{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.7212405337179949,
  "eval_steps": 200,
  "global_step": 6000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0006010337780983291,
      "grad_norm": 0.9436860084533691,
      "learning_rate": 3.0048076923076927e-06,
      "loss": 4.4875,
      "step": 5
    },
    {
      "epoch": 0.0012020675561966582,
      "grad_norm": 0.49743083119392395,
      "learning_rate": 6.0096153846153855e-06,
      "loss": 4.2438,
      "step": 10
    },
    {
      "epoch": 0.0018031013342949874,
      "grad_norm": 0.7305557131767273,
      "learning_rate": 9.014423076923076e-06,
      "loss": 4.7438,
      "step": 15
    },
    {
      "epoch": 0.0024041351123933164,
      "grad_norm": 1.2842742204666138,
      "learning_rate": 1.2019230769230771e-05,
      "loss": 4.3125,
      "step": 20
    },
    {
      "epoch": 0.0030051688904916456,
      "grad_norm": 1.0128940343856812,
      "learning_rate": 1.5024038461538462e-05,
      "loss": 4.2969,
      "step": 25
    },
    {
      "epoch": 0.003606202668589975,
      "grad_norm": 1.6097222566604614,
      "learning_rate": 1.8028846153846152e-05,
      "loss": 4.3625,
      "step": 30
    },
    {
      "epoch": 0.004207236446688304,
      "grad_norm": 0.7380394339561462,
      "learning_rate": 2.103365384615385e-05,
      "loss": 3.6562,
      "step": 35
    },
    {
      "epoch": 0.004808270224786633,
      "grad_norm": 2.499553918838501,
      "learning_rate": 2.4038461538461542e-05,
      "loss": 3.9656,
      "step": 40
    },
    {
      "epoch": 0.005409304002884962,
      "grad_norm": 0.9382426142692566,
      "learning_rate": 2.704326923076923e-05,
      "loss": 3.7906,
      "step": 45
    },
    {
      "epoch": 0.006010337780983291,
      "grad_norm": 0.4448552429676056,
      "learning_rate": 3.0048076923076925e-05,
      "loss": 3.4531,
      "step": 50
    },
    {
      "epoch": 0.00661137155908162,
      "grad_norm": 0.6187996864318848,
      "learning_rate": 3.3052884615384615e-05,
      "loss": 3.0406,
      "step": 55
    },
    {
      "epoch": 0.00721240533717995,
      "grad_norm": 0.4894959032535553,
      "learning_rate": 3.6057692307692304e-05,
      "loss": 2.9844,
      "step": 60
    },
    {
      "epoch": 0.007813439115278278,
      "grad_norm": 0.523160994052887,
      "learning_rate": 3.90625e-05,
      "loss": 2.825,
      "step": 65
    },
    {
      "epoch": 0.008414472893376608,
      "grad_norm": 0.41818058490753174,
      "learning_rate": 4.20673076923077e-05,
      "loss": 2.5125,
      "step": 70
    },
    {
      "epoch": 0.009015506671474938,
      "grad_norm": 0.37683457136154175,
      "learning_rate": 4.507211538461539e-05,
      "loss": 2.6844,
      "step": 75
    },
    {
      "epoch": 0.009616540449573266,
      "grad_norm": 0.428375780582428,
      "learning_rate": 4.8076923076923084e-05,
      "loss": 2.6719,
      "step": 80
    },
    {
      "epoch": 0.010217574227671595,
      "grad_norm": 0.3897765576839447,
      "learning_rate": 5.108173076923077e-05,
      "loss": 2.2531,
      "step": 85
    },
    {
      "epoch": 0.010818608005769925,
      "grad_norm": 0.2265370637178421,
      "learning_rate": 5.408653846153846e-05,
      "loss": 2.2844,
      "step": 90
    },
    {
      "epoch": 0.011419641783868253,
      "grad_norm": 0.2113611400127411,
      "learning_rate": 5.709134615384615e-05,
      "loss": 2.1922,
      "step": 95
    },
    {
      "epoch": 0.012020675561966582,
      "grad_norm": 0.1886824667453766,
      "learning_rate": 6.009615384615385e-05,
      "loss": 2.3516,
      "step": 100
    },
    {
      "epoch": 0.012621709340064912,
      "grad_norm": 0.25855502486228943,
      "learning_rate": 6.310096153846154e-05,
      "loss": 2.4,
      "step": 105
    },
    {
      "epoch": 0.01322274311816324,
      "grad_norm": 0.22833962738513947,
      "learning_rate": 6.610576923076923e-05,
      "loss": 2.2844,
      "step": 110
    },
    {
      "epoch": 0.01382377689626157,
      "grad_norm": 0.30784738063812256,
      "learning_rate": 6.911057692307693e-05,
      "loss": 2.2016,
      "step": 115
    },
    {
      "epoch": 0.0144248106743599,
      "grad_norm": 0.3998744487762451,
      "learning_rate": 7.211538461538461e-05,
      "loss": 2.4125,
      "step": 120
    },
    {
      "epoch": 0.015025844452458229,
      "grad_norm": 0.24773858487606049,
      "learning_rate": 7.512019230769231e-05,
      "loss": 2.4156,
      "step": 125
    },
    {
      "epoch": 0.015626878230556557,
      "grad_norm": 0.26020580530166626,
      "learning_rate": 7.8125e-05,
      "loss": 2.0094,
      "step": 130
    },
    {
      "epoch": 0.016227912008654886,
      "grad_norm": 0.25112366676330566,
      "learning_rate": 8.112980769230769e-05,
      "loss": 2.5969,
      "step": 135
    },
    {
      "epoch": 0.016828945786753216,
      "grad_norm": 0.3155271112918854,
      "learning_rate": 8.41346153846154e-05,
      "loss": 1.9844,
      "step": 140
    },
    {
      "epoch": 0.017429979564851546,
      "grad_norm": 0.2684473693370819,
      "learning_rate": 8.713942307692307e-05,
      "loss": 2.3594,
      "step": 145
    },
    {
      "epoch": 0.018031013342949875,
      "grad_norm": 0.19519321620464325,
      "learning_rate": 9.014423076923077e-05,
      "loss": 2.1906,
      "step": 150
    },
    {
      "epoch": 0.0186320471210482,
      "grad_norm": 0.29595857858657837,
      "learning_rate": 9.314903846153846e-05,
      "loss": 2.4844,
      "step": 155
    },
    {
      "epoch": 0.01923308089914653,
      "grad_norm": 0.21725840866565704,
      "learning_rate": 9.615384615384617e-05,
      "loss": 1.9969,
      "step": 160
    },
    {
      "epoch": 0.01983411467724486,
      "grad_norm": 0.250431627035141,
      "learning_rate": 9.915865384615384e-05,
      "loss": 2.1469,
      "step": 165
    },
    {
      "epoch": 0.02043514845534319,
      "grad_norm": 0.22979402542114258,
      "learning_rate": 0.00010216346153846153,
      "loss": 2.0891,
      "step": 170
    },
    {
      "epoch": 0.02103618223344152,
      "grad_norm": 0.29841649532318115,
      "learning_rate": 0.00010516826923076924,
      "loss": 2.0891,
      "step": 175
    },
    {
      "epoch": 0.02163721601153985,
      "grad_norm": 0.3121524155139923,
      "learning_rate": 0.00010817307692307693,
      "loss": 2.2938,
      "step": 180
    },
    {
      "epoch": 0.02223824978963818,
      "grad_norm": 0.25094497203826904,
      "learning_rate": 0.00011117788461538462,
      "loss": 2.0672,
      "step": 185
    },
    {
      "epoch": 0.022839283567736506,
      "grad_norm": 0.32229083776474,
      "learning_rate": 0.0001141826923076923,
      "loss": 1.9781,
      "step": 190
    },
    {
      "epoch": 0.023440317345834835,
      "grad_norm": 0.30247944593429565,
      "learning_rate": 0.0001171875,
      "loss": 2.4469,
      "step": 195
    },
    {
      "epoch": 0.024041351123933165,
      "grad_norm": 0.3992522358894348,
      "learning_rate": 0.0001201923076923077,
      "loss": 2.1609,
      "step": 200
    },
    {
      "epoch": 0.024041351123933165,
      "eval_loss": 2.7308592796325684,
      "eval_model_preparation_time": 0.0053,
      "eval_runtime": 35.262,
      "eval_samples_per_second": 4.537,
      "eval_steps_per_second": 1.134,
      "step": 200
    },
    {
      "epoch": 0.024642384902031494,
      "grad_norm": 0.28425589203834534,
      "learning_rate": 0.0001231971153846154,
      "loss": 2.625,
      "step": 205
    },
    {
      "epoch": 0.025243418680129824,
      "grad_norm": 0.31964734196662903,
      "learning_rate": 0.00012620192307692308,
      "loss": 1.9328,
      "step": 210
    },
    {
      "epoch": 0.025844452458228154,
      "grad_norm": 0.37272173166275024,
      "learning_rate": 0.00012920673076923078,
      "loss": 2.1641,
      "step": 215
    },
    {
      "epoch": 0.02644548623632648,
      "grad_norm": 0.32725071907043457,
      "learning_rate": 0.00013221153846153846,
      "loss": 2.225,
      "step": 220
    },
    {
      "epoch": 0.02704652001442481,
      "grad_norm": 0.25303465127944946,
      "learning_rate": 0.00013521634615384616,
      "loss": 2.2375,
      "step": 225
    },
    {
      "epoch": 0.02764755379252314,
      "grad_norm": 0.4098326861858368,
      "learning_rate": 0.00013822115384615386,
      "loss": 2.0203,
      "step": 230
    },
    {
      "epoch": 0.02824858757062147,
      "grad_norm": 0.3435593545436859,
      "learning_rate": 0.00014122596153846154,
      "loss": 2.3016,
      "step": 235
    },
    {
      "epoch": 0.0288496213487198,
      "grad_norm": 0.4556426703929901,
      "learning_rate": 0.00014423076923076922,
      "loss": 2.2969,
      "step": 240
    },
    {
      "epoch": 0.029450655126818128,
      "grad_norm": 0.39692002534866333,
      "learning_rate": 0.00014723557692307692,
      "loss": 2.3031,
      "step": 245
    },
    {
      "epoch": 0.030051688904916458,
      "grad_norm": 0.31686538457870483,
      "learning_rate": 0.00015024038461538462,
      "loss": 2.2984,
      "step": 250
    },
    {
      "epoch": 0.030652722683014784,
      "grad_norm": 0.30815422534942627,
      "learning_rate": 0.00015324519230769233,
      "loss": 2.1734,
      "step": 255
    },
    {
      "epoch": 0.031253756461113114,
      "grad_norm": 0.3927950859069824,
      "learning_rate": 0.00015625,
      "loss": 2.0031,
      "step": 260
    },
    {
      "epoch": 0.03185479023921144,
      "grad_norm": 0.3010413944721222,
      "learning_rate": 0.00015925480769230768,
      "loss": 2.1875,
      "step": 265
    },
    {
      "epoch": 0.03245582401730977,
      "grad_norm": 0.39929866790771484,
      "learning_rate": 0.00016225961538461538,
      "loss": 2.2266,
      "step": 270
    },
    {
      "epoch": 0.0330568577954081,
      "grad_norm": 0.3709786832332611,
      "learning_rate": 0.00016526442307692309,
      "loss": 2.2344,
      "step": 275
    },
    {
      "epoch": 0.03365789157350643,
      "grad_norm": 0.38551804423332214,
      "learning_rate": 0.0001682692307692308,
      "loss": 2.0391,
      "step": 280
    },
    {
      "epoch": 0.03425892535160476,
      "grad_norm": 0.3497028350830078,
      "learning_rate": 0.00017127403846153847,
      "loss": 2.1328,
      "step": 285
    },
    {
      "epoch": 0.03485995912970309,
      "grad_norm": 0.22066070139408112,
      "learning_rate": 0.00017427884615384614,
      "loss": 1.9891,
      "step": 290
    },
    {
      "epoch": 0.03546099290780142,
      "grad_norm": 0.3861188590526581,
      "learning_rate": 0.00017728365384615385,
      "loss": 2.0266,
      "step": 295
    },
    {
      "epoch": 0.03606202668589975,
      "grad_norm": 0.43038997054100037,
      "learning_rate": 0.00018028846153846155,
      "loss": 2.2062,
      "step": 300
    },
    {
      "epoch": 0.03666306046399807,
      "grad_norm": 0.4089072644710541,
      "learning_rate": 0.00018329326923076922,
      "loss": 2.2016,
      "step": 305
    },
    {
      "epoch": 0.0372640942420964,
      "grad_norm": 0.40281516313552856,
      "learning_rate": 0.00018629807692307693,
      "loss": 2.2578,
      "step": 310
    },
    {
      "epoch": 0.03786512802019473,
      "grad_norm": 0.33316513895988464,
      "learning_rate": 0.0001893028846153846,
      "loss": 2.1844,
      "step": 315
    },
    {
      "epoch": 0.03846616179829306,
      "grad_norm": 0.4020228087902069,
      "learning_rate": 0.00019230769230769233,
      "loss": 2.2109,
      "step": 320
    },
    {
      "epoch": 0.03906719557639139,
      "grad_norm": 0.36403888463974,
      "learning_rate": 0.0001953125,
      "loss": 2.0063,
      "step": 325
    },
    {
      "epoch": 0.03966822935448972,
      "grad_norm": 0.4289080500602722,
      "learning_rate": 0.0001983173076923077,
      "loss": 2.1641,
      "step": 330
    },
    {
      "epoch": 0.04026926313258805,
      "grad_norm": 0.3827407658100128,
      "learning_rate": 0.0002013221153846154,
      "loss": 2.4125,
      "step": 335
    },
    {
      "epoch": 0.04087029691068638,
      "grad_norm": 0.28297996520996094,
      "learning_rate": 0.00020432692307692307,
      "loss": 2.2047,
      "step": 340
    },
    {
      "epoch": 0.04147133068878471,
      "grad_norm": 0.3654349744319916,
      "learning_rate": 0.0002073317307692308,
      "loss": 2.0344,
      "step": 345
    },
    {
      "epoch": 0.04207236446688304,
      "grad_norm": 0.44768983125686646,
      "learning_rate": 0.00021033653846153847,
      "loss": 2.0469,
      "step": 350
    },
    {
      "epoch": 0.04267339824498137,
      "grad_norm": 0.36050865054130554,
      "learning_rate": 0.00021334134615384615,
      "loss": 1.8203,
      "step": 355
    },
    {
      "epoch": 0.0432744320230797,
      "grad_norm": 0.41343504190444946,
      "learning_rate": 0.00021634615384615385,
      "loss": 1.9031,
      "step": 360
    },
    {
      "epoch": 0.04387546580117803,
      "grad_norm": 0.33549779653549194,
      "learning_rate": 0.00021935096153846153,
      "loss": 1.9859,
      "step": 365
    },
    {
      "epoch": 0.04447649957927636,
      "grad_norm": 0.39200559258461,
      "learning_rate": 0.00022235576923076923,
      "loss": 2.0672,
      "step": 370
    },
    {
      "epoch": 0.04507753335737468,
      "grad_norm": 0.5816010236740112,
      "learning_rate": 0.00022536057692307694,
      "loss": 2.1625,
      "step": 375
    },
    {
      "epoch": 0.04567856713547301,
      "grad_norm": 0.4004225432872772,
      "learning_rate": 0.0002283653846153846,
      "loss": 2.1297,
      "step": 380
    },
    {
      "epoch": 0.04627960091357134,
      "grad_norm": 0.3329584300518036,
      "learning_rate": 0.00023137019230769232,
      "loss": 1.8969,
      "step": 385
    },
    {
      "epoch": 0.04688063469166967,
      "grad_norm": 0.3800398111343384,
      "learning_rate": 0.000234375,
      "loss": 1.875,
      "step": 390
    },
    {
      "epoch": 0.047481668469768,
      "grad_norm": 0.5345351696014404,
      "learning_rate": 0.0002373798076923077,
      "loss": 2.0641,
      "step": 395
    },
    {
      "epoch": 0.04808270224786633,
      "grad_norm": 0.31537583470344543,
      "learning_rate": 0.0002403846153846154,
      "loss": 2.0828,
      "step": 400
    },
    {
      "epoch": 0.04808270224786633,
      "eval_loss": 2.657031297683716,
      "eval_model_preparation_time": 0.0053,
      "eval_runtime": 35.2197,
      "eval_samples_per_second": 4.543,
      "eval_steps_per_second": 1.136,
      "step": 400
    },
    {
      "epoch": 0.04868373602596466,
      "grad_norm": 0.3651765286922455,
      "learning_rate": 0.00024338942307692307,
      "loss": 2.2188,
      "step": 405
    },
    {
      "epoch": 0.04928476980406299,
      "grad_norm": 0.42044126987457275,
      "learning_rate": 0.0002463942307692308,
      "loss": 1.9625,
      "step": 410
    },
    {
      "epoch": 0.04988580358216132,
      "grad_norm": 0.3405047357082367,
      "learning_rate": 0.00024939903846153845,
      "loss": 2.1203,
      "step": 415
    },
    {
      "epoch": 0.05048683736025965,
      "grad_norm": 0.5022028088569641,
      "learning_rate": 0.00025240384615384616,
      "loss": 1.7672,
      "step": 420
    },
    {
      "epoch": 0.05108787113835798,
      "grad_norm": 0.31208300590515137,
      "learning_rate": 0.00025540865384615386,
      "loss": 1.9266,
      "step": 425
    },
    {
      "epoch": 0.05168890491645631,
      "grad_norm": 0.39399516582489014,
      "learning_rate": 0.00025841346153846156,
      "loss": 1.8828,
      "step": 430
    },
    {
      "epoch": 0.05228993869455464,
      "grad_norm": 0.42515093088150024,
      "learning_rate": 0.0002614182692307692,
      "loss": 1.7656,
      "step": 435
    },
    {
      "epoch": 0.05289097247265296,
      "grad_norm": 0.3947089910507202,
      "learning_rate": 0.0002644230769230769,
      "loss": 2.0484,
      "step": 440
    },
    {
      "epoch": 0.05349200625075129,
      "grad_norm": 0.6280628442764282,
      "learning_rate": 0.0002674278846153846,
      "loss": 2.1422,
      "step": 445
    },
    {
      "epoch": 0.05409304002884962,
      "grad_norm": 0.3639807105064392,
      "learning_rate": 0.0002704326923076923,
      "loss": 1.9781,
      "step": 450
    },
    {
      "epoch": 0.05469407380694795,
      "grad_norm": 0.3984295427799225,
      "learning_rate": 0.0002734375,
      "loss": 2.2359,
      "step": 455
    },
    {
      "epoch": 0.05529510758504628,
      "grad_norm": 0.33954715728759766,
      "learning_rate": 0.00027644230769230773,
      "loss": 2.3547,
      "step": 460
    },
    {
      "epoch": 0.05589614136314461,
      "grad_norm": 0.4361511468887329,
      "learning_rate": 0.0002794471153846154,
      "loss": 2.0859,
      "step": 465
    },
    {
      "epoch": 0.05649717514124294,
      "grad_norm": 0.471563458442688,
      "learning_rate": 0.0002824519230769231,
      "loss": 2.1703,
      "step": 470
    },
    {
      "epoch": 0.05709820891934127,
      "grad_norm": 0.2517772614955902,
      "learning_rate": 0.0002854567307692308,
      "loss": 2.0281,
      "step": 475
    },
    {
      "epoch": 0.0576992426974396,
      "grad_norm": 0.3190082907676697,
      "learning_rate": 0.00028846153846153843,
      "loss": 2.0234,
      "step": 480
    },
    {
      "epoch": 0.05830027647553793,
      "grad_norm": 0.37972012162208557,
      "learning_rate": 0.00029146634615384614,
      "loss": 2.15,
      "step": 485
    },
    {
      "epoch": 0.058901310253636256,
      "grad_norm": 0.37980136275291443,
      "learning_rate": 0.00029447115384615384,
      "loss": 2.1219,
      "step": 490
    },
    {
      "epoch": 0.059502344031734586,
      "grad_norm": 0.32648953795433044,
      "learning_rate": 0.00029747596153846154,
      "loss": 1.9703,
      "step": 495
    },
    {
      "epoch": 0.060103377809832916,
      "grad_norm": 0.28836116194725037,
      "learning_rate": 0.00030048076923076925,
      "loss": 2.0406,
      "step": 500
    },
    {
      "epoch": 0.060704411587931245,
      "grad_norm": 0.2953934967517853,
      "learning_rate": 0.00030348557692307695,
      "loss": 2.2156,
      "step": 505
    },
    {
      "epoch": 0.06130544536602957,
      "grad_norm": 0.4778139889240265,
      "learning_rate": 0.00030649038461538465,
      "loss": 2.0672,
      "step": 510
    },
    {
      "epoch": 0.0619064791441279,
      "grad_norm": 0.27339640259742737,
      "learning_rate": 0.0003094951923076923,
      "loss": 1.8953,
      "step": 515
    },
    {
      "epoch": 0.06250751292222623,
      "grad_norm": 0.3127667009830475,
      "learning_rate": 0.0003125,
      "loss": 2.0859,
      "step": 520
    },
    {
      "epoch": 0.06310854670032456,
      "grad_norm": 0.2676738500595093,
      "learning_rate": 0.0003155048076923077,
      "loss": 1.9656,
      "step": 525
    },
    {
      "epoch": 0.06370958047842289,
      "grad_norm": 0.3519584834575653,
      "learning_rate": 0.00031850961538461536,
      "loss": 2.0828,
      "step": 530
    },
    {
      "epoch": 0.06431061425652122,
      "grad_norm": 0.38000714778900146,
      "learning_rate": 0.00032151442307692306,
      "loss": 1.8656,
      "step": 535
    },
    {
      "epoch": 0.06491164803461955,
      "grad_norm": 0.5076779127120972,
