epoch": 0.8108322115765302,
      "grad_norm": 0.23649707436561584,
      "learning_rate": 2.5e-05,
      "loss": 1.3212,
      "num_input_tokens_seen": 293339376,
      "step": 3200
    },
    {
      "epoch": 0.8133660622377069,
      "grad_norm": 0.22773633897304535,
      "learning_rate": 2.5e-05,
      "loss": 1.3124,
      "num_input_tokens_seen": 294273900,
      "step": 3210
    },
    {
      "epoch": 0.8158999128988835,
      "grad_norm": 0.23439520597457886,
      "learning_rate": 2.5e-05,
      "loss": 1.3104,
      "num_input_tokens_seen": 295167620,
      "step": 3220
    },
    {
      "epoch": 0.8184337635600601,
      "grad_norm": 0.2587607800960541,
      "learning_rate": 2.5e-05,
      "loss": 1.3378,
      "num_input_tokens_seen": 296070252,
      "step": 3230
    },
    {
      "epoch": 0.8209676142212369,
      "grad_norm": 0.2375950813293457,
      "learning_rate": 2.5e-05,
      "loss": 1.3608,
      "num_input_tokens_seen": 296964880,
      "step": 3240
    },
    {
      "epoch": 0.8235014648824135,
      "grad_norm": 0.217642143368721,
      "learning_rate": 2.5e-05,
      "loss": 1.3711,
      "num_input_tokens_seen": 297861584,
      "step": 3250
    },
    {
      "epoch": 0.8260353155435901,
      "grad_norm": 0.24903365969657898,
      "learning_rate": 2.5e-05,
      "loss": 1.3759,
      "num_input_tokens_seen": 298763600,
      "step": 3260
    },
    {
      "epoch": 0.8285691662047668,
      "grad_norm": 0.25492629408836365,
      "learning_rate": 2.5e-05,
      "loss": 1.336,
      "num_input_tokens_seen": 299655852,
      "step": 3270
    },
    {
      "epoch": 0.8311030168659435,
      "grad_norm": 0.26514139771461487,
      "learning_rate": 2.5e-05,
      "loss": 1.3294,
      "num_input_tokens_seen": 300539872,
      "step": 3280
    },
    {
      "epoch": 0.8336368675271201,
      "grad_norm": 0.23889844119548798,
      "learning_rate": 2.5e-05,
      "loss": 1.3845,
      "num_input_tokens_seen": 301433356,
      "step": 3290
    },
    {
      "epoch": 0.8361707181882968,
      "grad_norm": 0.23075729608535767,
      "learning_rate": 2.5e-05,
      "loss": 1.3359,
      "num_input_tokens_seen": 302358284,
      "step": 3300
    },
    {
      "epoch": 0.8387045688494734,
      "grad_norm": 0.28124797344207764,
      "learning_rate": 2.5e-05,
      "loss": 1.3663,
      "num_input_tokens_seen": 303293764,
      "step": 3310
    },
    {
      "epoch": 0.84123841951065,
      "grad_norm": 0.30670827627182007,
      "learning_rate": 2.5e-05,
      "loss": 1.335,
      "num_input_tokens_seen": 304171336,
      "step": 3320
    },
    {
      "epoch": 0.8437722701718268,
      "grad_norm": 0.22578497231006622,
      "learning_rate": 2.5e-05,
      "loss": 1.327,
      "num_input_tokens_seen": 305091264,
      "step": 3330
    },
    {
      "epoch": 0.8463061208330034,
      "grad_norm": 0.22120265662670135,
      "learning_rate": 2.5e-05,
      "loss": 1.3509,
      "num_input_tokens_seen": 306010588,
      "step": 3340
    },
    {
      "epoch": 0.8488399714941801,
      "grad_norm": 0.2477473020553589,
      "learning_rate": 2.5e-05,
      "loss": 1.3565,
      "num_input_tokens_seen": 306940328,
      "step": 3350
    },
    {
      "epoch": 0.8513738221553567,
      "grad_norm": 0.2530181109905243,
      "learning_rate": 2.5e-05,
      "loss": 1.2936,
      "num_input_tokens_seen": 307838056,
      "step": 3360
    },
    {
      "epoch": 0.8539076728165333,
      "grad_norm": 0.2556324303150177,
      "learning_rate": 2.5e-05,
      "loss": 1.3002,
      "num_input_tokens_seen": 308773220,
      "step": 3370
    },
    {
      "epoch": 0.8564415234777101,
      "grad_norm": 0.24870575964450836,
      "learning_rate": 2.5e-05,
      "loss": 1.3086,
      "num_input_tokens_seen": 309713036,
      "step": 3380
    },
    {
      "epoch": 0.8589753741388867,
      "grad_norm": 0.22579419612884521,