      "learning_rate": 0.00032451923076923077,
      "loss": 1.8938,
      "step": 540
    },
    {
      "epoch": 0.06551268181271787,
      "grad_norm": 0.3919801414012909,
      "learning_rate": 0.00032752403846153847,
      "loss": 2.1203,
      "step": 545
    },
    {
      "epoch": 0.0661137155908162,
      "grad_norm": 0.3263305425643921,
      "learning_rate": 0.00033052884615384617,
      "loss": 2.0344,
      "step": 550
    },
    {
      "epoch": 0.06671474936891453,
      "grad_norm": 0.4196506440639496,
      "learning_rate": 0.0003335336538461539,
      "loss": 2.1078,
      "step": 555
    },
    {
      "epoch": 0.06731578314701286,
      "grad_norm": 0.3997637927532196,
      "learning_rate": 0.0003365384615384616,
      "loss": 1.8844,
      "step": 560
    },
    {
      "epoch": 0.06791681692511119,
      "grad_norm": 0.39547184109687805,
      "learning_rate": 0.00033954326923076923,
      "loss": 1.9406,
      "step": 565
    },
    {
      "epoch": 0.06851785070320952,
      "grad_norm": 0.36170271039009094,
      "learning_rate": 0.00034254807692307693,
      "loss": 2.2469,
      "step": 570
    },
    {
      "epoch": 0.06911888448130785,
      "grad_norm": 0.3041069507598877,
      "learning_rate": 0.00034555288461538463,
      "loss": 1.7734,
      "step": 575
    },
    {
      "epoch": 0.06971991825940618,
      "grad_norm": 0.31936579942703247,
      "learning_rate": 0.0003485576923076923,
      "loss": 2.1141,
      "step": 580
    },
    {
      "epoch": 0.0703209520375045,
      "grad_norm": 0.5643404722213745,
      "learning_rate": 0.0003515625,
      "loss": 1.9,
      "step": 585
    },
    {
      "epoch": 0.07092198581560284,
      "grad_norm": 0.43453335762023926,
      "learning_rate": 0.0003545673076923077,
      "loss": 1.5203,
      "step": 590
    },
    {
      "epoch": 0.07152301959370116,
      "grad_norm": 0.28918376564979553,
      "learning_rate": 0.0003575721153846154,
      "loss": 1.9859,
      "step": 595
    },
    {
      "epoch": 0.0721240533717995,
      "grad_norm": 0.4441574215888977,
      "learning_rate": 0.0003605769230769231,
      "loss": 1.7422,
      "step": 600
    },
    {
      "epoch": 0.0721240533717995,
      "eval_loss": 2.598828077316284,
      "eval_model_preparation_time": 0.0053,
      "eval_runtime": 35.2182,
      "eval_samples_per_second": 4.543,
      "eval_steps_per_second": 1.136,
      "step": 600
    },
    {
      "epoch": 0.07272508714989782,
      "grad_norm": 0.47054970264434814,
      "learning_rate": 0.0003635817307692308,
      "loss": 1.975,
      "step": 605
    },
    {
      "epoch": 0.07332612092799615,
      "grad_norm": 0.4808221459388733,
      "learning_rate": 0.00036658653846153845,
      "loss": 2.0969,
      "step": 610
    },
    {
      "epoch": 0.07392715470609448,
      "grad_norm": 0.40260159969329834,
      "learning_rate": 0.00036959134615384615,
      "loss": 2.0781,
      "step": 615
    },
    {
      "epoch": 0.0745281884841928,
      "grad_norm": 0.3565881848335266,
      "learning_rate": 0.00037259615384615386,
      "loss": 1.8562,
      "step": 620
    },
    {
      "epoch": 0.07512922226229114,
      "grad_norm": 0.41623455286026,
      "learning_rate": 0.00037560096153846156,
      "loss": 1.9688,
      "step": 625
    },
    {
      "epoch": 0.07573025604038947,
      "grad_norm": 0.442056804895401,
      "learning_rate": 0.0003786057692307692,
      "loss": 2.0531,
      "step": 630
    },
    {
      "epoch": 0.0763312898184878,
      "grad_norm": 0.5474425554275513,
      "learning_rate": 0.0003816105769230769,
      "loss": 2.1391,
      "step": 635
    },
    {
      "epoch": 0.07693232359658612,
      "grad_norm": 0.29002273082733154,
      "learning_rate": 0.00038461538461538467,
      "loss": 1.6906,
      "step": 640
    },
    {
      "epoch": 0.07753335737468446,
      "grad_norm": 0.30469194054603577,
      "learning_rate": 0.0003876201923076923,
      "loss": 1.6859,
      "step": 645
    },
    {
      "epoch": 0.07813439115278278,
      "grad_norm": 0.3932645618915558,
      "learning_rate": 0.000390625,
      "loss": 1.8328,
      "step": 650
    },
    {
      "epoch": 0.07873542493088112,
      "grad_norm": 0.4049251079559326,
      "learning_rate": 0.0003936298076923077,
      "loss": 1.8672,
      "step": 655
    },
    {
      "epoch": 0.07933645870897944,
      "grad_norm": 0.4889291524887085,
      "learning_rate": 0.0003966346153846154,
      "loss": 2.0531,
      "step": 660
    },
    {
      "epoch": 0.07993749248707778,
      "grad_norm": 0.38475117087364197,
      "learning_rate": 0.0003996394230769231,
      "loss": 1.8422,
      "step": 665
    },
    {
      "epoch": 0.0805385262651761,
      "grad_norm": 0.34599217772483826,
      "learning_rate": 0.0004026442307692308,
      "loss": 1.8391,
      "step": 670
    },
    {
      "epoch": 0.08113956004327443,
      "grad_norm": 0.39600178599357605,
      "learning_rate": 0.00040564903846153843,
      "loss": 1.8484,
      "step": 675
    },
    {
      "epoch": 0.08174059382137276,
      "grad_norm": 0.3293285071849823,
      "learning_rate": 0.00040865384615384613,
      "loss": 1.6656,
      "step": 680
    },
    {
      "epoch": 0.08234162759947108,
      "grad_norm": 0.37310031056404114,
      "learning_rate": 0.00041165865384615384,
      "loss": 1.9609,
      "step": 685
    },
    {
      "epoch": 0.08294266137756942,
      "grad_norm": 0.41512343287467957,
      "learning_rate": 0.0004146634615384616,
      "loss": 1.9937,
      "step": 690
    },
    {
      "epoch": 0.08354369515566774,
      "grad_norm": 0.47950249910354614,
      "learning_rate": 0.00041766826923076924,
      "loss": 1.9109,
      "step": 695
    },
    {
      "epoch": 0.08414472893376608,
      "grad_norm": 0.4324653744697571,
      "learning_rate": 0.00042067307692307695,
      "loss": 1.9953,
      "step": 700
    },
    {
      "epoch": 0.0847457627118644,
      "grad_norm": 0.3693973422050476,
      "learning_rate": 0.00042367788461538465,
      "loss": 1.9016,
      "step": 705
    },
    {
      "epoch": 0.08534679648996274,
      "grad_norm": 0.33113107085227966,
      "learning_rate": 0.0004266826923076923,
      "loss": 2.2266,
      "step": 710
    },
    {
      "epoch": 0.08594783026806106,
      "grad_norm": 0.5808571577072144,
      "learning_rate": 0.0004296875,
      "loss": 1.5063,
      "step": 715
    },
    {
      "epoch": 0.0865488640461594,
      "grad_norm": 0.3792312443256378,
      "learning_rate": 0.0004326923076923077,
      "loss": 1.8016,
      "step": 720
    },
    {
      "epoch": 0.08714989782425772,
      "grad_norm": 0.43698450922966003,
      "learning_rate": 0.00043569711538461535,
      "loss": 1.7219,
      "step": 725
    },
    {
      "epoch": 0.08775093160235606,
      "grad_norm": 0.43264222145080566,
      "learning_rate": 0.00043870192307692306,
      "loss": 1.7234,
      "step": 730
    },
    {
      "epoch": 0.08835196538045438,
      "grad_norm": 0.5246540307998657,
      "learning_rate": 0.0004417067307692308,
      "loss": 1.7531,
      "step": 735
    },
    {
      "epoch": 0.08895299915855272,
      "grad_norm": 0.2953200936317444,
      "learning_rate": 0.00044471153846153846,
      "loss": 1.9438,
      "step": 740
    },
    {
      "epoch": 0.08955403293665104,
      "grad_norm": 0.39238616824150085,
      "learning_rate": 0.00044771634615384617,
      "loss": 1.7172,
      "step": 745
    },
    {
      "epoch": 0.09015506671474936,
      "grad_norm": 0.4887576401233673,
      "learning_rate": 0.00045072115384615387,
      "loss": 1.9594,
      "step": 750
    },
    {
      "epoch": 0.0907561004928477,
      "grad_norm": 0.391634076833725,
      "learning_rate": 0.0004537259615384616,
      "loss": 1.8406,
      "step": 755
    },
    {
      "epoch": 0.09135713427094602,
      "grad_norm": 0.4006985127925873,
      "learning_rate": 0.0004567307692307692,
      "loss": 1.7984,
      "step": 760
    },
    {
      "epoch": 0.09195816804904436,
      "grad_norm": 0.3601657748222351,
      "learning_rate": 0.0004597355769230769,
      "loss": 1.9844,
      "step": 765
    },
    {
      "epoch": 0.09255920182714268,
      "grad_norm": 0.5057326555252075,
      "learning_rate": 0.00046274038461538463,
      "loss": 1.6703,
      "step": 770
    },
    {
      "epoch": 0.09316023560524102,
      "grad_norm": 0.5787122845649719,
      "learning_rate": 0.0004657451923076923,
      "loss": 1.8984,
      "step": 775
    },
    {
      "epoch": 0.09376126938333934,
      "grad_norm": 0.4849441945552826,
      "learning_rate": 0.00046875,
      "loss": 1.8672,
      "step": 780
    },
    {
      "epoch": 0.09436230316143768,
      "grad_norm": 0.44167378544807434,
      "learning_rate": 0.00047175480769230774,
      "loss": 1.6422,
      "step": 785
    },
    {
      "epoch": 0.094963336939536,
      "grad_norm": 0.6295076608657837,
      "learning_rate": 0.0004747596153846154,
      "loss": 1.6875,
      "step": 790
    },
    {
      "epoch": 0.09556437071763434,
      "grad_norm": 0.4804101586341858,
      "learning_rate": 0.0004777644230769231,
      "loss": 1.8203,
      "step": 795
    },
    {
      "epoch": 0.09616540449573266,
      "grad_norm": 0.4898495674133301,
      "learning_rate": 0.0004807692307692308,
      "loss": 1.9891,
      "step": 800
    },
    {
      "epoch": 0.09616540449573266,
      "eval_loss": 2.535351514816284,
      "eval_model_preparation_time": 0.0053,
      "eval_runtime": 35.1754,
      "eval_samples_per_second": 4.549,
      "eval_steps_per_second": 1.137,
      "step": 800
    },
    {
      "epoch": 0.096766438273831,
      "grad_norm": 0.43688085675239563,
      "learning_rate": 0.00048377403846153845,
      "loss": 1.7234,
      "step": 805
    },
    {
      "epoch": 0.09736747205192932,
      "grad_norm": 0.5891087651252747,
      "learning_rate": 0.00048677884615384615,
      "loss": 1.7969,
      "step": 810
    },
    {
      "epoch": 0.09796850583002764,
      "grad_norm": 0.5140319466590881,
      "learning_rate": 0.0004897836538461539,
      "loss": 2.0719,
      "step": 815
    },
    {
      "epoch": 0.09856953960812598,
      "grad_norm": 0.40886375308036804,
      "learning_rate": 0.0004927884615384616,
      "loss": 2.0891,
      "step": 820
    },
    {
      "epoch": 0.0991705733862243,
      "grad_norm": 0.3513309955596924,
      "learning_rate": 0.0004957932692307692,
      "loss": 1.8453,
      "step": 825
    },
    {
      "epoch": 0.09977160716432264,
      "grad_norm": 0.5530559420585632,
      "learning_rate": 0.0004987980769230769,
      "loss": 1.675,
      "step": 830
    },
    {
      "epoch": 0.10037264094242096,
      "grad_norm": 0.4348265528678894,
      "learning_rate": 0.0004999999983630302,
      "loss": 1.7891,
      "step": 835
    },
    {
      "epoch": 0.1009736747205193,
      "grad_norm": 0.5396342277526855,
      "learning_rate": 0.0004999999883593255,
      "loss": 1.9047,
      "step": 840
    },
    {
      "epoch": 0.10157470849861762,
      "grad_norm": 0.5154384970664978,
      "learning_rate": 0.0004999999692613442,
      "loss": 1.8844,
      "step": 845
    },
    {
      "epoch": 0.10217574227671596,
      "grad_norm": 0.29072120785713196,
      "learning_rate": 0.0004999999410690872,
      "loss": 1.6531,
      "step": 850
    },
    {
      "epoch": 0.10277677605481428,
      "grad_norm": 0.4125816822052002,
      "learning_rate": 0.0004999999037825552,
      "loss": 1.9031,
      "step": 855
    },
    {
      "epoch": 0.10337780983291261,
      "grad_norm": 0.34915369749069214,
      "learning_rate": 0.0004999998574017497,
      "loss": 1.8609,
      "step": 860
    },
    {
      "epoch": 0.10397884361101094,
      "grad_norm": 0.3622804284095764,
      "learning_rate": 0.0004999998019266724,
      "loss": 1.7484,
      "step": 865
    },
    {
      "epoch": 0.10457987738910927,
      "grad_norm": 0.36787149310112,
      "learning_rate": 0.0004999997373573254,
      "loss": 1.7812,
      "step": 870
    },
    {
      "epoch": 0.1051809111672076,
      "grad_norm": 0.4469545781612396,
      "learning_rate": 0.0004999996636937108,
      "loss": 1.5484,
      "step": 875
    },
    {
      "epoch": 0.10578194494530592,
      "grad_norm": 0.30026400089263916,
      "learning_rate": 0.0004999995809358316,
      "loss": 1.6703,
      "step": 880
    },
    {
      "epoch": 0.10638297872340426,
      "grad_norm": 0.4870736002922058,
      "learning_rate": 0.0004999994890836904,
      "loss": 1.7547,
      "step": 885
    },
    {
      "epoch": 0.10698401250150258,
      "grad_norm": 0.6516287326812744,
      "learning_rate": 0.000499999388137291,
      "loss": 1.7891,
      "step": 890
    },
    {
      "epoch": 0.10758504627960092,
      "grad_norm": 0.2974604368209839,
      "learning_rate": 0.0004999992780966368,
      "loss": 1.8359,
      "step": 895
    },
    {
      "epoch": 0.10818608005769924,
      "grad_norm": 0.3521243929862976,
      "learning_rate": 0.0004999991589617318,
      "loss": 1.9141,
      "step": 900
    },
    {
      "epoch": 0.10878711383579757,
      "grad_norm": 0.38353726267814636,
      "learning_rate": 0.0004999990307325803,
      "loss": 1.775,
      "step": 905
    },
    {
      "epoch": 0.1093881476138959,
      "grad_norm": 0.46048542857170105,
      "learning_rate": 0.0004999988934091872,
      "loss": 1.7297,
      "step": 910
    },
    {
      "epoch": 0.10998918139199423,
      "grad_norm": 0.4313719570636749,
      "learning_rate": 0.0004999987469915573,
      "loss": 1.2891,
      "step": 915
    },
    {
      "epoch": 0.11059021517009256,
      "grad_norm": 0.5933486223220825,
      "learning_rate": 0.0004999985914796961,
      "loss": 1.6938,
      "step": 920
    },
    {
      "epoch": 0.1111912489481909,
      "grad_norm": 0.5271236300468445,
      "learning_rate": 0.000499998426873609,
      "loss": 1.8,
      "step": 925
    },
    {
      "epoch": 0.11179228272628922,
      "grad_norm": 0.3807511031627655,
      "learning_rate": 0.0004999982531733022,
      "loss": 1.3086,
      "step": 930
    },
    {
      "epoch": 0.11239331650438755,
      "grad_norm": 0.4684934914112091,
      "learning_rate": 0.0004999980703787819,
      "loss": 1.4875,
      "step": 935
    },
    {
      "epoch": 0.11299435028248588,
      "grad_norm": 0.5648980140686035,
      "learning_rate": 0.0004999978784900549,
      "loss": 1.6578,
      "step": 940
    },
    {
      "epoch": 0.1135953840605842,
      "grad_norm": 0.4021349549293518,
      "learning_rate": 0.0004999976775071278,
      "loss": 1.8266,
      "step": 945
    },
    {
      "epoch": 0.11419641783868253,
      "grad_norm": 0.3722395598888397,
      "learning_rate": 0.0004999974674300084,
      "loss": 1.8969,
      "step": 950
    },
    {
      "epoch": 0.11479745161678086,
      "grad_norm": 0.407781720161438,
      "learning_rate": 0.000499997248258704,
      "loss": 1.6562,
      "step": 955
    },
    {
      "epoch": 0.1153984853948792,
      "grad_norm": 0.44156748056411743,
      "learning_rate": 0.0004999970199932229,
      "loss": 2.0688,
      "step": 960
    },
    {
      "epoch": 0.11599951917297752,
      "grad_norm": 0.40020808577537537,
      "learning_rate": 0.000499996782633573,
      "loss": 1.5047,
      "step": 965
    },
    {
      "epoch": 0.11660055295107585,
      "grad_norm": 0.38710176944732666,
      "learning_rate": 0.0004999965361797633,
      "loss": 1.7367,
      "step": 970
    },
    {
      "epoch": 0.11720158672917418,
      "grad_norm": 0.344836562871933,
      "learning_rate": 0.0004999962806318025,
      "loss": 1.7828,
      "step": 975
    },
    {
      "epoch": 0.11780262050727251,
      "grad_norm": 0.3811284899711609,
      "learning_rate": 0.0004999960159897,
      "loss": 1.7766,
      "step": 980
    },
    {
      "epoch": 0.11840365428537084,
      "grad_norm": 0.5141933560371399,
      "learning_rate": 0.0004999957422534654,
      "loss": 1.75,
      "step": 985
    },
    {
      "epoch": 0.11900468806346917,
      "grad_norm": 0.37530529499053955,
      "learning_rate": 0.0004999954594231088,
      "loss": 2.0922,
      "step": 990
    },
    {
      "epoch": 0.1196057218415675,
      "grad_norm": 0.41129302978515625,
      "learning_rate": 0.0004999951674986401,
      "loss": 1.5781,
      "step": 995
    },
    {
      "epoch": 0.12020675561966583,
      "grad_norm": 0.3869934380054474,
      "learning_rate": 0.0004999948664800704,
      "loss": 1.7422,
      "step": 1000
    },
    {
      "epoch": 0.12020675561966583,
      "eval_loss": 2.4908204078674316,
      "eval_model_preparation_time": 0.0053,
      "eval_runtime": 35.1997,
      "eval_samples_per_second": 4.545,
      "eval_steps_per_second": 1.136,
      "step": 1000
    },
    {
      "epoch": 0.12080778939776415,
      "grad_norm": 0.36643335223197937,
      "learning_rate": 0.0004999945563674105,
      "loss": 1.6797,
      "step": 1005
    },
    {
      "epoch": 0.12140882317586249,
      "grad_norm": 0.45910894870758057,
      "learning_rate": 0.0004999942371606714,
      "loss": 1.7063,
      "step": 1010
    },
    {
      "epoch": 0.12200985695396081,
      "grad_norm": 0.350729763507843,
      "learning_rate": 0.0004999939088598652,
      "loss": 1.6344,
      "step": 1015
    },
    {
      "epoch": 0.12261089073205914,
      "grad_norm": 0.46493440866470337,
      "learning_rate": 0.0004999935714650034,
      "loss": 1.9641,
      "step": 1020
    },
    {
      "epoch": 0.12321192451015747,
      "grad_norm": 0.42726650834083557,
      "learning_rate": 0.0004999932249760984,
      "loss": 1.7094,
      "step": 1025
    },
    {
      "epoch": 0.1238129582882558,
      "grad_norm": 0.28014904260635376,
      "learning_rate": 0.000499992869393163,
      "loss": 1.8516,
      "step": 1030
    },
    {
      "epoch": 0.12441399206635413,
      "grad_norm": 0.4522114098072052,
      "learning_rate": 0.0004999925047162099,
      "loss": 1.3961,
      "step": 1035
    },
    {
      "epoch": 0.12501502584445245,
      "grad_norm": 0.46475955843925476,
      "learning_rate": 0.0004999921309452526,
      "loss": 1.4062,
      "step": 1040
    },
    {
      "epoch": 0.1256160596225508,
      "grad_norm": 0.44490954279899597,
      "learning_rate": 0.0004999917480803044,
      "loss": 1.6719,
      "step": 1045
    },
    {
      "epoch": 0.12621709340064913,
      "grad_norm": 0.40904587507247925,
      "learning_rate": 0.0004999913561213793,
      "loss": 1.7734,
      "step": 1050
    },
    {
      "epoch": 0.12681812717874744,
      "grad_norm": 0.36412525177001953,
      "learning_rate": 0.0004999909550684918,
      "loss": 1.2594,
      "step": 1055
    },
    {
      "epoch": 0.12741916095684577,
      "grad_norm": 0.7560976147651672,
      "learning_rate": 0.0004999905449216563,
      "loss": 1.6047,
      "step": 1060
    },
    {
      "epoch": 0.1280201947349441,
      "grad_norm": 0.5383388996124268,
      "learning_rate": 0.0004999901256808878,
      "loss": 1.6016,
      "step": 1065
    },
    {
      "epoch": 0.12862122851304245,
      "grad_norm": 0.5255587100982666,
      "learning_rate": 0.0004999896973462012,
      "loss": 1.7828,
      "step": 1070
    },
    {
      "epoch": 0.12922226229114075,
      "grad_norm": 0.4830612242221832,
      "learning_rate": 0.0004999892599176127,
      "loss": 1.8781,
      "step": 1075
    },
    {
      "epoch": 0.1298232960692391,
      "grad_norm": 0.3687385618686676,
      "learning_rate": 0.0004999888133951377,
      "loss": 1.4797,
      "step": 1080
    },
    {
      "epoch": 0.13042432984733743,
      "grad_norm": 0.3518010675907135,
      "learning_rate": 0.0004999883577787927,
      "loss": 1.7234,
      "step": 1085
    },
    {
      "epoch": 0.13102536362543574,
      "grad_norm": 0.4522668719291687,
      "learning_rate": 0.0004999878930685943,
      "loss": 1.675,
      "step": 1090
    },
    {
      "epoch": 0.13162639740353407,
      "grad_norm": 0.3153088390827179,
      "learning_rate": 0.0004999874192645592,
      "loss": 1.7328,
      "step": 1095
    },
    {
      "epoch": 0.1322274311816324,
      "grad_norm": 0.4520825147628784,
      "learning_rate": 0.0004999869363667048,
      "loss": 1.925,
      "step": 1100
    },
    {
      "epoch": 0.13282846495973075,
      "grad_norm": 0.3040079176425934,
      "learning_rate": 0.0004999864443750486,
      "loss": 1.6922,
      "step": 1105
    },
    {
      "epoch": 0.13342949873782906,
      "grad_norm": 0.5198135375976562,
      "learning_rate": 0.0004999859432896084,
      "loss": 1.6562,
      "step": 1110
    },
    {
      "epoch": 0.1340305325159274,
      "grad_norm": 0.30772989988327026,
      "learning_rate": 0.0004999854331104028,
      "loss": 1.8078,
      "step": 1115
    },
    {
      "epoch": 0.13463156629402573,
      "grad_norm": 0.39027324318885803,
      "learning_rate": 0.0004999849138374498,
      "loss": 1.625,
      "step": 1120
    },
    {
      "epoch": 0.13523260007212407,
      "grad_norm": 0.4438004195690155,
      "learning_rate": 0.0004999843854707688,
      "loss": 1.5414,
      "step": 1125
    },
    {
      "epoch": 0.13583363385022237,
      "grad_norm": 0.4966782033443451,
      "learning_rate": 0.0004999838480103787,
      "loss": 1.4836,
      "step": 1130
    },
    {
      "epoch": 0.1364346676283207,
      "grad_norm": 0.5602577328681946,
      "learning_rate": 0.0004999833014562992,
      "loss": 1.3961,
      "step": 1135
    },
    {
      "epoch": 0.13703570140641905,
      "grad_norm": 0.5276179909706116,
      "learning_rate": 0.0004999827458085502,
      "loss": 1.8422,
      "step": 1140
    },
    {
      "epoch": 0.13763673518451738,
      "grad_norm": 0.4706065058708191,
      "learning_rate": 0.0004999821810671518,
      "loss": 1.7109,
      "step": 1145
    },
    {
      "epoch": 0.1382377689626157,
      "grad_norm": 0.38341307640075684,
      "learning_rate": 0.0004999816072321245,
      "loss": 1.8859,
      "step": 1150
    },
    {
      "epoch": 0.13883880274071403,
      "grad_norm": 0.5754373073577881,
      "learning_rate": 0.0004999810243034894,
      "loss": 1.8,
      "step": 1155
    },
    {
      "epoch": 0.13943983651881237,
      "grad_norm": 0.5003094673156738,
      "learning_rate": 0.0004999804322812676,
      "loss": 1.6766,
      "step": 1160
    },
    {
      "epoch": 0.14004087029691067,
      "grad_norm": 0.31239280104637146,
      "learning_rate": 0.0004999798311654805,
      "loss": 1.775,
      "step": 1165
    },
    {
      "epoch": 0.140641904075009,
      "grad_norm": 0.3998953700065613,
      "learning_rate": 0.0004999792209561501,
      "loss": 1.7516,
      "step": 1170
    },
    {
      "epoch": 0.14124293785310735,
      "grad_norm": 0.3099336624145508,
      "learning_rate": 0.0004999786016532986,
      "loss": 1.8422,
      "step": 1175
    },
    {
      "epoch": 0.14184397163120568,
      "grad_norm": 0.48160257935523987,
      "learning_rate": 0.0004999779732569485,
      "loss": 1.6062,
      "step": 1180
    },
    {
      "epoch": 0.142445005409304,
      "grad_norm": 0.5494711399078369,
      "learning_rate": 0.0004999773357671227,
      "loss": 1.5906,
      "step": 1185
    },
    {
      "epoch": 0.14304603918740233,
      "grad_norm": 0.7721512913703918,
      "learning_rate": 0.0004999766891838444,
      "loss": 1.7734,
      "step": 1190
    },
    {
      "epoch": 0.14364707296550067,
      "grad_norm": 0.5135265588760376,
      "learning_rate": 0.000499976033507137,
      "loss": 1.4812,
      "step": 1195
    },
    {
      "epoch": 0.144248106743599,
      "grad_norm": 0.7913392186164856,
      "learning_rate": 0.0004999753687370245,
      "loss": 1.5484,
      "step": 1200
    },
    {
      "epoch": 0.144248106743599,
      "eval_loss": 2.5082030296325684,
      "eval_model_preparation_time": 0.0053,
      "eval_runtime": 35.2256,
      "eval_samples_per_second": 4.542,
      "eval_steps_per_second": 1.136,
      "step": 1200
    },
    {
      "epoch": 0.1448491405216973,
      "grad_norm": 0.6069223880767822,
      "learning_rate": 0.0004999746948735308,
      "loss": 1.4484,
      "step": 1205
    },
    {
      "epoch": 0.14545017429979565,
      "grad_norm": 0.4137849807739258,
      "learning_rate": 0.0004999740119166809,
      "loss": 1.6719,
      "step": 1210
    },
    {
      "epoch": 0.14605120807789398,
      "grad_norm": 0.7047042846679688,
      "learning_rate": 0.0004999733198664992,
      "loss": 1.5312,
      "step": 1215
    },
    {
      "epoch": 0.1466522418559923,
      "grad_norm": 0.5389900207519531,
      "learning_rate": 0.0004999726187230111,
      "loss": 1.4297,
      "step": 1220
    },
    {
      "epoch": 0.14725327563409063,
      "grad_norm": 0.5395992994308472,
      "learning_rate": 0.0004999719084862421,
      "loss": 1.6328,
      "step": 1225
    },
    {
      "epoch": 0.14785430941218897,
      "grad_norm": 0.43566471338272095,
      "learning_rate": 0.0004999711891562179,
      "loss": 1.7094,
      "step": 1230
    },
    {
      "epoch": 0.1484553431902873,
      "grad_norm": 0.3409474194049835,
      "learning_rate": 0.0004999704607329648,
      "loss": 1.6656,
      "step": 1235
    },
    {
      "epoch": 0.1490563769683856,
      "grad_norm": 0.5498088002204895,
      "learning_rate": 0.0004999697232165092,
      "loss": 1.6016,
      "step": 1240
    },
    {
      "epoch": 0.14965741074648395,
      "grad_norm": 0.567551851272583,
      "learning_rate": 0.000499968976606878,
      "loss": 1.6828,
      "step": 1245
    },
    {
      "epoch": 0.15025844452458229,
      "grad_norm": 0.4866923987865448,
      "learning_rate": 0.0004999682209040983,
      "loss": 1.6547,
      "step": 1250
    },
    {
      "epoch": 0.15085947830268062,
      "grad_norm": 0.3780736029148102,
      "learning_rate": 0.0004999674561081977,
      "loss": 1.6719,
      "step": 1255
    },
    {
      "epoch": 0.15146051208077893,
      "grad_norm": 0.3219822347164154,
      "learning_rate": 0.0004999666822192039,
      "loss": 1.4195,
      "step": 1260
    },
    {
      "epoch": 0.15206154585887727,
      "grad_norm": 0.3056913912296295,
      "learning_rate": 0.0004999658992371451,
      "loss": 1.7484,
      "step": 1265
    },
    {
      "epoch": 0.1526625796369756,
      "grad_norm": 0.4860096573829651,
      "learning_rate": 0.0004999651071620499,
      "loss": 1.6516,
      "step": 1270
    },
    {
      "epoch": 0.15326361341507394,
      "grad_norm": 0.4047755002975464,
      "learning_rate": 0.0004999643059939469,
      "loss": 1.6984,
      "step": 1275
    },
    {
      "epoch": 0.15386464719317225,
      "grad_norm": 0.27880361676216125,
      "learning_rate": 0.0004999634957328652,
      "loss": 1.8078,
      "step": 1280
    },
    {
      "epoch": 0.15446568097127059,
      "grad_norm": 0.4087715148925781,
      "learning_rate": 0.0004999626763788346,
      "loss": 1.6422,
      "step": 1285
    },
    {
      "epoch": 0.15506671474936892,
      "grad_norm": 0.556612491607666,
      "learning_rate": 0.0004999618479318847,
      "loss": 1.5359,
      "step": 1290
    },
    {
      "epoch": 0.15566774852746723,
      "grad_norm": 0.5415599346160889,
      "learning_rate": 0.0004999610103920457,
      "loss": 1.5641,
      "step": 1295
    },
    {
      "epoch": 0.15626878230556557,
      "grad_norm": 0.48660141229629517,
      "learning_rate": 0.0004999601637593479,
      "loss": 1.5,
      "step": 1300
    },
    {
      "epoch": 0.1568698160836639,
      "grad_norm": 0.5874481797218323,
      "learning_rate": 0.0004999593080338224,
      "loss": 1.3844,
      "step": 1305
    },
    {
      "epoch": 0.15747084986176224,
      "grad_norm": 0.3727753460407257,
      "learning_rate": 0.0004999584432155,
      "loss": 1.8125,
      "step": 1310
    },
    {
      "epoch": 0.15807188363986055,
      "grad_norm": 0.35395169258117676,
      "learning_rate": 0.0004999575693044124,
      "loss": 1.3305,
      "step": 1315
    },
    {
      "epoch": 0.1586729174179589,
      "grad_norm": 0.7356476783752441,
      "learning_rate": 0.0004999566863005912,
      "loss": 1.7078,
      "step": 1320
    },
    {
      "epoch": 0.15927395119605722,
      "grad_norm": 0.4838991165161133,
      "learning_rate": 0.0004999557942040687,
      "loss": 1.5969,
      "step": 1325
    },
    {
      "epoch": 0.15987498497415556,
      "grad_norm": 0.4504292905330658,
      "learning_rate": 0.0004999548930148773,
      "loss": 1.4555,
      "step": 1330
    },
    {
      "epoch": 0.16047601875225387,
      "grad_norm": 0.5174041390419006,
      "learning_rate": 0.0004999539827330497,
      "loss": 1.5266,
      "step": 1335
    },
    {
      "epoch": 0.1610770525303522,
      "grad_norm": 0.42709511518478394,
      "learning_rate": 0.000499953063358619,
      "loss": 1.4344,
      "step": 1340
    },
    {
      "epoch": 0.16167808630845054,
      "grad_norm": 0.34575751423835754,
      "learning_rate": 0.0004999521348916189,
      "loss": 1.5219,
      "step": 1345
    },
    {
      "epoch": 0.16227912008654885,
      "grad_norm": 0.4409041404724121,
      "learning_rate": 0.0004999511973320829,
      "loss": 1.6172,
      "step": 1350
    },
    {
      "epoch": 0.1628801538646472,
      "grad_norm": 0.37874963879585266,
      "learning_rate": 0.0004999502506800452,
      "loss": 1.3156,
      "step": 1355
    },
    {
      "epoch": 0.16348118764274552,
      "grad_norm": 0.39675143361091614,
      "learning_rate": 0.0004999492949355401,
      "loss": 1.7672,
      "step": 1360
    },
    {
      "epoch": 0.16408222142084386,
      "grad_norm": 0.4887191951274872,
      "learning_rate": 0.0004999483300986027,
      "loss": 1.6578,
      "step": 1365
    },
    {
      "epoch": 0.16468325519894217,
      "grad_norm": 0.5052289366722107,
      "learning_rate": 0.000499947356169268,
      "loss": 1.5766,
      "step": 1370
    },
    {
      "epoch": 0.1652842889770405,
      "grad_norm": 0.3420865833759308,
      "learning_rate": 0.000499946373147571,
      "loss": 1.4281,
      "step": 1375
    },
    {
      "epoch": 0.16588532275513884,
      "grad_norm": 0.6112978458404541,
      "learning_rate": 0.0004999453810335479,
      "loss": 1.4234,
      "step": 1380
    },
    {
      "epoch": 0.16648635653323718,
      "grad_norm": 0.46144208312034607,
      "learning_rate": 0.0004999443798272348,
      "loss": 1.4609,
      "step": 1385
    },
    {
      "epoch": 0.1670873903113355,
      "grad_norm": 0.5132108926773071,
      "learning_rate": 0.000499943369528668,
      "loss": 1.5656,
      "step": 1390
    },
    {
      "epoch": 0.16768842408943382,
      "grad_norm": 0.5717546939849854,
      "learning_rate": 0.000499942350137884,
      "loss": 1.3617,
      "step": 1395
    },
    {
      "epoch": 0.16828945786753216,
      "grad_norm": 0.4766351580619812,
      "learning_rate": 0.0004999413216549203,
      "loss": 1.5016,
      "step": 1400
    },
    {
      "epoch": 0.16828945786753216,
      "eval_loss": 2.3990235328674316,
      "eval_model_preparation_time": 0.0053,
      "eval_runtime": 35.19,
      "eval_samples_per_second": 4.547,
      "eval_steps_per_second": 1.137,
      "step": 1400
    },
    {
      "epoch": 0.1688904916456305,
      "grad_norm": 0.43601974844932556,
      "learning_rate": 0.0004999402840798142,
      "loss": 1.4156,
      "step": 1405
    },
    {
      "epoch": 0.1694915254237288,
      "grad_norm": 0.6105599403381348,
      "learning_rate": 0.0004999392374126034,
      "loss": 1.7,
      "step": 1410
    },
    {
      "epoch": 0.17009255920182714,
      "grad_norm": 0.4957026243209839,
      "learning_rate": 0.0004999381816533259,
      "loss": 1.7969,
      "step": 1415
    },
    {
      "epoch": 0.17069359297992548,
      "grad_norm": 0.44195666909217834,
      "learning_rate": 0.0004999371168020201,
      "loss": 1.4375,
      "step": 1420
    },
    {
      "epoch": 0.1712946267580238,
      "grad_norm": 0.45855048298835754,
      "learning_rate": 0.0004999360428587249,
      "loss": 1.6141,
      "step": 1425
    },
    {
      "epoch": 0.17189566053612212,
      "grad_norm": 0.6269901990890503,
      "learning_rate": 0.0004999349598234792,
      "loss": 1.3953,
      "step": 1430
    },
    {
      "epoch": 0.17249669431422046,
      "grad_norm": 0.3805680274963379,
      "learning_rate": 0.0004999338676963225,
      "loss": 1.6484,
      "step": 1435
    },
    {
      "epoch": 0.1730977280923188,
      "grad_norm": 0.6604627966880798,
      "learning_rate": 0.0004999327664772945,
      "loss": 1.5969,
      "step": 1440
    },
    {
      "epoch": 0.1736987618704171,
      "grad_norm": 0.4411623179912567,
      "learning_rate": 0.0004999316561664353,
      "loss": 1.2609,
      "step": 1445
    },
    {
      "epoch": 0.17429979564851544,
      "grad_norm": 0.5301747918128967,
      "learning_rate": 0.0004999305367637852,
      "loss": 1.6141,
      "step": 1450
    },
    {
      "epoch": 0.17490082942661378,
      "grad_norm": 0.5128594040870667,
      "learning_rate": 0.000499929408269385,
      "loss": 1.6187,
      "step": 1455
    },
    {
      "epoch": 0.17550186320471212,
      "grad_norm": 0.596217155456543,
      "learning_rate": 0.0004999282706832758,
      "loss": 1.4531,
      "step": 1460
    },
    {
      "epoch": 0.17610289698281043,
      "grad_norm": 0.45486292243003845,
      "learning_rate": 0.0004999271240054987,
      "loss": 1.4012,
      "step": 1465
    },
    {
      "epoch": 0.17670393076090876,
      "grad_norm": 0.6031058430671692,
      "learning_rate": 0.0004999259682360957,
      "loss": 1.6203,
      "step": 1470
    },
    {
      "epoch": 0.1773049645390071,
      "grad_norm": 0.4107096493244171,
      "learning_rate": 0.0004999248033751088,
      "loss": 1.7312,
      "step": 1475
    },
    {
      "epoch": 0.17790599831710543,
      "grad_norm": 0.46700888872146606,
      "learning_rate": 0.0004999236294225803,
      "loss": 1.5234,
      "step": 1480
    },
    {
      "epoch": 0.17850703209520374,
      "grad_norm": 0.7690737247467041,
      "learning_rate": 0.000499922446378553,
      "loss": 1.307,
      "step": 1485
    },
    {
      "epoch": 0.17910806587330208,
      "grad_norm": 0.5420579314231873,
      "learning_rate": 0.0004999212542430698,
      "loss": 1.6562,
      "step": 1490
    },
    {
      "epoch": 0.17970909965140042,
      "grad_norm": 0.4624311625957489,
      "learning_rate": 0.0004999200530161742,
      "loss": 1.3234,
      "step": 1495
    },
    {
      "epoch": 0.18031013342949873,
      "grad_norm": 0.4610016345977783,
      "learning_rate": 0.0004999188426979097,
      "loss": 1.5516,
      "step": 1500
    },
    {
      "epoch": 0.18091116720759706,
      "grad_norm": 0.5131213068962097,
      "learning_rate": 0.0004999176232883206,
      "loss": 1.5867,
      "step": 1505
    },
    {
      "epoch": 0.1815122009856954,
      "grad_norm": 0.5673689842224121,
      "learning_rate": 0.0004999163947874511,
      "loss": 1.5078,
      "step": 1510
    },
    {
      "epoch": 0.18211323476379374,
      "grad_norm": 0.7008316516876221,
      "learning_rate": 0.000499915157195346,
      "loss": 1.5562,
      "step": 1515
    },
    {
      "epoch": 0.18271426854189204,
      "grad_norm": 0.5652767419815063,
      "learning_rate": 0.00049991391051205,
      "loss": 1.4469,
      "step": 1520
    },
    {
      "epoch": 0.18331530231999038,
      "grad_norm": 0.5506184101104736,
      "learning_rate": 0.0004999126547376089,
      "loss": 1.4531,
      "step": 1525
    },
    {
      "epoch": 0.18391633609808872,
      "grad_norm": 0.5806117057800293,
      "learning_rate": 0.000499911389872068,
      "loss": 1.7594,
      "step": 1530
    },
    {
      "epoch": 0.18451736987618705,
      "grad_norm": 0.5860136151313782,
      "learning_rate": 0.0004999101159154736,
      "loss": 1.5562,
      "step": 1535
    },
    {
      "epoch": 0.18511840365428536,
      "grad_norm": 0.5575783252716064,
      "learning_rate": 0.000499908832867872,
      "loss": 1.6391,
      "step": 1540
    },
    {
      "epoch": 0.1857194374323837,
      "grad_norm": 0.3992920219898224,
      "learning_rate": 0.0004999075407293096,
      "loss": 1.3859,
      "step": 1545
    },
    {
      "epoch": 0.18632047121048204,
      "grad_norm": 0.8294938206672668,
      "learning_rate": 0.0004999062394998336,
      "loss": 1.25,
      "step": 1550
    },
    {
      "epoch": 0.18692150498858034,
      "grad_norm": 0.6560512185096741,
      "learning_rate": 0.0004999049291794915,
      "loss": 1.4453,
      "step": 1555
    },
    {
      "epoch": 0.18752253876667868,
      "grad_norm": 0.5583436489105225,
      "learning_rate": 0.0004999036097683307,
      "loss": 1.3969,
      "step": 1560
    },
    {
      "epoch": 0.18812357254477702,
      "grad_norm": 0.6256234645843506,
      "learning_rate": 0.0004999022812663993,
      "loss": 1.518,
      "step": 1565
    },
    {
      "epoch": 0.18872460632287535,
      "grad_norm": 0.5769176483154297,
      "learning_rate": 0.0004999009436737457,
      "loss": 1.6609,
      "step": 1570
    },
    {
      "epoch": 0.18932564010097366,
      "grad_norm": 0.6486324071884155,
      "learning_rate": 0.0004998995969904183,
      "loss": 1.3172,
      "step": 1575
    },
    {
      "epoch": 0.189926673879072,
      "grad_norm": 0.34935474395751953,
      "learning_rate": 0.0004998982412164663,
      "loss": 1.5562,
      "step": 1580
    },
    {
      "epoch": 0.19052770765717034,
      "grad_norm": 0.5806995630264282,
      "learning_rate": 0.000499896876351939,
      "loss": 1.6219,
      "step": 1585
    },
    {
      "epoch": 0.19112874143526867,
      "grad_norm": 0.6906558275222778,
      "learning_rate": 0.0004998955023968862,
      "loss": 1.5172,
      "step": 1590
    },
    {
      "epoch": 0.19172977521336698,
      "grad_norm": 0.49730750918388367,
      "learning_rate": 0.0004998941193513575,
      "loss": 1.6797,
      "step": 1595
    },
    {
      "epoch": 0.19233080899146532,
      "grad_norm": 0.5871158242225647,
      "learning_rate": 0.0004998927272154036,
      "loss": 1.6125,
      "step": 1600
    },
    {
      "epoch": 0.19233080899146532,
      "eval_loss": 2.360156297683716,
      "eval_model_preparation_time": 0.0053,
      "eval_runtime": 35.1647,
      "eval_samples_per_second": 4.55,
      "eval_steps_per_second": 1.138,
      "step": 1600
    },
    {
      "epoch": 0.19293184276956366,
      "grad_norm": 0.3994157910346985,
      "learning_rate": 0.000499891325989075,
      "loss": 1.3305,