      "learning_rate": 2.5e-05,
      "loss": 1.3238,
      "num_input_tokens_seen": 310676544,
      "step": 3390
    },
    {
      "epoch": 0.8615092248000633,
      "grad_norm": 0.26896366477012634,
      "learning_rate": 2.5e-05,
      "loss": 1.3518,
      "num_input_tokens_seen": 311609100,
      "step": 3400
    },
    {
      "epoch": 0.86404307546124,
      "grad_norm": 0.23491699993610382,
      "learning_rate": 2.5e-05,
      "loss": 1.3478,
      "num_input_tokens_seen": 312541060,
      "step": 3410
    },
    {
      "epoch": 0.8665769261224167,
      "grad_norm": 0.21398873627185822,
      "learning_rate": 2.5e-05,
      "loss": 1.317,
      "num_input_tokens_seen": 313464680,
      "step": 3420
    },
    {
      "epoch": 0.8691107767835933,
      "grad_norm": 0.2201145589351654,
      "learning_rate": 2.5e-05,
      "loss": 1.3203,
      "num_input_tokens_seen": 314362092,
      "step": 3430
    },
    {
      "epoch": 0.87164462744477,
      "grad_norm": 0.23937499523162842,
      "learning_rate": 2.5e-05,
      "loss": 1.3594,
      "num_input_tokens_seen": 315286788,
      "step": 3440
    },
    {
      "epoch": 0.8741784781059466,
      "grad_norm": 0.2299693375825882,
      "learning_rate": 2.5e-05,
      "loss": 1.359,
      "num_input_tokens_seen": 316199708,
      "step": 3450
    },
    {
      "epoch": 0.8767123287671232,
      "grad_norm": 0.21679440140724182,
      "learning_rate": 2.5e-05,
      "loss": 1.3372,
      "num_input_tokens_seen": 317082032,
      "step": 3460
    },
    {
      "epoch": 0.8792461794283,
      "grad_norm": 0.23869968950748444,
      "learning_rate": 2.5e-05,
      "loss": 1.2815,
      "num_input_tokens_seen": 317999160,
      "step": 3470
    },
    {
      "epoch": 0.8817800300894766,
      "grad_norm": 0.24342550337314606,
      "learning_rate": 2.5e-05,
      "loss": 1.3222,
      "num_input_tokens_seen": 318945628,
      "step": 3480
    },
    {
      "epoch": 0.8843138807506532,
      "grad_norm": 0.23146317899227142,
      "learning_rate": 2.5e-05,
      "loss": 1.3188,
      "num_input_tokens_seen": 319892264,
      "step": 3490
    },
    {
      "epoch": 0.8868477314118299,
      "grad_norm": 0.27557140588760376,
      "learning_rate": 2.5e-05,
      "loss": 1.3065,
      "num_input_tokens_seen": 320815992,
      "step": 3500
    },
    {
      "epoch": 0.8893815820730065,
      "grad_norm": 0.24911952018737793,
      "learning_rate": 2.5e-05,
      "loss": 1.3275,
      "num_input_tokens_seen": 321703172,
      "step": 3510
    },
    {
      "epoch": 0.8919154327341833,
      "grad_norm": 0.2727194130420685,
      "learning_rate": 2.5e-05,
      "loss": 1.3297,
      "num_input_tokens_seen": 322642588,
      "step": 3520
    },
    {
      "epoch": 0.8944492833953599,
      "grad_norm": 0.242356538772583,
      "learning_rate": 2.5e-05,
      "loss": 1.2881,
      "num_input_tokens_seen": 323529188,
      "step": 3530
    },
    {
      "epoch": 0.8969831340565365,
      "grad_norm": 0.21331574022769928,
      "learning_rate": 2.5e-05,
      "loss": 1.2861,
      "num_input_tokens_seen": 324438984,
      "step": 3540
    },
    {
      "epoch": 0.8995169847177132,
      "grad_norm": 0.28540030121803284,
      "learning_rate": 2.5e-05,
      "loss": 1.3361,
      "num_input_tokens_seen": 325302632,
      "step": 3550
    },
    {
      "epoch": 0.9020508353788899,
      "grad_norm": 0.2721042037010193,
      "learning_rate": 2.5e-05,
      "loss": 1.3377,
      "num_input_tokens_seen": 326223312,
      "step": 3560
    },
    {
      "epoch": 0.9045846860400665,
      "grad_norm": 0.235883429646492,
      "learning_rate": 2.5e-05,
      "loss": 1.3603,
      "num_input_tokens_seen": 327174992,
      "step": 3570
    },
    {
      "epoch": 0.9071185367012432,
      "grad_norm": 0.2746555507183075,
      "learning_rate": 2.5e-05,
      "loss": 1.3497,
      "num_input_tokens_seen": 328087740,