      "step": 1605
    },
    {
      "epoch": 0.193532876547662,
      "grad_norm": 0.3497280180454254,
      "learning_rate": 0.0004998899156724224,
      "loss": 1.3531,
      "step": 1610
    },
    {
      "epoch": 0.1941339103257603,
      "grad_norm": 0.4835513234138489,
      "learning_rate": 0.0004998884962654976,
      "loss": 1.293,
      "step": 1615
    },
    {
      "epoch": 0.19473494410385864,
      "grad_norm": 0.4717245101928711,
      "learning_rate": 0.0004998870677683519,
      "loss": 1.3742,
      "step": 1620
    },
    {
      "epoch": 0.19533597788195697,
      "grad_norm": 0.3917827308177948,
      "learning_rate": 0.0004998856301810373,
      "loss": 1.5719,
      "step": 1625
    },
    {
      "epoch": 0.19593701166005528,
      "grad_norm": 0.4725429117679596,
      "learning_rate": 0.0004998841835036061,
      "loss": 1.3859,
      "step": 1630
    },
    {
      "epoch": 0.19653804543815362,
      "grad_norm": 0.4728795289993286,
      "learning_rate": 0.0004998827277361111,
      "loss": 1.4203,
      "step": 1635
    },
    {
      "epoch": 0.19713907921625196,
      "grad_norm": 0.6246328949928284,
      "learning_rate": 0.000499881262878605,
      "loss": 1.7719,
      "step": 1640
    },
    {
      "epoch": 0.1977401129943503,
      "grad_norm": 0.7019891738891602,
      "learning_rate": 0.0004998797889311413,
      "loss": 1.3781,
      "step": 1645
    },
    {
      "epoch": 0.1983411467724486,
      "grad_norm": 0.2940036654472351,
      "learning_rate": 0.0004998783058937735,
      "loss": 1.4148,
      "step": 1650
    },
    {
      "epoch": 0.19894218055054694,
      "grad_norm": 0.434410959482193,
      "learning_rate": 0.0004998768137665556,
      "loss": 1.6094,
      "step": 1655
    },
    {
      "epoch": 0.19954321432864527,
      "grad_norm": 0.5853382349014282,
      "learning_rate": 0.0004998753125495418,
      "loss": 1.4125,
      "step": 1660
    },
    {
      "epoch": 0.2001442481067436,
      "grad_norm": 0.5105974078178406,
      "learning_rate": 0.0004998738022427867,
      "loss": 1.3313,
      "step": 1665
    },
    {
      "epoch": 0.20074528188484192,
      "grad_norm": 0.4266336262226105,
      "learning_rate": 0.0004998722828463455,
      "loss": 1.5953,
      "step": 1670
    },
    {
      "epoch": 0.20134631566294026,
      "grad_norm": 0.4918626844882965,
      "learning_rate": 0.0004998707543602731,
      "loss": 1.8383,
      "step": 1675
    },
    {
      "epoch": 0.2019473494410386,
      "grad_norm": 0.4804850220680237,
      "learning_rate": 0.0004998692167846253,
      "loss": 1.1484,
      "step": 1680
    },
    {
      "epoch": 0.20254838321913693,
      "grad_norm": 0.5131824612617493,
      "learning_rate": 0.0004998676701194581,
      "loss": 1.7109,
      "step": 1685
    },
    {
      "epoch": 0.20314941699723524,
      "grad_norm": 0.4895535111427307,
      "learning_rate": 0.0004998661143648277,
      "loss": 1.7453,
      "step": 1690
    },
    {
      "epoch": 0.20375045077533357,
      "grad_norm": 0.4180288314819336,
      "learning_rate": 0.0004998645495207906,
      "loss": 1.0766,
      "step": 1695
    },
    {
      "epoch": 0.2043514845534319,
      "grad_norm": 0.4888496696949005,
      "learning_rate": 0.0004998629755874037,
      "loss": 1.5359,
      "step": 1700
    },
    {
      "epoch": 0.20495251833153022,
      "grad_norm": 0.666147768497467,
      "learning_rate": 0.0004998613925647245,
      "loss": 1.5609,
      "step": 1705
    },
    {
      "epoch": 0.20555355210962856,
      "grad_norm": 0.563382625579834,
      "learning_rate": 0.0004998598004528103,
      "loss": 1.4187,
      "step": 1710
    },
    {
      "epoch": 0.2061545858877269,
      "grad_norm": 0.619296669960022,
      "learning_rate": 0.0004998581992517192,
      "loss": 1.3367,
      "step": 1715
    },
    {
      "epoch": 0.20675561966582523,
      "grad_norm": 0.928014874458313,
      "learning_rate": 0.0004998565889615096,
      "loss": 1.4094,
      "step": 1720
    },
    {
      "epoch": 0.20735665344392354,
      "grad_norm": 0.4932372272014618,
      "learning_rate": 0.0004998549695822397,
      "loss": 1.3719,
      "step": 1725
    },
    {
      "epoch": 0.20795768722202188,
      "grad_norm": 0.6022034287452698,
      "learning_rate": 0.0004998533411139685,
      "loss": 1.5781,
      "step": 1730
    },
    {
      "epoch": 0.2085587210001202,
      "grad_norm": 0.41716283559799194,
      "learning_rate": 0.0004998517035567554,
      "loss": 1.1914,
      "step": 1735
    },
    {
      "epoch": 0.20915975477821855,
      "grad_norm": 0.4988159239292145,
      "learning_rate": 0.0004998500569106599,
      "loss": 1.475,
      "step": 1740
    },
    {
      "epoch": 0.20976078855631686,
      "grad_norm": 0.4242478907108307,
      "learning_rate": 0.0004998484011757419,
      "loss": 1.2859,
      "step": 1745
    },
    {
      "epoch": 0.2103618223344152,
      "grad_norm": 0.5382992625236511,
      "learning_rate": 0.0004998467363520617,
      "loss": 1.3687,
      "step": 1750
    },
    {
      "epoch": 0.21096285611251353,
      "grad_norm": 0.31303003430366516,
      "learning_rate": 0.0004998450624396797,
      "loss": 1.9281,
      "step": 1755
    },
    {
      "epoch": 0.21156388989061184,
      "grad_norm": 0.5793948173522949,
      "learning_rate": 0.0004998433794386569,
      "loss": 1.457,
      "step": 1760
    },
    {
      "epoch": 0.21216492366871018,
      "grad_norm": 0.48824676871299744,
      "learning_rate": 0.0004998416873490544,
      "loss": 1.5359,
      "step": 1765
    },
    {
      "epoch": 0.2127659574468085,
      "grad_norm": 0.5384695529937744,
      "learning_rate": 0.000499839986170934,
      "loss": 1.4461,
      "step": 1770
    },
    {
      "epoch": 0.21336699122490685,
      "grad_norm": 0.5212387442588806,
      "learning_rate": 0.0004998382759043574,
      "loss": 1.2844,
      "step": 1775
    },
    {
      "epoch": 0.21396802500300516,
      "grad_norm": 0.5552918910980225,
      "learning_rate": 0.0004998365565493868,
      "loss": 1.5516,
      "step": 1780
    },
    {
      "epoch": 0.2145690587811035,
      "grad_norm": 0.5672168135643005,
      "learning_rate": 0.0004998348281060848,
      "loss": 1.6297,
      "step": 1785
    },
    {
      "epoch": 0.21517009255920183,
      "grad_norm": 0.620464026927948,
      "learning_rate": 0.0004998330905745143,
      "loss": 1.5047,
      "step": 1790
    },
    {
      "epoch": 0.21577112633730017,
      "grad_norm": 0.5900077819824219,
      "learning_rate": 0.0004998313439547384,
      "loss": 1.2367,
      "step": 1795
    },
    {
      "epoch": 0.21637216011539848,
      "grad_norm": 0.5305217504501343,
      "learning_rate": 0.0004998295882468209,
      "loss": 1.5906,
      "step": 1800
    },
    {
      "epoch": 0.21637216011539848,
      "eval_loss": 2.288867235183716,
      "eval_model_preparation_time": 0.0053,
      "eval_runtime": 35.2003,
      "eval_samples_per_second": 4.545,
      "eval_steps_per_second": 1.136,
      "step": 1800
    },
    {
      "epoch": 0.2169731938934968,
      "grad_norm": 0.5836020112037659,
      "learning_rate": 0.0004998278234508253,
      "loss": 1.4891,
      "step": 1805
    },
    {
      "epoch": 0.21757422767159515,
      "grad_norm": 0.3793884813785553,
      "learning_rate": 0.0004998260495668161,
      "loss": 1.3328,
      "step": 1810
    },
    {
      "epoch": 0.21817526144969349,
      "grad_norm": 0.5394117832183838,
      "learning_rate": 0.0004998242665948577,
      "loss": 1.368,
      "step": 1815
    },
    {
      "epoch": 0.2187762952277918,
      "grad_norm": 0.39613473415374756,
      "learning_rate": 0.0004998224745350148,
      "loss": 1.2285,
      "step": 1820
    },
    {
      "epoch": 0.21937732900589013,
      "grad_norm": 0.543116569519043,
      "learning_rate": 0.0004998206733873529,
      "loss": 1.5078,
      "step": 1825
    },
    {
      "epoch": 0.21997836278398847,
      "grad_norm": 0.4901551306247711,
      "learning_rate": 0.0004998188631519375,
      "loss": 1.4516,
      "step": 1830
    },
    {
      "epoch": 0.22057939656208678,
      "grad_norm": 0.5067916512489319,
      "learning_rate": 0.0004998170438288342,
      "loss": 1.5719,
      "step": 1835
    },
    {
      "epoch": 0.2211804303401851,
      "grad_norm": 0.4343029856681824,
      "learning_rate": 0.0004998152154181093,
      "loss": 1.3766,
      "step": 1840
    },
    {
      "epoch": 0.22178146411828345,
      "grad_norm": 0.5296164155006409,
      "learning_rate": 0.0004998133779198293,
      "loss": 1.3625,
      "step": 1845
    },
    {
      "epoch": 0.2223824978963818,
      "grad_norm": 0.4429774284362793,
      "learning_rate": 0.0004998115313340611,
      "loss": 1.3891,
      "step": 1850
    },
    {
      "epoch": 0.2229835316744801,
      "grad_norm": 0.5772582292556763,
      "learning_rate": 0.0004998096756608719,
      "loss": 1.5437,
      "step": 1855
    },
    {
      "epoch": 0.22358456545257843,
      "grad_norm": 0.5951064825057983,
      "learning_rate": 0.0004998078109003291,
      "loss": 1.4672,
      "step": 1860
    },
    {
      "epoch": 0.22418559923067677,
      "grad_norm": 0.3261686861515045,
      "learning_rate": 0.0004998059370525006,
      "loss": 1.5063,
      "step": 1865
    },
    {
      "epoch": 0.2247866330087751,
      "grad_norm": 0.3098331689834595,
      "learning_rate": 0.0004998040541174545,
      "loss": 1.5094,
      "step": 1870
    },
    {
      "epoch": 0.22538766678687341,
      "grad_norm": 0.8590214252471924,
      "learning_rate": 0.0004998021620952593,
      "loss": 1.3977,
      "step": 1875
    },
    {
      "epoch": 0.22598870056497175,
      "grad_norm": 0.5078855752944946,
      "learning_rate": 0.0004998002609859839,
      "loss": 1.2789,
      "step": 1880
    },
    {
      "epoch": 0.2265897343430701,
      "grad_norm": 0.4515461027622223,
      "learning_rate": 0.0004997983507896976,
      "loss": 1.368,
      "step": 1885
    },
    {
      "epoch": 0.2271907681211684,
      "grad_norm": 0.4937264025211334,
      "learning_rate": 0.0004997964315064695,
      "loss": 1.1953,
      "step": 1890
    },
    {
      "epoch": 0.22779180189926673,
      "grad_norm": 0.6028769612312317,
      "learning_rate": 0.0004997945031363697,
      "loss": 1.4859,
      "step": 1895
    },
    {
      "epoch": 0.22839283567736507,
      "grad_norm": 0.4746128022670746,
      "learning_rate": 0.0004997925656794683,
      "loss": 1.6016,
      "step": 1900
    },
    {
      "epoch": 0.2289938694554634,
      "grad_norm": 0.519091010093689,
      "learning_rate": 0.0004997906191358358,
      "loss": 1.3906,
      "step": 1905
    },
    {
      "epoch": 0.22959490323356171,
      "grad_norm": 0.4584903419017792,
      "learning_rate": 0.0004997886635055429,
      "loss": 1.3258,
      "step": 1910
    },
    {
      "epoch": 0.23019593701166005,
      "grad_norm": 0.7446622252464294,
      "learning_rate": 0.0004997866987886608,
      "loss": 1.2141,
      "step": 1915
    },
    {
      "epoch": 0.2307969707897584,
      "grad_norm": 0.5405495166778564,
      "learning_rate": 0.0004997847249852609,
      "loss": 1.4359,
      "step": 1920
    },
    {
      "epoch": 0.23139800456785672,
      "grad_norm": 0.38187775015830994,
      "learning_rate": 0.0004997827420954152,
      "loss": 1.7219,
      "step": 1925
    },
    {
      "epoch": 0.23199903834595503,
      "grad_norm": 0.503364622592926,
      "learning_rate": 0.0004997807501191957,
      "loss": 1.3586,
      "step": 1930
    },
    {
      "epoch": 0.23260007212405337,
      "grad_norm": 0.43855008482933044,
      "learning_rate": 0.0004997787490566749,
      "loss": 1.7625,
      "step": 1935
    },
    {
      "epoch": 0.2332011059021517,
      "grad_norm": 0.4955185651779175,
      "learning_rate": 0.0004997767389079255,
      "loss": 1.2281,
      "step": 1940
    },
    {
      "epoch": 0.23380213968025004,
      "grad_norm": 0.7726651430130005,
      "learning_rate": 0.0004997747196730206,
      "loss": 1.5445,
      "step": 1945
    },
    {
      "epoch": 0.23440317345834835,
      "grad_norm": 0.38199684023857117,
      "learning_rate": 0.000499772691352034,
      "loss": 1.4203,
      "step": 1950
    },
    {
      "epoch": 0.2350042072364467,
      "grad_norm": 0.4838792383670807,
      "learning_rate": 0.000499770653945039,
      "loss": 1.2484,
      "step": 1955
    },
    {
      "epoch": 0.23560524101454502,
      "grad_norm": 0.43874993920326233,
      "learning_rate": 0.00049976860745211,
      "loss": 1.3594,
      "step": 1960
    },
    {
      "epoch": 0.23620627479264333,
      "grad_norm": 0.4992177188396454,
      "learning_rate": 0.0004997665518733215,
      "loss": 1.1977,
      "step": 1965
    },
    {
      "epoch": 0.23680730857074167,
      "grad_norm": 0.526907742023468,
      "learning_rate": 0.000499764487208748,
      "loss": 1.1609,
      "step": 1970
    },
    {
      "epoch": 0.23740834234884,
      "grad_norm": 0.599902868270874,
      "learning_rate": 0.000499762413458465,
      "loss": 1.4203,
      "step": 1975
    },
    {
      "epoch": 0.23800937612693834,
      "grad_norm": 0.42601317167282104,
      "learning_rate": 0.0004997603306225475,
      "loss": 1.1516,
      "step": 1980
    },
    {
      "epoch": 0.23861040990503665,
      "grad_norm": 0.3787403404712677,
      "learning_rate": 0.0004997582387010715,
      "loss": 1.3391,
      "step": 1985
    },
    {
      "epoch": 0.239211443683135,
      "grad_norm": 0.5586139559745789,
      "learning_rate": 0.0004997561376941131,
      "loss": 1.6656,
      "step": 1990
    },
    {
      "epoch": 0.23981247746123333,
      "grad_norm": 0.44761109352111816,
      "learning_rate": 0.0004997540276017487,
      "loss": 1.5828,
      "step": 1995
    },
    {
      "epoch": 0.24041351123933166,
      "grad_norm": 0.4291538894176483,
      "learning_rate": 0.000499751908424055,
      "loss": 1.4539,
      "step": 2000
    },
    {
      "epoch": 0.24041351123933166,
      "eval_loss": 2.2626953125,
      "eval_model_preparation_time": 0.0053,
      "eval_runtime": 35.2012,
      "eval_samples_per_second": 4.545,
      "eval_steps_per_second": 1.136,
      "step": 2000
    },
    {
      "epoch": 0.24101454501742997,
      "grad_norm": 0.46680477261543274,
      "learning_rate": 0.0004997497801611093,
      "loss": 1.2609,
      "step": 2005
    },
    {
      "epoch": 0.2416155787955283,
      "grad_norm": 0.42086416482925415,
      "learning_rate": 0.0004997476428129887,
      "loss": 1.1609,
      "step": 2010
    },
    {
      "epoch": 0.24221661257362664,
      "grad_norm": 0.7524279952049255,
      "learning_rate": 0.0004997454963797713,
      "loss": 1.0633,
      "step": 2015
    },
    {
      "epoch": 0.24281764635172498,
      "grad_norm": 0.43722498416900635,
      "learning_rate": 0.0004997433408615349,
      "loss": 1.2969,
      "step": 2020
    },
    {
      "epoch": 0.2434186801298233,
      "grad_norm": 0.2848932147026062,
      "learning_rate": 0.0004997411762583581,
      "loss": 1.2063,
      "step": 2025
    },
    {
      "epoch": 0.24401971390792163,
      "grad_norm": 0.4349381923675537,
      "learning_rate": 0.0004997390025703194,
      "loss": 1.3625,
      "step": 2030
    },
    {
      "epoch": 0.24462074768601996,
      "grad_norm": 0.4666562080383301,
      "learning_rate": 0.0004997368197974982,
      "loss": 1.4164,
      "step": 2035
    },
    {
      "epoch": 0.24522178146411827,
      "grad_norm": 0.5730391144752502,
      "learning_rate": 0.0004997346279399736,
      "loss": 1.1633,
      "step": 2040
    },
    {
      "epoch": 0.2458228152422166,
      "grad_norm": 0.5395126938819885,
      "learning_rate": 0.0004997324269978255,
      "loss": 1.2398,
      "step": 2045
    },
    {
      "epoch": 0.24642384902031494,
      "grad_norm": 0.3828608989715576,
      "learning_rate": 0.000499730216971134,
      "loss": 1.0594,
      "step": 2050
    },
    {
      "epoch": 0.24702488279841328,
      "grad_norm": 0.796903133392334,
      "learning_rate": 0.0004997279978599794,
      "loss": 1.3055,
      "step": 2055
    },
    {
      "epoch": 0.2476259165765116,
      "grad_norm": 0.35091638565063477,
      "learning_rate": 0.0004997257696644424,
      "loss": 1.2023,
      "step": 2060
    },
    {
      "epoch": 0.24822695035460993,
      "grad_norm": 0.46753543615341187,
      "learning_rate": 0.000499723532384604,
      "loss": 1.3203,
      "step": 2065
    },
    {
      "epoch": 0.24882798413270826,
      "grad_norm": 0.5231248736381531,
      "learning_rate": 0.0004997212860205459,
      "loss": 1.3438,
      "step": 2070
    },
    {
      "epoch": 0.2494290179108066,
      "grad_norm": 0.470639169216156,
      "learning_rate": 0.0004997190305723495,
      "loss": 1.3031,
      "step": 2075
    },
    {
      "epoch": 0.2500300516889049,
      "grad_norm": 0.4669177234172821,
      "learning_rate": 0.000499716766040097,
      "loss": 1.5188,
      "step": 2080
    },
    {
      "epoch": 0.25063108546700325,
      "grad_norm": 0.5113319754600525,
      "learning_rate": 0.0004997144924238706,
      "loss": 1.0992,
      "step": 2085
    },
    {
      "epoch": 0.2512321192451016,
      "grad_norm": 0.5395264625549316,
      "learning_rate": 0.0004997122097237533,
      "loss": 1.3281,
      "step": 2090
    },
    {
      "epoch": 0.2518331530231999,
      "grad_norm": 0.47676244378089905,
      "learning_rate": 0.0004997099179398279,
      "loss": 1.2898,
      "step": 2095
    },
    {
      "epoch": 0.25243418680129825,
      "grad_norm": 0.3385642468929291,
      "learning_rate": 0.0004997076170721778,
      "loss": 1.2078,
      "step": 2100
    },
    {
      "epoch": 0.25303522057939654,
      "grad_norm": 0.3868078887462616,
      "learning_rate": 0.0004997053071208868,
      "loss": 1.4563,
      "step": 2105
    },
    {
      "epoch": 0.2536362543574949,
      "grad_norm": 0.437321275472641,
      "learning_rate": 0.0004997029880860389,
      "loss": 1.3977,
      "step": 2110
    },
    {
      "epoch": 0.2542372881355932,
      "grad_norm": 0.6515981554985046,
      "learning_rate": 0.0004997006599677183,
      "loss": 1.2461,
      "step": 2115
    },
    {
      "epoch": 0.25483832191369155,
      "grad_norm": 0.3654949367046356,
      "learning_rate": 0.0004996983227660099,
      "loss": 1.4187,
      "step": 2120
    },
    {
      "epoch": 0.2554393556917899,
      "grad_norm": 0.5203860998153687,
      "learning_rate": 0.0004996959764809987,
      "loss": 1.4328,
      "step": 2125
    },
    {
      "epoch": 0.2560403894698882,
      "grad_norm": 0.5454062223434448,
      "learning_rate": 0.00049969362111277,
      "loss": 1.5125,
      "step": 2130
    },
    {
      "epoch": 0.25664142324798656,
      "grad_norm": 0.5460174679756165,
      "learning_rate": 0.0004996912566614094,
      "loss": 1.4344,
      "step": 2135
    },
    {
      "epoch": 0.2572424570260849,
      "grad_norm": 0.4798714816570282,
      "learning_rate": 0.000499688883127003,
      "loss": 1.1953,
      "step": 2140
    },
    {
      "epoch": 0.2578434908041832,
      "grad_norm": 0.679547905921936,
      "learning_rate": 0.0004996865005096372,
      "loss": 1.2688,
      "step": 2145
    },
    {
      "epoch": 0.2584445245822815,
      "grad_norm": 0.42334336042404175,
      "learning_rate": 0.0004996841088093985,
      "loss": 1.1516,
      "step": 2150
    },
    {
      "epoch": 0.25904555836037985,
      "grad_norm": 0.4171724021434784,
      "learning_rate": 0.000499681708026374,
      "loss": 1.0961,
      "step": 2155
    },
    {
      "epoch": 0.2596465921384782,
      "grad_norm": 0.6091195940971375,
      "learning_rate": 0.0004996792981606511,
      "loss": 1.0164,
      "step": 2160
    },
    {
      "epoch": 0.2602476259165765,
      "grad_norm": 0.7312507033348083,
      "learning_rate": 0.0004996768792123173,
      "loss": 1.3031,
      "step": 2165
    },
    {
      "epoch": 0.26084865969467486,
      "grad_norm": 0.8120207786560059,
      "learning_rate": 0.0004996744511814609,
      "loss": 1.1641,
      "step": 2170
    },
    {
      "epoch": 0.2614496934727732,
      "grad_norm": 0.4702399969100952,
      "learning_rate": 0.0004996720140681699,
      "loss": 1.2805,
      "step": 2175
    },
    {
      "epoch": 0.2620507272508715,
      "grad_norm": 0.45239925384521484,
      "learning_rate": 0.0004996695678725331,
      "loss": 1.5539,
      "step": 2180
    },
    {
      "epoch": 0.2626517610289698,
      "grad_norm": 0.6370692253112793,
      "learning_rate": 0.0004996671125946394,
      "loss": 1.2156,
      "step": 2185
    },
    {
      "epoch": 0.26325279480706815,
      "grad_norm": 0.6115698218345642,
      "learning_rate": 0.0004996646482345781,
      "loss": 1.2891,
      "step": 2190
    },
    {
      "epoch": 0.2638538285851665,
      "grad_norm": 0.611488401889801,
      "learning_rate": 0.0004996621747924391,
      "loss": 1.3023,
      "step": 2195
    },
    {
      "epoch": 0.2644548623632648,
      "grad_norm": 0.6977550983428955,
      "learning_rate": 0.0004996596922683122,
      "loss": 1.3555,
      "step": 2200
    },
    {
      "epoch": 0.2644548623632648,
      "eval_loss": 2.2613282203674316,
      "eval_model_preparation_time": 0.0053,
      "eval_runtime": 35.2156,
      "eval_samples_per_second": 4.543,
      "eval_steps_per_second": 1.136,
      "step": 2200
    },
    {
      "epoch": 0.26505589614136316,
      "grad_norm": 0.6270340085029602,
      "learning_rate": 0.0004996572006622876,
      "loss": 1.5938,
      "step": 2205
    },
    {
      "epoch": 0.2656569299194615,
      "grad_norm": 0.5670061707496643,
      "learning_rate": 0.0004996546999744561,
      "loss": 1.5016,
      "step": 2210
    },
    {
      "epoch": 0.26625796369755983,
      "grad_norm": 0.38163918256759644,
      "learning_rate": 0.0004996521902049086,
      "loss": 1.2812,
      "step": 2215
    },
    {
      "epoch": 0.2668589974756581,
      "grad_norm": 0.45828545093536377,
      "learning_rate": 0.0004996496713537365,
      "loss": 1.3023,
      "step": 2220
    },
    {
      "epoch": 0.26746003125375645,
      "grad_norm": 0.4318217933177948,
      "learning_rate": 0.0004996471434210312,
      "loss": 1.6039,
      "step": 2225
    },
    {
      "epoch": 0.2680610650318548,
      "grad_norm": 0.5099067091941833,
      "learning_rate": 0.0004996446064068848,
      "loss": 1.5562,
      "step": 2230
    },
    {
      "epoch": 0.2686620988099531,
      "grad_norm": 0.7253368496894836,
      "learning_rate": 0.0004996420603113897,
      "loss": 1.2523,
      "step": 2235
    },
    {
      "epoch": 0.26926313258805146,
      "grad_norm": 0.6101372838020325,
      "learning_rate": 0.0004996395051346384,
      "loss": 1.4125,
      "step": 2240
    },
    {
      "epoch": 0.2698641663661498,
      "grad_norm": 0.5073166489601135,
      "learning_rate": 0.0004996369408767238,
      "loss": 1.1109,
      "step": 2245
    },
    {
      "epoch": 0.27046520014424813,
      "grad_norm": 0.4978417456150055,
      "learning_rate": 0.0004996343675377393,
      "loss": 1.3438,
      "step": 2250
    },
    {
      "epoch": 0.2710662339223464,
      "grad_norm": 0.695686936378479,
      "learning_rate": 0.0004996317851177784,
      "loss": 1.0445,
      "step": 2255
    },
    {
      "epoch": 0.27166726770044475,
      "grad_norm": 0.5276048183441162,
      "learning_rate": 0.000499629193616935,
      "loss": 1.2703,
      "step": 2260
    },
    {
      "epoch": 0.2722683014785431,
      "grad_norm": 0.7686821222305298,
      "learning_rate": 0.0004996265930353036,
      "loss": 1.2656,
      "step": 2265
    },
    {
      "epoch": 0.2728693352566414,
      "grad_norm": 0.673497200012207,
      "learning_rate": 0.0004996239833729786,
      "loss": 1.4055,
      "step": 2270
    },
    {
      "epoch": 0.27347036903473976,
      "grad_norm": 0.4770069122314453,
      "learning_rate": 0.000499621364630055,
      "loss": 1.1227,
      "step": 2275
    },
    {
      "epoch": 0.2740714028128381,
      "grad_norm": 0.630565881729126,
      "learning_rate": 0.000499618736806628,
      "loss": 1.293,
      "step": 2280
    },
    {
      "epoch": 0.27467243659093643,
      "grad_norm": 0.5288265943527222,
      "learning_rate": 0.0004996160999027933,
      "loss": 1.5109,
      "step": 2285
    },
    {
      "epoch": 0.27527347036903477,
      "grad_norm": 0.35486194491386414,
      "learning_rate": 0.0004996134539186469,
      "loss": 1.5078,
      "step": 2290
    },
    {
      "epoch": 0.27587450414713305,
      "grad_norm": 0.5654587745666504,
      "learning_rate": 0.0004996107988542847,
      "loss": 1.625,
      "step": 2295
    },
    {
      "epoch": 0.2764755379252314,
      "grad_norm": 0.40694040060043335,
      "learning_rate": 0.0004996081347098037,
      "loss": 1.4531,
      "step": 2300
    },
    {
      "epoch": 0.2770765717033297,
      "grad_norm": 0.5765879154205322,
      "learning_rate": 0.0004996054614853005,
      "loss": 1.343,
      "step": 2305
    },
    {
      "epoch": 0.27767760548142806,
      "grad_norm": 0.49710384011268616,
      "learning_rate": 0.0004996027791808725,
      "loss": 1.3266,
      "step": 2310
    },
    {
      "epoch": 0.2782786392595264,
      "grad_norm": 0.5011634826660156,
      "learning_rate": 0.0004996000877966172,
      "loss": 1.3438,
      "step": 2315
    },
    {
      "epoch": 0.27887967303762473,
      "grad_norm": 0.6307665705680847,
      "learning_rate": 0.0004995973873326326,
      "loss": 1.5703,
      "step": 2320
    },
    {
      "epoch": 0.27948070681572307,
      "grad_norm": 0.46662095189094543,
      "learning_rate": 0.0004995946777890169,
      "loss": 1.4414,
      "step": 2325
    },
    {
      "epoch": 0.28008174059382135,
      "grad_norm": 0.49989181756973267,
      "learning_rate": 0.0004995919591658687,
      "loss": 1.3789,
      "step": 2330
    },
    {
      "epoch": 0.2806827743719197,
      "grad_norm": 0.4880094528198242,
      "learning_rate": 0.0004995892314632867,
      "loss": 1.2633,
      "step": 2335
    },
    {
      "epoch": 0.281283808150018,
      "grad_norm": 0.6314132213592529,
      "learning_rate": 0.0004995864946813703,
      "loss": 1.5539,
      "step": 2340
    },
    {
      "epoch": 0.28188484192811636,
      "grad_norm": 0.7073726654052734,
      "learning_rate": 0.0004995837488202191,
      "loss": 1.3766,
      "step": 2345
    },
    {
      "epoch": 0.2824858757062147,
      "grad_norm": 0.5559587478637695,
      "learning_rate": 0.0004995809938799329,
      "loss": 1.4875,
      "step": 2350
    },
    {
      "epoch": 0.28308690948431303,
      "grad_norm": 0.4955267906188965,
      "learning_rate": 0.0004995782298606119,
      "loss": 1.3156,
      "step": 2355
    },
    {
      "epoch": 0.28368794326241137,
      "grad_norm": 0.4989592432975769,
      "learning_rate": 0.0004995754567623567,
      "loss": 1.2484,
      "step": 2360
    },
    {
      "epoch": 0.28428897704050965,
      "grad_norm": 0.5886387228965759,
      "learning_rate": 0.0004995726745852681,
      "loss": 1.2344,
      "step": 2365
    },
    {
      "epoch": 0.284890010818608,
      "grad_norm": 0.5085893273353577,
      "learning_rate": 0.0004995698833294474,
      "loss": 1.407,
      "step": 2370
    },
    {
      "epoch": 0.2854910445967063,
      "grad_norm": 0.4706375002861023,
      "learning_rate": 0.000499567082994996,
      "loss": 1.3117,
      "step": 2375
    },
    {
      "epoch": 0.28609207837480466,
      "grad_norm": 0.5287367701530457,
      "learning_rate": 0.000499564273582016,
      "loss": 1.2594,
      "step": 2380
    },
    {
      "epoch": 0.286693112152903,
      "grad_norm": 0.5483081936836243,
      "learning_rate": 0.0004995614550906093,
      "loss": 1.1008,
      "step": 2385
    },
    {
      "epoch": 0.28729414593100133,
      "grad_norm": 0.8154200911521912,
      "learning_rate": 0.0004995586275208788,
      "loss": 1.5164,
      "step": 2390
    },
    {
      "epoch": 0.28789517970909967,
      "grad_norm": 0.837818443775177,
      "learning_rate": 0.000499555790872927,
      "loss": 1.2531,
      "step": 2395
    },
    {
      "epoch": 0.288496213487198,
      "grad_norm": 0.4989728033542633,
      "learning_rate": 0.0004995529451468574,
      "loss": 1.3719,
      "step": 2400
    },
    {
      "epoch": 0.288496213487198,
      "eval_loss": 2.189453125,
      "eval_model_preparation_time": 0.0053,
      "eval_runtime": 35.202,
      "eval_samples_per_second": 4.545,
      "eval_steps_per_second": 1.136,
      "step": 2400
    },
    {
      "epoch": 0.2890972472652963,
      "grad_norm": 0.5615506768226624,
      "learning_rate": 0.0004995500903427732,
      "loss": 1.1023,
      "step": 2405
    },
    {
      "epoch": 0.2896982810433946,
      "grad_norm": 0.7758134007453918,
      "learning_rate": 0.0004995472264607784,
      "loss": 1.2625,
      "step": 2410
    },
    {
      "epoch": 0.29029931482149296,
      "grad_norm": 0.6751444935798645,
      "learning_rate": 0.0004995443535009773,
      "loss": 1.5734,
      "step": 2415
    },
    {
      "epoch": 0.2909003485995913,
      "grad_norm": 0.5839786529541016,
      "learning_rate": 0.0004995414714634743,
      "loss": 1.3625,
      "step": 2420
    },
    {
      "epoch": 0.29150138237768963,
      "grad_norm": 0.5906524062156677,
      "learning_rate": 0.0004995385803483742,
      "loss": 1.0875,
      "step": 2425
    },
    {
      "epoch": 0.29210241615578797,
      "grad_norm": 0.7597156763076782,
      "learning_rate": 0.0004995356801557821,
      "loss": 1.4781,
      "step": 2430
    },
    {
      "epoch": 0.2927034499338863,
      "grad_norm": 0.5112520456314087,
      "learning_rate": 0.0004995327708858038,
      "loss": 1.2758,
      "step": 2435
    },
    {
      "epoch": 0.2933044837119846,
      "grad_norm": 0.44212523102760315,
      "learning_rate": 0.0004995298525385447,
      "loss": 1.5094,
      "step": 2440
    },
    {
      "epoch": 0.2939055174900829,
      "grad_norm": 0.43641284108161926,
      "learning_rate": 0.0004995269251141114,
      "loss": 1.1656,
      "step": 2445
    },
    {
      "epoch": 0.29450655126818126,
      "grad_norm": 0.4382478892803192,
      "learning_rate": 0.0004995239886126102,
      "loss": 1.5023,
      "step": 2450
    },
    {
      "epoch": 0.2951075850462796,
      "grad_norm": 0.6196442246437073,
      "learning_rate": 0.0004995210430341478,
      "loss": 1.2875,
      "step": 2455
    },
    {
      "epoch": 0.29570861882437793,
      "grad_norm": 0.6048389673233032,
      "learning_rate": 0.0004995180883788316,
      "loss": 0.9516,
      "step": 2460
    },
    {
      "epoch": 0.29630965260247627,
      "grad_norm": 0.5682608485221863,
      "learning_rate": 0.0004995151246467689,
      "loss": 1.3422,
      "step": 2465
    },
    {
      "epoch": 0.2969106863805746,
      "grad_norm": 0.5677405595779419,
      "learning_rate": 0.0004995121518380674,
      "loss": 1.5016,
      "step": 2470
    },
    {
      "epoch": 0.29751172015867294,
      "grad_norm": 0.48005715012550354,
      "learning_rate": 0.0004995091699528355,
      "loss": 1.3219,
      "step": 2475
    },
    {
      "epoch": 0.2981127539367712,
      "grad_norm": 0.48294246196746826,
      "learning_rate": 0.0004995061789911817,
      "loss": 1.2516,
      "step": 2480
    },
    {
      "epoch": 0.29871378771486956,
      "grad_norm": 0.7167287468910217,
      "learning_rate": 0.0004995031789532147,
      "loss": 1.3531,
      "step": 2485
    },
    {
      "epoch": 0.2993148214929679,
      "grad_norm": 0.5675193667411804,
      "learning_rate": 0.0004995001698390434,
      "loss": 1.3648,
      "step": 2490
    },
    {
      "epoch": 0.29991585527106623,
      "grad_norm": 0.5264390707015991,
      "learning_rate": 0.0004994971516487775,
      "loss": 1.1133,
      "step": 2495
    },
    {
      "epoch": 0.30051688904916457,
      "grad_norm": 0.5506901144981384,
      "learning_rate": 0.0004994941243825269,
      "loss": 1.1594,
      "step": 2500
    },
    {
      "epoch": 0.3011179228272629,
      "grad_norm": 0.9272066950798035,
      "learning_rate": 0.0004994910880404015,
      "loss": 1.4906,
      "step": 2505
    },
    {
      "epoch": 0.30171895660536124,
      "grad_norm": 0.5853176712989807,
      "learning_rate": 0.0004994880426225119,
      "loss": 1.3508,
      "step": 2510
    },
    {
      "epoch": 0.3023199903834595,
      "grad_norm": 0.4796172082424164,
      "learning_rate": 0.0004994849881289687,
      "loss": 1.3484,
      "step": 2515
    },
    {
      "epoch": 0.30292102416155786,
      "grad_norm": 0.6331420540809631,
      "learning_rate": 0.0004994819245598833,
      "loss": 1.2188,
      "step": 2520
    },
    {
      "epoch": 0.3035220579396562,
      "grad_norm": 0.6519079208374023,
      "learning_rate": 0.000499478851915367,
      "loss": 1.3531,
      "step": 2525
    },
    {
      "epoch": 0.30412309171775453,
      "grad_norm": 0.6366649866104126,
      "learning_rate": 0.0004994757701955314,
      "loss": 1.1703,
      "step": 2530
    },
    {
      "epoch": 0.30472412549585287,
      "grad_norm": 0.5621868371963501,
      "learning_rate": 0.0004994726794004888,
      "loss": 1.0441,
      "step": 2535
    },
    {
      "epoch": 0.3053251592739512,
      "grad_norm": 0.6726334095001221,
      "learning_rate": 0.0004994695795303517,
      "loss": 1.2984,
      "step": 2540
    },
    {
      "epoch": 0.30592619305204954,
      "grad_norm": 0.5448851585388184,
      "learning_rate": 0.0004994664705852326,
      "loss": 0.8781,
      "step": 2545
    },
    {
      "epoch": 0.3065272268301479,
      "grad_norm": 0.6853761076927185,
      "learning_rate": 0.0004994633525652448,
      "loss": 1.6891,
      "step": 2550
    },
    {
      "epoch": 0.30712826060824616,
      "grad_norm": 0.5627267956733704,
      "learning_rate": 0.0004994602254705017,
      "loss": 1.368,
      "step": 2555
    },
    {
      "epoch": 0.3077292943863445,
      "grad_norm": 0.38999640941619873,
      "learning_rate": 0.0004994570893011171,
      "loss": 1.3789,
      "step": 2560
    },
    {
      "epoch": 0.30833032816444284,
      "grad_norm": 0.6671114563941956,
      "learning_rate": 0.000499453944057205,
      "loss": 1.4078,
      "step": 2565
    },
    {
      "epoch": 0.30893136194254117,
      "grad_norm": 0.5521063208580017,
      "learning_rate": 0.0004994507897388798,
      "loss": 1.5859,
      "step": 2570
    },
    {
      "epoch": 0.3095323957206395,
      "grad_norm": 0.6885313391685486,
      "learning_rate": 0.0004994476263462563,
      "loss": 1.2578,
      "step": 2575
    },
    {
      "epoch": 0.31013342949873784,
      "grad_norm": 0.45498156547546387,
      "learning_rate": 0.0004994444538794495,
      "loss": 1.3914,
      "step": 2580
    },
    {
      "epoch": 0.3107344632768362,
      "grad_norm": 0.5482655167579651,
      "learning_rate": 0.0004994412723385749,
      "loss": 1.3391,
      "step": 2585
    },
    {
      "epoch": 0.31133549705493446,
      "grad_norm": 0.5240392684936523,
      "learning_rate": 0.0004994380817237482,
      "loss": 1.25,
      "step": 2590
    },
    {
      "epoch": 0.3119365308330328,
      "grad_norm": 0.5129856467247009,
      "learning_rate": 0.0004994348820350854,
      "loss": 1.4406,
      "step": 2595
    },
    {
      "epoch": 0.31253756461113114,
      "grad_norm": 0.5252668261528015,
      "learning_rate": 0.000499431673272703,
      "loss": 1.0773,
      "step": 2600
    },
    {
      "epoch": 0.31253756461113114,
      "eval_loss": 2.1500000953674316,
      "eval_model_preparation_time": 0.0053,
      "eval_runtime": 35.1975,
      "eval_samples_per_second": 4.546,
      "eval_steps_per_second": 1.136,
      "step": 2600
    },
    {
      "epoch": 0.31313859838922947,
      "grad_norm": 0.6648097634315491,
      "learning_rate": 0.0004994284554367176,
      "loss": 1.1133,
      "step": 2605
    },
    {
      "epoch": 0.3137396321673278,
      "grad_norm": 0.6218547224998474,
      "learning_rate": 0.0004994252285272465,
      "loss": 1.2937,
      "step": 2610
    },
    {
      "epoch": 0.31434066594542615,
      "grad_norm": 0.6880519390106201,
      "learning_rate": 0.0004994219925444068,
      "loss": 1.8039,
      "step": 2615
    },
    {
      "epoch": 0.3149416997235245,
      "grad_norm": 0.6464706063270569,
      "learning_rate": 0.0004994187474883164,
      "loss": 1.5594,
      "step": 2620
    },
    {
      "epoch": 0.3155427335016228,
      "grad_norm": 0.7200093865394592,
      "learning_rate": 0.0004994154933590932,
      "loss": 1.0945,
      "step": 2625
    },
    {
      "epoch": 0.3161437672797211,
      "grad_norm": 0.6853864789009094,
      "learning_rate": 0.0004994122301568557,
      "loss": 1.268,
      "step": 2630
    },
    {
      "epoch": 0.31674480105781944,
      "grad_norm": 0.5081961750984192,
      "learning_rate": 0.0004994089578817226,
      "loss": 1.4062,
      "step": 2635
    },
    {
      "epoch": 0.3173458348359178,
      "grad_norm": 0.4750553071498871,
      "learning_rate": 0.0004994056765338129,
      "loss": 1.2828,
      "step": 2640
    },
    {
      "epoch": 0.3179468686140161,
      "grad_norm": 0.5867997407913208,
      "learning_rate": 0.0004994023861132459,
      "loss": 1.2484,
      "step": 2645
    },
    {
      "epoch": 0.31854790239211445,
      "grad_norm": 0.7348740696907043,
      "learning_rate": 0.0004993990866201414,
      "loss": 1.2258,
      "step": 2650
    },
    {
      "epoch": 0.3191489361702128,
      "grad_norm": 0.5523998141288757,
      "learning_rate": 0.0004993957780546193,
      "loss": 1.2805,
      "step": 2655
    },
    {
      "epoch": 0.3197499699483111,
      "grad_norm": 0.5308116674423218,
      "learning_rate": 0.0004993924604168001,
      "loss": 1.3188,
      "step": 2660
    },
    {
      "epoch": 0.3203510037264094,
      "grad_norm": 0.40592867136001587,
      "learning_rate": 0.0004993891337068046,
      "loss": 1.2148,
      "step": 2665
    },
    {
      "epoch": 0.32095203750450774,
      "grad_norm": 0.6522583365440369,
      "learning_rate": 0.0004993857979247535,
      "loss": 1.175,
      "step": 2670
    },
    {
      "epoch": 0.3215530712826061,
      "grad_norm": 0.5981694459915161,
      "learning_rate": 0.0004993824530707682,
      "loss": 1.143,
      "step": 2675
    },
    {
      "epoch": 0.3221541050607044,
      "grad_norm": 0.6832042932510376,
      "learning_rate": 0.0004993790991449707,