      "step": 3580
    },
    {
      "epoch": 0.9096523873624198,
      "grad_norm": 0.21206247806549072,
      "learning_rate": 2.5e-05,
      "loss": 1.3192,
      "num_input_tokens_seen": 329015496,
      "step": 3590
    },
    {
      "epoch": 0.9121862380235964,
      "grad_norm": 0.24580571055412292,
      "learning_rate": 2.5e-05,
      "loss": 1.2958,
      "num_input_tokens_seen": 329914504,
      "step": 3600
    },
    {
      "epoch": 0.9147200886847732,
      "grad_norm": 0.2298029512166977,
      "learning_rate": 2.5e-05,
      "loss": 1.2955,
      "num_input_tokens_seen": 330861412,
      "step": 3610
    },
    {
      "epoch": 0.9172539393459498,
      "grad_norm": 0.20944957435131073,
      "learning_rate": 2.5e-05,
      "loss": 1.3413,
      "num_input_tokens_seen": 331705132,
      "step": 3620
    },
    {
      "epoch": 0.9197877900071264,
      "grad_norm": 0.26745468378067017,
      "learning_rate": 2.5e-05,
      "loss": 1.3528,
      "num_input_tokens_seen": 332613612,
      "step": 3630
    },
    {
      "epoch": 0.9223216406683031,
      "grad_norm": 0.23441898822784424,
      "learning_rate": 2.5e-05,
      "loss": 1.3125,
      "num_input_tokens_seen": 333546464,
      "step": 3640
    },
    {
      "epoch": 0.9248554913294798,
      "grad_norm": 0.25231051445007324,
      "learning_rate": 2.5e-05,
      "loss": 1.3264,
      "num_input_tokens_seen": 334449860,
      "step": 3650
    },
    {
      "epoch": 0.9273893419906565,
      "grad_norm": 0.22412322461605072,
      "learning_rate": 2.5e-05,
      "loss": 1.3159,
      "num_input_tokens_seen": 335390600,
      "step": 3660
    },
    {
      "epoch": 0.9299231926518331,
      "grad_norm": 0.23513691127300262,
      "learning_rate": 2.5e-05,
      "loss": 1.3115,
      "num_input_tokens_seen": 336327464,
      "step": 3670
    },
    {
      "epoch": 0.9324570433130097,
      "grad_norm": 0.22470693290233612,
      "learning_rate": 2.5e-05,
      "loss": 1.3214,
      "num_input_tokens_seen": 337241700,
      "step": 3680
    },
    {
      "epoch": 0.9349908939741864,
      "grad_norm": 0.24091310799121857,
      "learning_rate": 2.5e-05,
      "loss": 1.3306,
      "num_input_tokens_seen": 338184552,
      "step": 3690
    },
    {
      "epoch": 0.9375247446353631,
      "grad_norm": 0.23601089417934418,
      "learning_rate": 2.5e-05,
      "loss": 1.2856,
      "num_input_tokens_seen": 339109296,
      "step": 3700
    },
    {
      "epoch": 0.9400585952965397,
      "grad_norm": 0.23559744656085968,
      "learning_rate": 2.5e-05,
      "loss": 1.293,
      "num_input_tokens_seen": 340010148,
      "step": 3710
    },
    {
      "epoch": 0.9425924459577164,
      "grad_norm": 0.2477143257856369,
      "learning_rate": 2.5e-05,
      "loss": 1.3226,
      "num_input_tokens_seen": 340905016,
      "step": 3720
    },
    {
      "epoch": 0.945126296618893,
      "grad_norm": 0.2724590599536896,
      "learning_rate": 2.5e-05,
      "loss": 1.3063,
      "num_input_tokens_seen": 341861552,
      "step": 3730
    },
    {
      "epoch": 0.9476601472800696,
      "grad_norm": 0.23112662136554718,
      "learning_rate": 2.5e-05,
      "loss": 1.3099,
      "num_input_tokens_seen": 342806136,
      "step": 3740
    },
    {
      "epoch": 0.9501939979412464,
      "grad_norm": 0.2522134780883789,
      "learning_rate": 2.5e-05,
      "loss": 1.2874,
      "num_input_tokens_seen": 343741672,
      "step": 3750
    },
    {
      "epoch": 0.952727848602423,
      "grad_norm": 0.23056572675704956,
      "learning_rate": 2.5e-05,
      "loss": 1.3069,
      "num_input_tokens_seen": 344641984,
      "step": 3760
    },
    {
      "epoch": 0.9552616992635996,
      "grad_norm": 0.2758452892303467,
      "learning_rate": 2.5e-05,
      "loss": 1.2951,
      "num_input_tokens_seen": 345553040,
      "step": 3770
    },
    {
      "epoch": 0.9577955499247763,