      "loss": 1.2242,
      "step": 2680
    },
    {
      "epoch": 0.32275513883880275,
      "grad_norm": 0.6935708522796631,
      "learning_rate": 0.0004993757361474825,
      "loss": 0.9617,
      "step": 2685
    },
    {
      "epoch": 0.3233561726169011,
      "grad_norm": 0.5491186380386353,
      "learning_rate": 0.0004993723640784265,
      "loss": 1.3672,
      "step": 2690
    },
    {
      "epoch": 0.3239572063949994,
      "grad_norm": 0.4743538498878479,
      "learning_rate": 0.0004993689829379249,
      "loss": 1.1547,
      "step": 2695
    },
    {
      "epoch": 0.3245582401730977,
      "grad_norm": 0.641859769821167,
      "learning_rate": 0.0004993655927261008,
      "loss": 1.4078,
      "step": 2700
    },
    {
      "epoch": 0.32515927395119604,
      "grad_norm": 0.5002933144569397,
      "learning_rate": 0.0004993621934430778,
      "loss": 0.9492,
      "step": 2705
    },
    {
      "epoch": 0.3257603077292944,
      "grad_norm": 0.7241799831390381,
      "learning_rate": 0.0004993587850889793,
      "loss": 1.575,
      "step": 2710
    },
    {
      "epoch": 0.3263613415073927,
      "grad_norm": 0.5693483948707581,
      "learning_rate": 0.0004993553676639292,
      "loss": 0.9961,
      "step": 2715
    },
    {
      "epoch": 0.32696237528549105,
      "grad_norm": 0.43130815029144287,
      "learning_rate": 0.000499351941168052,
      "loss": 1.2984,
      "step": 2720
    },
    {
      "epoch": 0.3275634090635894,
      "grad_norm": 0.5054978728294373,
      "learning_rate": 0.0004993485056014724,
      "loss": 1.1375,
      "step": 2725
    },
    {
      "epoch": 0.3281644428416877,
      "grad_norm": 0.5581235289573669,
      "learning_rate": 0.0004993450609643152,
      "loss": 1.0164,
      "step": 2730
    },
    {
      "epoch": 0.32876547661978606,
      "grad_norm": 0.6733124256134033,
      "learning_rate": 0.0004993416072567059,
      "loss": 1.4078,
      "step": 2735
    },
    {
      "epoch": 0.32936651039788434,
      "grad_norm": 0.5003538727760315,
      "learning_rate": 0.0004993381444787699,
      "loss": 0.8742,
      "step": 2740
    },
    {
      "epoch": 0.3299675441759827,
      "grad_norm": 0.6292559504508972,
      "learning_rate": 0.0004993346726306333,
      "loss": 1.007,
      "step": 2745
    },
    {
      "epoch": 0.330568577954081,
      "grad_norm": 0.6760239005088806,
      "learning_rate": 0.0004993311917124224,
      "loss": 1.25,
      "step": 2750
    },
    {
      "epoch": 0.33116961173217935,
      "grad_norm": 0.6075654625892639,
      "learning_rate": 0.0004993277017242638,
      "loss": 1.5766,
      "step": 2755
    },
    {
      "epoch": 0.3317706455102777,
      "grad_norm": 0.5432557463645935,
      "learning_rate": 0.0004993242026662846,
      "loss": 1.0883,
      "step": 2760
    },
    {
      "epoch": 0.332371679288376,
      "grad_norm": 0.6972253918647766,
      "learning_rate": 0.0004993206945386118,
      "loss": 0.9992,
      "step": 2765
    },
    {
      "epoch": 0.33297271306647436,
      "grad_norm": 0.45837146043777466,
      "learning_rate": 0.0004993171773413731,
      "loss": 1.6766,
      "step": 2770
    },
    {
      "epoch": 0.33357374684457264,
      "grad_norm": 0.5207621455192566,
      "learning_rate": 0.0004993136510746966,
      "loss": 1.2578,
      "step": 2775
    },
    {
      "epoch": 0.334174780622671,
      "grad_norm": 0.7034028768539429,
      "learning_rate": 0.0004993101157387106,
      "loss": 1.3578,
      "step": 2780
    },
    {
      "epoch": 0.3347758144007693,
      "grad_norm": 0.544851541519165,
      "learning_rate": 0.0004993065713335434,
      "loss": 1.2836,
      "step": 2785
    },
    {
      "epoch": 0.33537684817886765,
      "grad_norm": 0.705143928527832,
      "learning_rate": 0.0004993030178593241,
      "loss": 1.4453,
      "step": 2790
    },
    {
      "epoch": 0.335977881956966,
      "grad_norm": 0.6619438529014587,
      "learning_rate": 0.0004992994553161823,
      "loss": 1.0547,
      "step": 2795
    },
    {
      "epoch": 0.3365789157350643,
      "grad_norm": 0.5982903242111206,
      "learning_rate": 0.000499295883704247,
      "loss": 1.4281,
      "step": 2800
    },
    {
      "epoch": 0.3365789157350643,
      "eval_loss": 2.189453125,
      "eval_model_preparation_time": 0.0053,
      "eval_runtime": 35.1965,
      "eval_samples_per_second": 4.546,
      "eval_steps_per_second": 1.136,
      "step": 2800
    },
    {
      "epoch": 0.33717994951316266,
      "grad_norm": 0.589056670665741,
      "learning_rate": 0.0004992923030236485,
      "loss": 1.3727,
      "step": 2805
    },
    {
      "epoch": 0.337780983291261,
      "grad_norm": 0.39378607273101807,
      "learning_rate": 0.000499288713274517,
      "loss": 1.1789,
      "step": 2810
    },
    {
      "epoch": 0.3383820170693593,
      "grad_norm": 0.5460519790649414,
      "learning_rate": 0.000499285114456983,
      "loss": 1.1719,
      "step": 2815
    },
    {
      "epoch": 0.3389830508474576,
      "grad_norm": 0.4953864812850952,
      "learning_rate": 0.0004992815065711774,
      "loss": 1.1672,
      "step": 2820
    },
    {
      "epoch": 0.33958408462555595,
      "grad_norm": 0.5705846548080444,
      "learning_rate": 0.0004992778896172317,
      "loss": 1.5328,
      "step": 2825
    },
    {
      "epoch": 0.3401851184036543,
      "grad_norm": 0.5687447190284729,
      "learning_rate": 0.0004992742635952771,
      "loss": 1.0063,
      "step": 2830
    },
    {
      "epoch": 0.3407861521817526,
      "grad_norm": 0.516343891620636,
      "learning_rate": 0.0004992706285054458,
      "loss": 1.1492,
      "step": 2835
    },
    {
      "epoch": 0.34138718595985096,
      "grad_norm": 0.6128392815589905,
      "learning_rate": 0.0004992669843478699,
      "loss": 1.325,
      "step": 2840
    },
    {
      "epoch": 0.3419882197379493,
      "grad_norm": 0.5104270577430725,
      "learning_rate": 0.000499263331122682,
      "loss": 1.0195,
      "step": 2845
    },
    {
      "epoch": 0.3425892535160476,
      "grad_norm": 0.38332250714302063,
      "learning_rate": 0.0004992596688300149,
      "loss": 1.302,
      "step": 2850
    },
    {
      "epoch": 0.3431902872941459,
      "grad_norm": 0.5674039125442505,
      "learning_rate": 0.000499255997470002,
      "loss": 1.2094,
      "step": 2855
    },
    {
      "epoch": 0.34379132107224425,
      "grad_norm": 0.7987366914749146,
      "learning_rate": 0.0004992523170427766,
      "loss": 1.2047,
      "step": 2860
    },
    {
      "epoch": 0.3443923548503426,
      "grad_norm": 0.45501282811164856,
      "learning_rate": 0.0004992486275484729,
      "loss": 1.1539,
      "step": 2865
    },
    {
      "epoch": 0.3449933886284409,
      "grad_norm": 0.5390669703483582,
      "learning_rate": 0.0004992449289872249,
      "loss": 1.2102,
      "step": 2870
    },
    {
      "epoch": 0.34559442240653926,
      "grad_norm": 0.6710581183433533,
      "learning_rate": 0.0004992412213591672,
      "loss": 1.4297,
      "step": 2875
    },
    {
      "epoch": 0.3461954561846376,
      "grad_norm": 0.6371570825576782,
      "learning_rate": 0.0004992375046644347,
      "loss": 1.0164,
      "step": 2880
    },
    {
      "epoch": 0.34679648996273593,
      "grad_norm": 0.49934741854667664,
      "learning_rate": 0.0004992337789031625,
      "loss": 1.1313,
      "step": 2885
    },
    {
      "epoch": 0.3473975237408342,
      "grad_norm": 0.41756120324134827,
      "learning_rate": 0.0004992300440754862,
      "loss": 1.1969,
      "step": 2890
    },
    {
      "epoch": 0.34799855751893255,
      "grad_norm": 0.8102174997329712,
      "learning_rate": 0.0004992263001815418,
      "loss": 1.2719,
      "step": 2895
    },
    {
      "epoch": 0.3485995912970309,
      "grad_norm": 0.45573851466178894,
      "learning_rate": 0.0004992225472214653,
      "loss": 1.1375,
      "step": 2900
    },
    {
      "epoch": 0.3492006250751292,
      "grad_norm": 0.5512142777442932,
      "learning_rate": 0.0004992187851953932,
      "loss": 1.4781,
      "step": 2905
    },
    {
      "epoch": 0.34980165885322756,
      "grad_norm": 0.6429489850997925,
      "learning_rate": 0.0004992150141034624,
      "loss": 1.3453,
      "step": 2910
    },
    {
      "epoch": 0.3504026926313259,
      "grad_norm": 0.6230481266975403,
      "learning_rate": 0.0004992112339458103,
      "loss": 1.2766,
      "step": 2915
    },
    {
      "epoch": 0.35100372640942423,
      "grad_norm": 0.6134311556816101,
      "learning_rate": 0.0004992074447225741,
      "loss": 1.0664,
      "step": 2920
    },
    {
      "epoch": 0.3516047601875225,
      "grad_norm": 0.5894711017608643,
      "learning_rate": 0.0004992036464338918,
      "loss": 0.9,
      "step": 2925
    },
    {
      "epoch": 0.35220579396562085,
      "grad_norm": 0.525947630405426,
      "learning_rate": 0.0004991998390799016,
      "loss": 1.4844,
      "step": 2930
    },
    {
      "epoch": 0.3528068277437192,
      "grad_norm": 0.5282221436500549,
      "learning_rate": 0.0004991960226607418,
      "loss": 1.0398,
      "step": 2935
    },
    {
      "epoch": 0.3534078615218175,
      "grad_norm": 0.5808281302452087,
      "learning_rate": 0.0004991921971765514,
      "loss": 0.943,
      "step": 2940
    },
    {
      "epoch": 0.35400889529991586,
      "grad_norm": 0.6163946390151978,
      "learning_rate": 0.0004991883626274696,
      "loss": 1.1086,
      "step": 2945
    },
    {
      "epoch": 0.3546099290780142,
      "grad_norm": 0.4154224991798401,
      "learning_rate": 0.0004991845190136357,
      "loss": 1.2703,
      "step": 2950
    },
    {
      "epoch": 0.35521096285611253,
      "grad_norm": 0.4030189514160156,
      "learning_rate": 0.0004991806663351897,
      "loss": 1.1086,
      "step": 2955
    },
    {
      "epoch": 0.35581199663421087,
      "grad_norm": 0.5927292704582214,
      "learning_rate": 0.0004991768045922718,
      "loss": 0.9758,
      "step": 2960
    },
    {
      "epoch": 0.35641303041230915,
      "grad_norm": 0.476971834897995,
      "learning_rate": 0.0004991729337850223,
      "loss": 1.525,
      "step": 2965
    },
    {
      "epoch": 0.3570140641904075,
      "grad_norm": 0.5584660768508911,
      "learning_rate": 0.000499169053913582,
      "loss": 1.2125,
      "step": 2970
    },
    {
      "epoch": 0.3576150979685058,
      "grad_norm": 0.5617804527282715,
      "learning_rate": 0.0004991651649780922,
      "loss": 1.3102,
      "step": 2975
    },
    {
      "epoch": 0.35821613174660416,
      "grad_norm": 0.3463181257247925,
      "learning_rate": 0.0004991612669786942,
      "loss": 1.4227,
      "step": 2980
    },
    {
      "epoch": 0.3588171655247025,
      "grad_norm": 0.5156741142272949,
      "learning_rate": 0.0004991573599155299,
      "loss": 1.3828,
      "step": 2985
    },
    {
      "epoch": 0.35941819930280083,
      "grad_norm": 0.6080055832862854,
      "learning_rate": 0.0004991534437887414,
      "loss": 1.0102,
      "step": 2990
    },
    {
      "epoch": 0.36001923308089917,
      "grad_norm": 0.5142369866371155,
      "learning_rate": 0.0004991495185984711,
      "loss": 1.3469,
      "step": 2995
    },
    {
      "epoch": 0.36062026685899745,
      "grad_norm": 0.4148232638835907,
      "learning_rate": 0.000499145584344862,
      "loss": 1.3102,
      "step": 3000
    },
    {
      "epoch": 0.36062026685899745,
      "eval_loss": 2.1435546875,
      "eval_model_preparation_time": 0.0053,
      "eval_runtime": 35.2124,
      "eval_samples_per_second": 4.544,
      "eval_steps_per_second": 1.136,
      "step": 3000
    },
    {
      "epoch": 0.3612213006370958,
      "grad_norm": 0.43302541971206665,
      "learning_rate": 0.000499141641028057,
      "loss": 1.1625,
      "step": 3005
    },
    {
      "epoch": 0.3618223344151941,
      "grad_norm": 0.5702094435691833,
      "learning_rate": 0.0004991376886481996,
      "loss": 1.4937,
      "step": 3010
    },
    {
      "epoch": 0.36242336819329246,
      "grad_norm": 0.602148175239563,
      "learning_rate": 0.0004991337272054336,
      "loss": 1.4453,
      "step": 3015
    },
    {
      "epoch": 0.3630244019713908,
      "grad_norm": 0.7094736099243164,
      "learning_rate": 0.0004991297566999031,
      "loss": 1.1242,
      "step": 3020
    },
    {
      "epoch": 0.36362543574948913,
      "grad_norm": 1.0491294860839844,
      "learning_rate": 0.0004991257771317525,
      "loss": 1.2945,
      "step": 3025
    },
    {
      "epoch": 0.36422646952758747,
      "grad_norm": 0.5383880734443665,
      "learning_rate": 0.0004991217885011266,
      "loss": 1.4398,
      "step": 3030
    },
    {
      "epoch": 0.36482750330568575,
      "grad_norm": 0.5303685665130615,
      "learning_rate": 0.0004991177908081706,
      "loss": 1.357,
      "step": 3035
    },
    {
      "epoch": 0.3654285370837841,
      "grad_norm": 0.5658953785896301,
      "learning_rate": 0.0004991137840530297,
      "loss": 1.625,
      "step": 3040
    },
    {
      "epoch": 0.3660295708618824,
      "grad_norm": 0.6413328051567078,
      "learning_rate": 0.0004991097682358498,
      "loss": 1.0664,
      "step": 3045
    },
    {
      "epoch": 0.36663060463998076,
      "grad_norm": 0.40573734045028687,
      "learning_rate": 0.000499105743356777,
      "loss": 1.4219,
      "step": 3050
    },
    {
      "epoch": 0.3672316384180791,
      "grad_norm": 0.4775443375110626,
      "learning_rate": 0.0004991017094159576,
      "loss": 1.1383,
      "step": 3055
    },
    {
      "epoch": 0.36783267219617743,
      "grad_norm": 0.4478199779987335,
      "learning_rate": 0.0004990976664135384,
      "loss": 1.1906,
      "step": 3060
    },
    {
      "epoch": 0.36843370597427577,
      "grad_norm": 0.5353224277496338,
      "learning_rate": 0.0004990936143496664,
      "loss": 1.0695,
      "step": 3065
    },
    {
      "epoch": 0.3690347397523741,
      "grad_norm": 0.6140496730804443,
      "learning_rate": 0.0004990895532244893,
      "loss": 1.45,
      "step": 3070
    },
    {
      "epoch": 0.3696357735304724,
      "grad_norm": 0.674697995185852,
      "learning_rate": 0.0004990854830381545,
      "loss": 1.2617,
      "step": 3075
    },
    {
      "epoch": 0.3702368073085707,
      "grad_norm": 0.793539822101593,
      "learning_rate": 0.0004990814037908102,
      "loss": 1.0797,
      "step": 3080
    },
    {
      "epoch": 0.37083784108666906,
      "grad_norm": 0.4356972873210907,
      "learning_rate": 0.0004990773154826048,
      "loss": 1.257,
      "step": 3085
    },
    {
      "epoch": 0.3714388748647674,
      "grad_norm": 0.4007517099380493,
      "learning_rate": 0.000499073218113687,
      "loss": 1.6695,
      "step": 3090
    },
    {
      "epoch": 0.37203990864286574,
      "grad_norm": 0.7213647961616516,
      "learning_rate": 0.0004990691116842058,
      "loss": 1.1039,
      "step": 3095
    },
    {
      "epoch": 0.37264094242096407,
      "grad_norm": 0.5188817977905273,
      "learning_rate": 0.0004990649961943105,
      "loss": 1.3844,
      "step": 3100
    },
    {
      "epoch": 0.3732419761990624,
      "grad_norm": 0.5494900345802307,
      "learning_rate": 0.0004990608716441511,
      "loss": 1.3094,
      "step": 3105
    },
    {
      "epoch": 0.3738430099771607,
      "grad_norm": 0.6364961266517639,
      "learning_rate": 0.0004990567380338774,
      "loss": 1.432,
      "step": 3110
    },
    {
      "epoch": 0.374444043755259,
      "grad_norm": 0.7705929279327393,
      "learning_rate": 0.0004990525953636399,
      "loss": 1.3594,
      "step": 3115
    },
    {
      "epoch": 0.37504507753335736,
      "grad_norm": 0.6256239414215088,
      "learning_rate": 0.0004990484436335892,
      "loss": 1.4969,
      "step": 3120
    },
    {
      "epoch": 0.3756461113114557,
      "grad_norm": 0.8040322661399841,
      "learning_rate": 0.0004990442828438764,
      "loss": 1.3664,
      "step": 3125
    },
    {
      "epoch": 0.37624714508955404,
      "grad_norm": 0.7349644303321838,
      "learning_rate": 0.0004990401129946528,
      "loss": 1.3016,
      "step": 3130
    },
    {
      "epoch": 0.3768481788676524,
      "grad_norm": 0.7406263947486877,
      "learning_rate": 0.0004990359340860701,
      "loss": 1.2672,
      "step": 3135
    },
    {
      "epoch": 0.3774492126457507,
      "grad_norm": 0.6488270163536072,
      "learning_rate": 0.0004990317461182803,
      "loss": 1.5125,
      "step": 3140
    },
    {
      "epoch": 0.37805024642384905,
      "grad_norm": 0.450980544090271,
      "learning_rate": 0.0004990275490914358,
      "loss": 0.9531,
      "step": 3145
    },
    {
      "epoch": 0.3786512802019473,
      "grad_norm": 0.5992457866668701,
      "learning_rate": 0.0004990233430056892,
      "loss": 1.2563,
      "step": 3150
    },
    {
      "epoch": 0.37925231398004566,
      "grad_norm": 0.4323638677597046,
      "learning_rate": 0.0004990191278611936,
      "loss": 1.1438,
      "step": 3155
    },
    {
      "epoch": 0.379853347758144,
      "grad_norm": 0.9859302639961243,
      "learning_rate": 0.0004990149036581023,
      "loss": 1.2555,
      "step": 3160
    },
    {
      "epoch": 0.38045438153624234,
      "grad_norm": 0.8136280179023743,
      "learning_rate": 0.0004990106703965689,
      "loss": 1.3172,
      "step": 3165
    },
    {
      "epoch": 0.3810554153143407,
      "grad_norm": 0.5532881021499634,
      "learning_rate": 0.0004990064280767475,
      "loss": 0.9656,
      "step": 3170
    },
    {
      "epoch": 0.381656449092439,
      "grad_norm": 0.5996264219284058,
      "learning_rate": 0.0004990021766987923,
      "loss": 1.2688,
      "step": 3175
    },
    {
      "epoch": 0.38225748287053735,
      "grad_norm": 0.6474243402481079,
      "learning_rate": 0.0004989979162628582,
      "loss": 0.9461,
      "step": 3180
    },
    {
      "epoch": 0.3828585166486356,
      "grad_norm": 0.5693522691726685,
      "learning_rate": 0.0004989936467690998,
      "loss": 1.0906,
      "step": 3185
    },
    {
      "epoch": 0.38345955042673396,
      "grad_norm": 0.5393794775009155,
      "learning_rate": 0.0004989893682176727,
      "loss": 1.2445,
      "step": 3190
    },
    {
      "epoch": 0.3840605842048323,
      "grad_norm": 0.6597670316696167,
      "learning_rate": 0.0004989850806087325,
      "loss": 1.4289,
      "step": 3195
    },
    {
      "epoch": 0.38466161798293064,
      "grad_norm": 0.6367236971855164,
      "learning_rate": 0.0004989807839424352,
      "loss": 1.2266,
      "step": 3200
    },
    {
      "epoch": 0.38466161798293064,
      "eval_loss": 2.1474609375,
      "eval_model_preparation_time": 0.0053,
      "eval_runtime": 35.1942,
      "eval_samples_per_second": 4.546,
      "eval_steps_per_second": 1.137,
      "step": 3200
    },
    {
      "epoch": 0.385262651761029,
      "grad_norm": 0.5790736079216003,
      "learning_rate": 0.0004989764782189369,
      "loss": 1.2164,
      "step": 3205
    },
    {
      "epoch": 0.3858636855391273,
      "grad_norm": 0.502346932888031,
      "learning_rate": 0.0004989721634383943,
      "loss": 1.3391,
      "step": 3210
    },
    {
      "epoch": 0.38646471931722565,
      "grad_norm": 0.6391408443450928,
      "learning_rate": 0.0004989678396009645,
      "loss": 1.2711,
      "step": 3215
    },
    {
      "epoch": 0.387065753095324,
      "grad_norm": 0.40186697244644165,
      "learning_rate": 0.0004989635067068047,
      "loss": 1.1691,
      "step": 3220
    },
    {
      "epoch": 0.38766678687342226,
      "grad_norm": 0.5028790235519409,
      "learning_rate": 0.0004989591647560726,
      "loss": 1.1609,
      "step": 3225
    },
    {
      "epoch": 0.3882678206515206,
      "grad_norm": 0.4072652757167816,
      "learning_rate": 0.0004989548137489259,
      "loss": 1.5672,
      "step": 3230
    },
    {
      "epoch": 0.38886885442961894,
      "grad_norm": 0.4738612174987793,
      "learning_rate": 0.0004989504536855232,
      "loss": 1.3344,
      "step": 3235
    },
    {
      "epoch": 0.3894698882077173,
      "grad_norm": 0.9375470876693726,
      "learning_rate": 0.0004989460845660229,
      "loss": 1.0484,
      "step": 3240
    },
    {
      "epoch": 0.3900709219858156,
      "grad_norm": 0.4789920449256897,
      "learning_rate": 0.000498941706390584,
      "loss": 1.1109,
      "step": 3245
    },
    {
      "epoch": 0.39067195576391395,
      "grad_norm": 0.6214213967323303,
      "learning_rate": 0.0004989373191593658,
      "loss": 1.3516,
      "step": 3250
    },
    {
      "epoch": 0.3912729895420123,
      "grad_norm": 0.49041748046875,
      "learning_rate": 0.0004989329228725277,
      "loss": 1.2977,
      "step": 3255
    },
    {
      "epoch": 0.39187402332011056,
      "grad_norm": 0.5892441272735596,
      "learning_rate": 0.00049892851753023,
      "loss": 1.1109,
      "step": 3260
    },
    {
      "epoch": 0.3924750570982089,
      "grad_norm": 0.4488331973552704,
      "learning_rate": 0.0004989241031326326,
      "loss": 1.2734,
      "step": 3265
    },
    {
      "epoch": 0.39307609087630724,
      "grad_norm": 0.43196800351142883,
      "learning_rate": 0.0004989196796798963,
      "loss": 1.1059,
      "step": 3270
    },
    {
      "epoch": 0.3936771246544056,
      "grad_norm": 0.3739017844200134,
      "learning_rate": 0.0004989152471721819,
      "loss": 0.8492,
      "step": 3275
    },
    {
      "epoch": 0.3942781584325039,
      "grad_norm": 0.5097094178199768,
      "learning_rate": 0.0004989108056096505,
      "loss": 1.1836,
      "step": 3280
    },
    {
      "epoch": 0.39487919221060225,
      "grad_norm": 0.7751893401145935,
      "learning_rate": 0.000498906354992464,
      "loss": 1.443,
      "step": 3285
    },
    {
      "epoch": 0.3954802259887006,
      "grad_norm": 0.6941812634468079,
      "learning_rate": 0.0004989018953207841,
      "loss": 1.2094,
      "step": 3290
    },
    {
      "epoch": 0.3960812597667989,
      "grad_norm": 0.5261918902397156,
      "learning_rate": 0.0004988974265947731,
      "loss": 1.1297,
      "step": 3295
    },
    {
      "epoch": 0.3966822935448972,
      "grad_norm": 0.6817163228988647,
      "learning_rate": 0.0004988929488145934,
      "loss": 1.5297,
      "step": 3300
    },
    {
      "epoch": 0.39728332732299554,
      "grad_norm": 0.6343579888343811,
      "learning_rate": 0.0004988884619804082,
      "loss": 1.1766,
      "step": 3305
    },
    {
      "epoch": 0.3978843611010939,
      "grad_norm": 0.4660755693912506,
      "learning_rate": 0.0004988839660923805,
      "loss": 0.9953,
      "step": 3310
    },
    {
      "epoch": 0.3984853948791922,
      "grad_norm": 0.6850960850715637,
      "learning_rate": 0.0004988794611506738,
      "loss": 1.4023,
      "step": 3315
    },
    {
      "epoch": 0.39908642865729055,
      "grad_norm": 0.7170994877815247,
      "learning_rate": 0.0004988749471554521,
      "loss": 1.1391,
      "step": 3320
    },
    {
      "epoch": 0.3996874624353889,
      "grad_norm": 0.9181567430496216,
      "learning_rate": 0.0004988704241068795,
      "loss": 1.3047,
      "step": 3325
    },
    {
      "epoch": 0.4002884962134872,
      "grad_norm": 0.5317991971969604,
      "learning_rate": 0.0004988658920051207,
      "loss": 1.3273,
      "step": 3330
    },
    {
      "epoch": 0.4008895299915855,
      "grad_norm": 0.5464211702346802,
      "learning_rate": 0.0004988613508503405,
      "loss": 1.0758,
      "step": 3335
    },
    {
      "epoch": 0.40149056376968384,
      "grad_norm": 0.664720892906189,
      "learning_rate": 0.0004988568006427039,
      "loss": 1.1383,
      "step": 3340
    },
    {
      "epoch": 0.4020915975477822,
      "grad_norm": 0.5137344002723694,
      "learning_rate": 0.0004988522413823767,
      "loss": 0.8992,
      "step": 3345
    },
    {
      "epoch": 0.4026926313258805,
      "grad_norm": 0.6021727919578552,
      "learning_rate": 0.0004988476730695246,
      "loss": 1.0391,
      "step": 3350
    },
    {
      "epoch": 0.40329366510397885,
      "grad_norm": 0.5493384003639221,
      "learning_rate": 0.0004988430957043138,
      "loss": 1.2063,
      "step": 3355
    },
    {
      "epoch": 0.4038946988820772,
      "grad_norm": 0.44344305992126465,
      "learning_rate": 0.0004988385092869109,
      "loss": 1.1539,
      "step": 3360
    },
    {
      "epoch": 0.4044957326601755,
      "grad_norm": 0.5496264100074768,
      "learning_rate": 0.0004988339138174827,
      "loss": 1.0008,
      "step": 3365
    },
    {
      "epoch": 0.40509676643827386,
      "grad_norm": 0.5868191719055176,
      "learning_rate": 0.0004988293092961962,
      "loss": 1.0273,
      "step": 3370
    },
    {
      "epoch": 0.40569780021637214,
      "grad_norm": 0.47252243757247925,
      "learning_rate": 0.0004988246957232191,
      "loss": 1.1547,
      "step": 3375
    },
    {
      "epoch": 0.4062988339944705,
      "grad_norm": 0.5384260416030884,
      "learning_rate": 0.0004988200730987192,
      "loss": 0.9969,
      "step": 3380
    },
    {
      "epoch": 0.4068998677725688,
      "grad_norm": 0.5157658457756042,
      "learning_rate": 0.0004988154414228645,
      "loss": 1.218,
      "step": 3385
    },
    {
      "epoch": 0.40750090155066715,
      "grad_norm": 0.609667181968689,
      "learning_rate": 0.0004988108006958237,
      "loss": 1.2977,
      "step": 3390
    },
    {
      "epoch": 0.4081019353287655,
      "grad_norm": 0.6255223751068115,
      "learning_rate": 0.0004988061509177656,
      "loss": 1.0859,
      "step": 3395
    },
    {
      "epoch": 0.4087029691068638,
      "grad_norm": 0.48681220412254333,
      "learning_rate": 0.0004988014920888592,
      "loss": 1.1094,
      "step": 3400
    },
    {
      "epoch": 0.4087029691068638,
      "eval_loss": 2.1009764671325684,
      "eval_model_preparation_time": 0.0053,
      "eval_runtime": 35.1797,
      "eval_samples_per_second": 4.548,
      "eval_steps_per_second": 1.137,
      "step": 3400
    },
    {
      "epoch": 0.40930400288496216,
      "grad_norm": 0.7313075065612793,
      "learning_rate": 0.0004987968242092741,
      "loss": 1.4648,
      "step": 3405
    },
    {
      "epoch": 0.40990503666306044,
      "grad_norm": 0.3973584771156311,
      "learning_rate": 0.00049879214727918,
      "loss": 1.4195,
      "step": 3410
    },
    {
      "epoch": 0.4105060704411588,
      "grad_norm": 0.5780074596405029,
      "learning_rate": 0.0004987874612987471,
      "loss": 1.1383,
      "step": 3415
    },
    {
      "epoch": 0.4111071042192571,
      "grad_norm": 0.5280987620353699,
      "learning_rate": 0.0004987827662681459,
      "loss": 1.2125,
      "step": 3420
    },
    {
      "epoch": 0.41170813799735545,
      "grad_norm": 0.672178328037262,
      "learning_rate": 0.0004987780621875471,
      "loss": 1.2563,
      "step": 3425
    },
    {
      "epoch": 0.4123091717754538,
      "grad_norm": 0.5133812427520752,
      "learning_rate": 0.0004987733490571218,
      "loss": 1.2266,
      "step": 3430
    },
    {
      "epoch": 0.4129102055535521,
      "grad_norm": 0.7771773338317871,
      "learning_rate": 0.0004987686268770415,
      "loss": 1.2648,
      "step": 3435
    },
    {
      "epoch": 0.41351123933165046,
      "grad_norm": 0.555830180644989,
      "learning_rate": 0.0004987638956474781,
      "loss": 1.2445,
      "step": 3440
    },
    {
      "epoch": 0.41411227310974874,
      "grad_norm": 0.377913236618042,
      "learning_rate": 0.0004987591553686035,
      "loss": 1.0813,
      "step": 3445
    },
    {
      "epoch": 0.4147133068878471,
      "grad_norm": 0.6742091774940491,
      "learning_rate": 0.0004987544060405903,
      "loss": 1.1516,
      "step": 3450
    },
    {
      "epoch": 0.4153143406659454,
      "grad_norm": 0.6832530498504639,
      "learning_rate": 0.0004987496476636112,
      "loss": 0.9953,
      "step": 3455
    },
    {
      "epoch": 0.41591537444404375,
      "grad_norm": 0.7047104835510254,
      "learning_rate": 0.0004987448802378393,
      "loss": 1.2781,
      "step": 3460
    },
    {
      "epoch": 0.4165164082221421,
      "grad_norm": 0.5788468718528748,
      "learning_rate": 0.000498740103763448,
      "loss": 1.1266,
      "step": 3465
    },
    {
      "epoch": 0.4171174420002404,
      "grad_norm": 0.6551057696342468,
      "learning_rate": 0.0004987353182406111,
      "loss": 0.9711,
      "step": 3470
    },
    {
      "epoch": 0.41771847577833876,
      "grad_norm": 0.7851700782775879,
      "learning_rate": 0.0004987305236695027,
      "loss": 1.2258,
      "step": 3475
    },
    {
      "epoch": 0.4183195095564371,
      "grad_norm": 0.45742517709732056,
      "learning_rate": 0.000498725720050297,
      "loss": 1.1375,
      "step": 3480
    },
    {
      "epoch": 0.4189205433345354,
      "grad_norm": 0.5413776636123657,
      "learning_rate": 0.0004987209073831691,
      "loss": 1.3938,
      "step": 3485
    },
    {
      "epoch": 0.4195215771126337,
      "grad_norm": 0.5107463598251343,
      "learning_rate": 0.0004987160856682938,
      "loss": 1.0172,
      "step": 3490
    },
    {
      "epoch": 0.42012261089073205,
      "grad_norm": 0.6791651844978333,
      "learning_rate": 0.0004987112549058466,
      "loss": 1.2555,
      "step": 3495
    },
    {
      "epoch": 0.4207236446688304,
      "grad_norm": 0.71052086353302,
      "learning_rate": 0.0004987064150960033,
      "loss": 1.057,
      "step": 3500
    },
    {
      "epoch": 0.4213246784469287,
      "grad_norm": 0.5631623268127441,
      "learning_rate": 0.0004987015662389398,
      "loss": 1.143,
      "step": 3505
    },
    {
      "epoch": 0.42192571222502706,
      "grad_norm": 0.7245007753372192,
      "learning_rate": 0.0004986967083348325,
      "loss": 1.0203,
      "step": 3510
    },
    {
      "epoch": 0.4225267460031254,
      "grad_norm": 0.5436373353004456,
      "learning_rate": 0.0004986918413838583,
      "loss": 1.0805,
      "step": 3515
    },
    {
      "epoch": 0.4231277797812237,
      "grad_norm": 0.4831235110759735,
      "learning_rate": 0.0004986869653861941,
      "loss": 1.2867,
      "step": 3520
    },
    {
      "epoch": 0.423728813559322,
      "grad_norm": 0.4958447813987732,
      "learning_rate": 0.0004986820803420172,
      "loss": 1.143,
      "step": 3525
    },
    {
      "epoch": 0.42432984733742035,
      "grad_norm": 0.7917611002922058,
      "learning_rate": 0.0004986771862515055,
      "loss": 1.2012,
      "step": 3530
    },
    {
      "epoch": 0.4249308811155187,
      "grad_norm": 0.6926260590553284,
      "learning_rate": 0.0004986722831148369,
      "loss": 1.3297,
      "step": 3535
    },
    {
      "epoch": 0.425531914893617,
      "grad_norm": 0.5297942757606506,
      "learning_rate": 0.0004986673709321898,
      "loss": 1.0195,
      "step": 3540
    },
    {
      "epoch": 0.42613294867171536,
      "grad_norm": 0.6709704399108887,
      "learning_rate": 0.0004986624497037429,
      "loss": 1.307,
      "step": 3545
    },
    {
      "epoch": 0.4267339824498137,
      "grad_norm": 0.586955189704895,
      "learning_rate": 0.0004986575194296752,
      "loss": 0.9141,
      "step": 3550
    },
    {
      "epoch": 0.42733501622791203,
      "grad_norm": 0.573215663433075,
      "learning_rate": 0.000498652580110166,
      "loss": 1.0344,
      "step": 3555
    },
    {
      "epoch": 0.4279360500060103,
      "grad_norm": 0.4756597578525543,
      "learning_rate": 0.0004986476317453951,
      "loss": 1.1664,
      "step": 3560
    },
    {
      "epoch": 0.42853708378410865,
      "grad_norm": 0.5231835246086121,
      "learning_rate": 0.0004986426743355425,
      "loss": 1.1281,
      "step": 3565
    },
    {
      "epoch": 0.429138117562207,
      "grad_norm": 0.4669915735721588,
      "learning_rate": 0.0004986377078807884,
      "loss": 1.2641,
      "step": 3570
    },
    {
      "epoch": 0.4297391513403053,
      "grad_norm": 0.49261239171028137,
      "learning_rate": 0.0004986327323813135,
      "loss": 1.1812,
      "step": 3575
    },
    {
      "epoch": 0.43034018511840366,
      "grad_norm": 0.5701055526733398,
      "learning_rate": 0.0004986277478372989,
      "loss": 0.943,
      "step": 3580
    },
    {
      "epoch": 0.430941218896502,
      "grad_norm": 0.8740291595458984,
      "learning_rate": 0.0004986227542489259,
      "loss": 1.4688,
      "step": 3585
    },
    {
      "epoch": 0.43154225267460034,
      "grad_norm": 0.5957376956939697,
      "learning_rate": 0.000498617751616376,
      "loss": 1.0031,
      "step": 3590
    },
    {
      "epoch": 0.4321432864526986,
      "grad_norm": 0.5455657243728638,
      "learning_rate": 0.0004986127399398315,
      "loss": 1.1375,
      "step": 3595
    },
    {
      "epoch": 0.43274432023079695,
      "grad_norm": 0.7837244868278503,
      "learning_rate": 0.0004986077192194743,
      "loss": 1.1492,
      "step": 3600
    },
    {
      "epoch": 0.43274432023079695,
      "eval_loss": 2.097851514816284,
      "eval_model_preparation_time": 0.0053,
      "eval_runtime": 35.2375,
      "eval_samples_per_second": 4.541,
      "eval_steps_per_second": 1.135,
      "step": 3600
    },
    {
      "epoch": 0.4333453540088953,
      "grad_norm": 0.6420156359672546,
      "learning_rate": 0.0004986026894554874,
      "loss": 1.6125,
      "step": 3605
    },
    {
      "epoch": 0.4339463877869936,
      "grad_norm": 0.37207821011543274,
      "learning_rate": 0.0004985976506480535,
      "loss": 1.0586,
      "step": 3610
    },
    {
      "epoch": 0.43454742156509196,
      "grad_norm": 0.4620397090911865,
      "learning_rate": 0.000498592602797356,
      "loss": 1.1328,
      "step": 3615
    },
    {
      "epoch": 0.4351484553431903,
      "grad_norm": 0.8707792162895203,
      "learning_rate": 0.0004985875459035786,
      "loss": 1.2578,
      "step": 3620
    },
    {
      "epoch": 0.43574948912128864,
      "grad_norm": 0.5785757303237915,
      "learning_rate": 0.0004985824799669052,
      "loss": 1.1313,
      "step": 3625
    },
    {
      "epoch": 0.43635052289938697,
      "grad_norm": 0.581447958946228,
      "learning_rate": 0.00049857740498752,
      "loss": 1.2219,
      "step": 3630
    },
    {
      "epoch": 0.43695155667748525,
      "grad_norm": 0.5232359170913696,
      "learning_rate": 0.0004985723209656078,
      "loss": 1.0891,
      "step": 3635
    },
    {
      "epoch": 0.4375525904555836,
      "grad_norm": 0.4474778175354004,
      "learning_rate": 0.0004985672279013534,
      "loss": 1.625,
      "step": 3640
    },
    {
      "epoch": 0.4381536242336819,
      "grad_norm": 0.7095141410827637,
      "learning_rate": 0.000498562125794942,
      "loss": 0.9273,
      "step": 3645
    },
    {
      "epoch": 0.43875465801178026,
      "grad_norm": 0.6456997394561768,
      "learning_rate": 0.0004985570146465593,
      "loss": 1.1453,
      "step": 3650
    },
    {
      "epoch": 0.4393556917898786,
      "grad_norm": 0.7230183482170105,
      "learning_rate": 0.0004985518944563914,
      "loss": 1.15,
      "step": 3655
    },
    {
      "epoch": 0.43995672556797694,
      "grad_norm": 0.6240507364273071,
      "learning_rate": 0.0004985467652246243,
      "loss": 1.0586,
      "step": 3660
    },
    {
      "epoch": 0.4405577593460753,
      "grad_norm": 1.0956453084945679,
      "learning_rate": 0.0004985416269514447,
      "loss": 1.3141,
      "step": 3665
    },
    {
      "epoch": 0.44115879312417355,
      "grad_norm": 0.4983053505420685,
      "learning_rate": 0.0004985364796370394,
      "loss": 1.0805,
      "step": 3670
    },
    {
      "epoch": 0.4417598269022719,
      "grad_norm": 0.5128179788589478,
      "learning_rate": 0.0004985313232815958,
      "loss": 0.9055,
      "step": 3675
    },
    {
      "epoch": 0.4423608606803702,
      "grad_norm": 0.535322368144989,
      "learning_rate": 0.0004985261578853014,
      "loss": 0.9563,
      "step": 3680
    },
    {
      "epoch": 0.44296189445846856,
      "grad_norm": 0.6636273264884949,
      "learning_rate": 0.000498520983448344,
      "loss": 1.2719,
      "step": 3685
    },
    {
      "epoch": 0.4435629282365669,
      "grad_norm": 0.7313540577888489,
      "learning_rate": 0.0004985157999709122,
      "loss": 1.5672,
      "step": 3690
    },
    {
      "epoch": 0.44416396201466524,
      "grad_norm": 0.45398956537246704,
      "learning_rate": 0.000498510607453194,
      "loss": 1.0688,
      "step": 3695
    },
    {
      "epoch": 0.4447649957927636,
      "grad_norm": 0.4940120577812195,
      "learning_rate": 0.0004985054058953788,
      "loss": 1.2211,
      "step": 3700
    },
    {
      "epoch": 0.4453660295708619,
      "grad_norm": 0.48061200976371765,
      "learning_rate": 0.0004985001952976556,
      "loss": 1.1328,
      "step": 3705
    },
    {
      "epoch": 0.4459670633489602,
      "grad_norm": 0.6256884336471558,
      "learning_rate": 0.0004984949756602139,
      "loss": 1.2969,
      "step": 3710
    },
    {
      "epoch": 0.4465680971270585,
      "grad_norm": 0.5207997560501099,
      "learning_rate": 0.0004984897469832437,
      "loss": 1.3641,
      "step": 3715
    },
    {
      "epoch": 0.44716913090515686,
      "grad_norm": 0.4868186414241791,
      "learning_rate": 0.000498484509266935,
      "loss": 0.8984,
      "step": 3720
    },
    {
      "epoch": 0.4477701646832552,
      "grad_norm": 0.5324142575263977,
      "learning_rate": 0.0004984792625114786,
      "loss": 1.1617,
      "step": 3725
    },
    {
      "epoch": 0.44837119846135354,
      "grad_norm": 0.5691832304000854,
      "learning_rate": 0.0004984740067170651,
      "loss": 1.2117,
      "step": 3730
    },
    {
      "epoch": 0.4489722322394519,
      "grad_norm": 0.419612318277359,
      "learning_rate": 0.0004984687418838859,
      "loss": 1.0063,
      "step": 3735
    },
    {
      "epoch": 0.4495732660175502,
      "grad_norm": 0.5638041496276855,
      "learning_rate": 0.0004984634680121325,
      "loss": 1.0805,
      "step": 3740
    },
    {
      "epoch": 0.4501742997956485,
      "grad_norm": 0.5611728429794312,
      "learning_rate": 0.0004984581851019966,
      "loss": 1.282,
      "step": 3745
    },
    {
      "epoch": 0.45077533357374683,
      "grad_norm": 0.6137074828147888,
      "learning_rate": 0.0004984528931536705,
      "loss": 1.2445,
      "step": 3750
    },
    {
      "epoch": 0.45137636735184516,
      "grad_norm": 0.4265820384025574,
      "learning_rate": 0.0004984475921673466,
      "loss": 0.9336,