      "grad_norm": 0.2210364043712616,
      "learning_rate": 2.5e-05,
      "loss": 1.288,
      "num_input_tokens_seen": 346455716,
      "step": 3780
    },
    {
      "epoch": 0.960329400585953,
      "grad_norm": 0.24254508316516876,
      "learning_rate": 2.5e-05,
      "loss": 1.3527,
      "num_input_tokens_seen": 347362188,
      "step": 3790
    },
    {
      "epoch": 0.9628632512471296,
      "grad_norm": 0.2317672073841095,
      "learning_rate": 2.5e-05,
      "loss": 1.2872,
      "num_input_tokens_seen": 348323636,
      "step": 3800
    },
    {
      "epoch": 0.9653971019083063,
      "grad_norm": 0.25921356678009033,
      "learning_rate": 2.5e-05,
      "loss": 1.326,
      "num_input_tokens_seen": 349187636,
      "step": 3810
    },
    {
      "epoch": 0.9679309525694829,
      "grad_norm": 0.24803981184959412,
      "learning_rate": 2.5e-05,
      "loss": 1.2919,
      "num_input_tokens_seen": 350146896,
      "step": 3820
    },
    {
      "epoch": 0.9704648032306596,
      "grad_norm": 0.27010080218315125,
      "learning_rate": 2.5e-05,
      "loss": 1.3511,
      "num_input_tokens_seen": 351082648,
      "step": 3830
    },
    {
      "epoch": 0.9729986538918363,
      "grad_norm": 0.3154395520687103,
      "learning_rate": 2.5e-05,
      "loss": 1.328,
      "num_input_tokens_seen": 351973288,
      "step": 3840
    },
    {
      "epoch": 0.9755325045530129,
      "grad_norm": 0.27058759331703186,
      "learning_rate": 2.5e-05,
      "loss": 1.2797,
      "num_input_tokens_seen": 352899120,
      "step": 3850
    },
    {
      "epoch": 0.9780663552141896,
      "grad_norm": 0.22412972152233124,
      "learning_rate": 2.5e-05,
      "loss": 1.3193,
      "num_input_tokens_seen": 353825356,
      "step": 3860
    },
    {
      "epoch": 0.9806002058753662,
      "grad_norm": 0.3295518755912781,
      "learning_rate": 2.5e-05,
      "loss": 1.324,
      "num_input_tokens_seen": 354778268,
      "step": 3870
    },
    {
      "epoch": 0.9831340565365428,
      "grad_norm": 0.20455938577651978,
      "learning_rate": 2.5e-05,
      "loss": 1.3359,
      "num_input_tokens_seen": 355687292,
      "step": 3880
    },
    {
      "epoch": 0.9856679071977196,
      "grad_norm": 0.22574731707572937,
      "learning_rate": 2.5e-05,
      "loss": 1.3081,
      "num_input_tokens_seen": 356581252,
      "step": 3890
    },
    {
      "epoch": 0.9882017578588962,
      "grad_norm": 0.25318706035614014,
      "learning_rate": 2.5e-05,
      "loss": 1.3327,
      "num_input_tokens_seen": 357531400,
      "step": 3900
    },
    {
      "epoch": 0.9907356085200728,
      "grad_norm": 0.25423163175582886,
      "learning_rate": 2.5e-05,
      "loss": 1.3269,
      "num_input_tokens_seen": 358429676,
      "step": 3910
    },
    {
      "epoch": 0.9932694591812495,
      "grad_norm": 0.23770791292190552,
      "learning_rate": 2.5e-05,
      "loss": 1.2942,
      "num_input_tokens_seen": 359328932,
      "step": 3920
    },
    {
      "epoch": 0.9958033098424262,
      "grad_norm": 0.23878265917301178,
      "learning_rate": 2.5e-05,
      "loss": 1.3295,
      "num_input_tokens_seen": 360264552,
      "step": 3930
    },
    {
      "epoch": 0.9983371605036028,
      "grad_norm": 0.2264624685049057,
      "learning_rate": 2.5e-05,
      "loss": 1.3224,
      "num_input_tokens_seen": 361179728,
      "step": 3940
    },
    {
      "epoch": 0.9998574709003089,
      "num_input_tokens_seen": 361724696,
      "step": 3946,
      "total_flos": 1.4115183327245763e+18,
      "train_loss": 1.3691888915302182,
      "train_runtime": 65409.2818,
      "train_samples_per_second": 15.446,
      "train_steps_per_second": 0.06
    }
  ],
  "logging_steps": 10,
  "max_steps": 3946,
  "num_input_tokens_seen": 361724696,
  "num_train_epochs": 1,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.4115183327245763e+18,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}