      "step": 3755
    },
    {
      "epoch": 0.4519774011299435,
      "grad_norm": 0.6829708218574524,
      "learning_rate": 0.0004984422821432178,
      "loss": 1.4375,
      "step": 3760
    },
    {
      "epoch": 0.45257843490804184,
      "grad_norm": 0.41247573494911194,
      "learning_rate": 0.0004984369630814773,
      "loss": 1.3852,
      "step": 3765
    },
    {
      "epoch": 0.4531794686861402,
      "grad_norm": 0.5414575338363647,
      "learning_rate": 0.0004984316349823186,
      "loss": 1.0758,
      "step": 3770
    },
    {
      "epoch": 0.4537805024642385,
      "grad_norm": 0.5673078298568726,
      "learning_rate": 0.0004984262978459355,
      "loss": 0.9523,
      "step": 3775
    },
    {
      "epoch": 0.4543815362423368,
      "grad_norm": 0.5617050528526306,
      "learning_rate": 0.0004984209516725221,
      "loss": 1.0805,
      "step": 3780
    },
    {
      "epoch": 0.45498257002043513,
      "grad_norm": 0.5333116054534912,
      "learning_rate": 0.0004984155964622729,
      "loss": 1.0984,
      "step": 3785
    },
    {
      "epoch": 0.45558360379853347,
      "grad_norm": 0.6070541739463806,
      "learning_rate": 0.0004984102322153827,
      "loss": 1.2125,
      "step": 3790
    },
    {
      "epoch": 0.4561846375766318,
      "grad_norm": 0.6610611081123352,
      "learning_rate": 0.0004984048589320467,
      "loss": 1.1281,
      "step": 3795
    },
    {
      "epoch": 0.45678567135473014,
      "grad_norm": 0.6735252737998962,
      "learning_rate": 0.0004983994766124602,
      "loss": 1.0371,
      "step": 3800
    },
    {
      "epoch": 0.45678567135473014,
      "eval_loss": 2.081835985183716,
      "eval_model_preparation_time": 0.0053,
      "eval_runtime": 35.1969,
      "eval_samples_per_second": 4.546,
      "eval_steps_per_second": 1.136,
      "step": 3800
    },
    {
      "epoch": 0.4573867051328285,
      "grad_norm": 0.48256927728652954,
      "learning_rate": 0.0004983940852568193,
      "loss": 1.0188,
      "step": 3805
    },
    {
      "epoch": 0.4579877389109268,
      "grad_norm": 0.4252387583255768,
      "learning_rate": 0.0004983886848653197,
      "loss": 1.1875,
      "step": 3810
    },
    {
      "epoch": 0.45858877268902515,
      "grad_norm": 0.5931745767593384,
      "learning_rate": 0.0004983832754381582,
      "loss": 1.0766,
      "step": 3815
    },
    {
      "epoch": 0.45918980646712343,
      "grad_norm": 0.5438306927680969,
      "learning_rate": 0.0004983778569755315,
      "loss": 1.4563,
      "step": 3820
    },
    {
      "epoch": 0.45979084024522177,
      "grad_norm": 0.4355262219905853,
      "learning_rate": 0.0004983724294776366,
      "loss": 1.0938,
      "step": 3825
    },
    {
      "epoch": 0.4603918740233201,
      "grad_norm": 0.44057753682136536,
      "learning_rate": 0.0004983669929446711,
      "loss": 1.0375,
      "step": 3830
    },
    {
      "epoch": 0.46099290780141844,
      "grad_norm": 0.5248241424560547,
      "learning_rate": 0.0004983615473768326,
      "loss": 1.1828,
      "step": 3835
    },
    {
      "epoch": 0.4615939415795168,
      "grad_norm": 0.6102302074432373,
      "learning_rate": 0.0004983560927743193,
      "loss": 1.2516,
      "step": 3840
    },
    {
      "epoch": 0.4621949753576151,
      "grad_norm": 0.6916151642799377,
      "learning_rate": 0.0004983506291373295,
      "loss": 1.6047,
      "step": 3845
    },
    {
      "epoch": 0.46279600913571345,
      "grad_norm": 0.5055292844772339,
      "learning_rate": 0.0004983451564660622,
      "loss": 1.0172,
      "step": 3850
    },
    {
      "epoch": 0.46339704291381173,
      "grad_norm": 0.6418437957763672,
      "learning_rate": 0.0004983396747607161,
      "loss": 1.1805,
      "step": 3855
    },
    {
      "epoch": 0.46399807669191007,
      "grad_norm": 0.5011945962905884,
      "learning_rate": 0.000498334184021491,
      "loss": 1.0055,
      "step": 3860
    },
    {
      "epoch": 0.4645991104700084,
      "grad_norm": 0.5504122376441956,
      "learning_rate": 0.0004983286842485864,
      "loss": 0.9742,
      "step": 3865
    },
    {
      "epoch": 0.46520014424810674,
      "grad_norm": 0.3638380467891693,
      "learning_rate": 0.0004983231754422024,
      "loss": 0.9906,
      "step": 3870
    },
    {
      "epoch": 0.4658011780262051,
      "grad_norm": 0.5893705487251282,
      "learning_rate": 0.0004983176576025394,
      "loss": 1.0367,
      "step": 3875
    },
    {
      "epoch": 0.4664022118043034,
      "grad_norm": 0.5191811323165894,
      "learning_rate": 0.0004983121307297983,
      "loss": 1.4234,
      "step": 3880
    },
    {
      "epoch": 0.46700324558240175,
      "grad_norm": 0.5002449750900269,
      "learning_rate": 0.0004983065948241799,
      "loss": 1.1242,
      "step": 3885
    },
    {
      "epoch": 0.4676042793605001,
      "grad_norm": 0.5473368167877197,
      "learning_rate": 0.0004983010498858857,
      "loss": 1.1219,
      "step": 3890
    },
    {
      "epoch": 0.46820531313859837,
      "grad_norm": 0.5295068621635437,
      "learning_rate": 0.0004982954959151174,
      "loss": 1.4508,
      "step": 3895
    },
    {
      "epoch": 0.4688063469166967,
      "grad_norm": 0.8546839356422424,
      "learning_rate": 0.000498289932912077,
      "loss": 0.9664,
      "step": 3900
    },
    {
      "epoch": 0.46940738069479504,
      "grad_norm": 0.6534102559089661,
      "learning_rate": 0.000498284360876967,
      "loss": 1.4344,
      "step": 3905
    },
    {
      "epoch": 0.4700084144728934,
      "grad_norm": 0.4570360779762268,
      "learning_rate": 0.0004982787798099898,
      "loss": 0.9531,
      "step": 3910
    },
    {
      "epoch": 0.4706094482509917,
      "grad_norm": 0.5392407178878784,
      "learning_rate": 0.0004982731897113488,
      "loss": 1.243,
      "step": 3915
    },
    {
      "epoch": 0.47121048202909005,
      "grad_norm": 0.7176635265350342,
      "learning_rate": 0.0004982675905812469,
      "loss": 0.968,
      "step": 3920
    },
    {
      "epoch": 0.4718115158071884,
      "grad_norm": 0.5123677253723145,
      "learning_rate": 0.0004982619824198882,
      "loss": 1.4969,
      "step": 3925
    },
    {
      "epoch": 0.47241254958528667,
      "grad_norm": 0.5915489196777344,
      "learning_rate": 0.0004982563652274766,
      "loss": 1.3805,
      "step": 3930
    },
    {
      "epoch": 0.473013583363385,
      "grad_norm": 0.5139390230178833,
      "learning_rate": 0.0004982507390042163,
      "loss": 1.057,
      "step": 3935
    },
    {
      "epoch": 0.47361461714148334,
      "grad_norm": 0.6480752229690552,
      "learning_rate": 0.0004982451037503121,
      "loss": 1.0063,
      "step": 3940
    },
    {
      "epoch": 0.4742156509195817,
      "grad_norm": 0.6889259219169617,
      "learning_rate": 0.0004982394594659689,
      "loss": 1.4828,
      "step": 3945
    },
    {
      "epoch": 0.47481668469768,
      "grad_norm": 0.5907976627349854,
      "learning_rate": 0.0004982338061513921,
      "loss": 1.2148,
      "step": 3950
    },
    {
      "epoch": 0.47541771847577835,
      "grad_norm": 0.6065306067466736,
      "learning_rate": 0.0004982281438067874,
      "loss": 1.3336,
      "step": 3955
    },
    {
      "epoch": 0.4760187522538767,
      "grad_norm": 0.5058939456939697,
      "learning_rate": 0.0004982224724323606,
      "loss": 1.2148,
      "step": 3960
    },
    {
      "epoch": 0.476619786031975,
      "grad_norm": 0.633465588092804,
      "learning_rate": 0.0004982167920283181,
      "loss": 1.1977,
      "step": 3965
    },
    {
      "epoch": 0.4772208198100733,
      "grad_norm": 0.48402997851371765,
      "learning_rate": 0.0004982111025948666,
      "loss": 1.0758,
      "step": 3970
    },
    {
      "epoch": 0.47782185358817164,
      "grad_norm": 0.4740035831928253,
      "learning_rate": 0.000498205404132213,
      "loss": 0.9914,
      "step": 3975
    },
    {
      "epoch": 0.47842288736627,
      "grad_norm": 0.6403579711914062,
      "learning_rate": 0.0004981996966405646,
      "loss": 0.9035,
      "step": 3980
    },
    {
      "epoch": 0.4790239211443683,
      "grad_norm": 0.6459212303161621,
      "learning_rate": 0.000498193980120129,
      "loss": 1.1273,
      "step": 3985
    },
    {
      "epoch": 0.47962495492246665,
      "grad_norm": 0.5674965977668762,
      "learning_rate": 0.0004981882545711142,
      "loss": 1.3828,
      "step": 3990
    },
    {
      "epoch": 0.480225988700565,
      "grad_norm": 0.6159428954124451,
      "learning_rate": 0.0004981825199937285,
      "loss": 1.2344,
      "step": 3995
    },
    {
      "epoch": 0.4808270224786633,
      "grad_norm": 0.6476476788520813,
      "learning_rate": 0.0004981767763881803,
      "loss": 1.4625,
      "step": 4000
    },
    {
      "epoch": 0.4808270224786633,
      "eval_loss": 2.0904297828674316,
      "eval_model_preparation_time": 0.0053,
      "eval_runtime": 35.1956,
      "eval_samples_per_second": 4.546,
      "eval_steps_per_second": 1.137,
      "step": 4000
    },
    {
      "epoch": 0.4814280562567616,
      "grad_norm": 0.7665592432022095,
      "learning_rate": 0.0004981710237546789,
      "loss": 1.2703,
      "step": 4005
    },
    {
      "epoch": 0.48202909003485994,
      "grad_norm": 0.5057299733161926,
      "learning_rate": 0.0004981652620934333,
      "loss": 1.3039,
      "step": 4010
    },
    {
      "epoch": 0.4826301238129583,
      "grad_norm": 0.46735164523124695,
      "learning_rate": 0.0004981594914046532,
      "loss": 1.0281,
      "step": 4015
    },
    {
      "epoch": 0.4832311575910566,
      "grad_norm": 0.6298277378082275,
      "learning_rate": 0.0004981537116885484,
      "loss": 1.0656,
      "step": 4020
    },
    {
      "epoch": 0.48383219136915495,
      "grad_norm": 0.6472790837287903,
      "learning_rate": 0.0004981479229453292,
      "loss": 1.0418,
      "step": 4025
    },
    {
      "epoch": 0.4844332251472533,
      "grad_norm": 0.5558010339736938,
      "learning_rate": 0.0004981421251752063,
      "loss": 1.0992,
      "step": 4030
    },
    {
      "epoch": 0.4850342589253516,
      "grad_norm": 0.46242016553878784,
      "learning_rate": 0.0004981363183783903,
      "loss": 1.002,
      "step": 4035
    },
    {
      "epoch": 0.48563529270344996,
      "grad_norm": 0.48881420493125916,
      "learning_rate": 0.0004981305025550929,
      "loss": 1.1867,
      "step": 4040
    },
    {
      "epoch": 0.48623632648154824,
      "grad_norm": 0.4354248642921448,
      "learning_rate": 0.0004981246777055252,
      "loss": 1.4141,
      "step": 4045
    },
    {
      "epoch": 0.4868373602596466,
      "grad_norm": 0.5004878044128418,
      "learning_rate": 0.0004981188438298995,
      "loss": 0.9684,
      "step": 4050
    },
    {
      "epoch": 0.4874383940377449,
      "grad_norm": 0.5613626837730408,
      "learning_rate": 0.0004981130009284277,
      "loss": 1.1555,
      "step": 4055
    },
    {
      "epoch": 0.48803942781584325,
      "grad_norm": 0.4278530776500702,
      "learning_rate": 0.0004981071490013225,
      "loss": 1.1289,
      "step": 4060
    },
    {
      "epoch": 0.4886404615939416,
      "grad_norm": 0.6259580850601196,
      "learning_rate": 0.0004981012880487968,
      "loss": 0.9828,
      "step": 4065
    },
    {
      "epoch": 0.4892414953720399,
      "grad_norm": 0.5686140656471252,
      "learning_rate": 0.0004980954180710636,
      "loss": 1.1805,
      "step": 4070
    },
    {
      "epoch": 0.48984252915013826,
      "grad_norm": 1.0444506406784058,
      "learning_rate": 0.0004980895390683367,
      "loss": 0.9242,
      "step": 4075
    },
    {
      "epoch": 0.49044356292823654,
      "grad_norm": 0.49180054664611816,
      "learning_rate": 0.0004980836510408297,
      "loss": 0.9906,
      "step": 4080
    },
    {
      "epoch": 0.4910445967063349,
      "grad_norm": 0.39391833543777466,
      "learning_rate": 0.000498077753988757,
      "loss": 0.9578,
      "step": 4085
    },
    {
      "epoch": 0.4916456304844332,
      "grad_norm": 0.6872186660766602,
      "learning_rate": 0.0004980718479123332,
      "loss": 1.2711,
      "step": 4090
    },
    {
      "epoch": 0.49224666426253155,
      "grad_norm": 0.4728960692882538,
      "learning_rate": 0.0004980659328117728,
      "loss": 1.0828,
      "step": 4095
    },
    {
      "epoch": 0.4928476980406299,
      "grad_norm": 0.4187394380569458,
      "learning_rate": 0.0004980600086872913,
      "loss": 1.0727,
      "step": 4100
    },
    {
      "epoch": 0.4934487318187282,
      "grad_norm": 0.5191169381141663,
      "learning_rate": 0.000498054075539104,
      "loss": 0.9246,
      "step": 4105
    },
    {
      "epoch": 0.49404976559682656,
      "grad_norm": 0.5990204811096191,
      "learning_rate": 0.0004980481333674269,
      "loss": 1.2625,
      "step": 4110
    },
    {
      "epoch": 0.49465079937492484,
      "grad_norm": 0.5898841619491577,
      "learning_rate": 0.0004980421821724759,
      "loss": 0.9133,
      "step": 4115
    },
    {
      "epoch": 0.4952518331530232,
      "grad_norm": 0.576693594455719,
      "learning_rate": 0.0004980362219544677,
      "loss": 1.2969,
      "step": 4120
    },
    {
      "epoch": 0.4958528669311215,
      "grad_norm": 0.5620527267456055,
      "learning_rate": 0.000498030252713619,
      "loss": 0.9883,
      "step": 4125
    },
    {
      "epoch": 0.49645390070921985,
      "grad_norm": 0.46309694647789,
      "learning_rate": 0.0004980242744501472,
      "loss": 1.3211,
      "step": 4130
    },
    {
      "epoch": 0.4970549344873182,
      "grad_norm": 0.6207218170166016,
      "learning_rate": 0.0004980182871642694,
      "loss": 1.2563,
      "step": 4135
    },
    {
      "epoch": 0.4976559682654165,
      "grad_norm": 0.5662885904312134,
      "learning_rate": 0.0004980122908562036,
      "loss": 1.3328,
      "step": 4140
    },
    {
      "epoch": 0.49825700204351486,
      "grad_norm": 0.6134732961654663,
      "learning_rate": 0.000498006285526168,
      "loss": 1.132,
      "step": 4145
    },
    {
      "epoch": 0.4988580358216132,
      "grad_norm": 0.7582330107688904,
      "learning_rate": 0.0004980002711743809,
      "loss": 0.8836,
      "step": 4150
    },
    {
      "epoch": 0.4994590695997115,
      "grad_norm": 0.6320111155509949,
      "learning_rate": 0.0004979942478010612,
      "loss": 1.1727,
      "step": 4155
    },
    {
      "epoch": 0.5000601033778098,
      "grad_norm": 0.5162654519081116,
      "learning_rate": 0.0004979882154064279,
      "loss": 1.2016,
      "step": 4160
    },
    {
      "epoch": 0.5006611371559082,
      "grad_norm": 0.25934234261512756,
      "learning_rate": 0.0004979821739907005,
      "loss": 1.0988,
      "step": 4165
    },
    {
      "epoch": 0.5012621709340065,
      "grad_norm": 0.6046193838119507,
      "learning_rate": 0.0004979761235540988,
      "loss": 1.3953,
      "step": 4170
    },
    {
      "epoch": 0.5018632047121048,
      "grad_norm": 0.5944646596908569,
      "learning_rate": 0.0004979700640968429,
      "loss": 1.1305,
      "step": 4175
    },
    {
      "epoch": 0.5024642384902032,
      "grad_norm": 0.5635605454444885,
      "learning_rate": 0.0004979639956191531,
      "loss": 1.2656,
      "step": 4180
    },
    {
      "epoch": 0.5030652722683014,
      "grad_norm": 0.5029686689376831,
      "learning_rate": 0.0004979579181212504,
      "loss": 1.2305,
      "step": 4185
    },
    {
      "epoch": 0.5036663060463998,
      "grad_norm": 0.8044158816337585,
      "learning_rate": 0.0004979518316033556,
      "loss": 1.225,
      "step": 4190
    },
    {
      "epoch": 0.5042673398244981,
      "grad_norm": 0.6193830966949463,
      "learning_rate": 0.0004979457360656902,
      "loss": 0.9453,
      "step": 4195
    },
    {
      "epoch": 0.5048683736025965,
      "grad_norm": 0.5999032258987427,
      "learning_rate": 0.0004979396315084761,
      "loss": 1.2148,
      "step": 4200
    },
    {
      "epoch": 0.5048683736025965,
      "eval_loss": 2.069140672683716,
      "eval_model_preparation_time": 0.0053,
      "eval_runtime": 35.1981,
      "eval_samples_per_second": 4.546,
      "eval_steps_per_second": 1.136,
      "step": 4200
    },
    {
      "epoch": 0.5054694073806948,
      "grad_norm": 0.6504601836204529,
      "learning_rate": 0.0004979335179319352,
      "loss": 1.082,
      "step": 4205
    },
    {
      "epoch": 0.5060704411587931,
      "grad_norm": 0.5235728025436401,
      "learning_rate": 0.00049792739533629,
      "loss": 0.9156,
      "step": 4210
    },
    {
      "epoch": 0.5066714749368915,
      "grad_norm": 0.46760815382003784,
      "learning_rate": 0.0004979212637217631,
      "loss": 1.007,
      "step": 4215
    },
    {
      "epoch": 0.5072725087149897,
      "grad_norm": 0.5325601696968079,
      "learning_rate": 0.0004979151230885776,
      "loss": 1.1094,
      "step": 4220
    },
    {
      "epoch": 0.5078735424930881,
      "grad_norm": 0.4098140299320221,
      "learning_rate": 0.0004979089734369568,
      "loss": 0.8449,
      "step": 4225
    },
    {
      "epoch": 0.5084745762711864,
      "grad_norm": 0.6042998433113098,
      "learning_rate": 0.0004979028147671246,
      "loss": 1.1672,
      "step": 4230
    },
    {
      "epoch": 0.5090756100492848,
      "grad_norm": 0.41027069091796875,
      "learning_rate": 0.0004978966470793049,
      "loss": 1.1992,
      "step": 4235
    },
    {
      "epoch": 0.5096766438273831,
      "grad_norm": 0.5344287753105164,
      "learning_rate": 0.0004978904703737221,
      "loss": 0.9934,
      "step": 4240
    },
    {
      "epoch": 0.5102776776054815,
      "grad_norm": 0.5234737992286682,
      "learning_rate": 0.000497884284650601,
      "loss": 1.0617,
      "step": 4245
    },
    {
      "epoch": 0.5108787113835798,
      "grad_norm": 0.4804365336894989,
      "learning_rate": 0.0004978780899101663,
      "loss": 1.0098,
      "step": 4250
    },
    {
      "epoch": 0.511479745161678,
      "grad_norm": 0.6976252198219299,
      "learning_rate": 0.0004978718861526438,
      "loss": 1.2852,
      "step": 4255
    },
    {
      "epoch": 0.5120807789397764,
      "grad_norm": 0.4136990010738373,
      "learning_rate": 0.0004978656733782588,
      "loss": 1.1062,
      "step": 4260
    },
    {
      "epoch": 0.5126818127178747,
      "grad_norm": 0.5344454646110535,
      "learning_rate": 0.0004978594515872373,
      "loss": 1.2984,
      "step": 4265
    },
    {
      "epoch": 0.5132828464959731,
      "grad_norm": 0.42136499285697937,
      "learning_rate": 0.0004978532207798059,
      "loss": 1.0766,
      "step": 4270
    },
    {
      "epoch": 0.5138838802740714,
      "grad_norm": 0.4809805750846863,
      "learning_rate": 0.0004978469809561911,
      "loss": 1.1469,
      "step": 4275
    },
    {
      "epoch": 0.5144849140521698,
      "grad_norm": 0.5629724264144897,
      "learning_rate": 0.0004978407321166199,
      "loss": 1.218,
      "step": 4280
    },
    {
      "epoch": 0.5150859478302681,
      "grad_norm": 0.34920796751976013,
      "learning_rate": 0.0004978344742613195,
      "loss": 0.8383,
      "step": 4285
    },
    {
      "epoch": 0.5156869816083663,
      "grad_norm": 0.5682929158210754,
      "learning_rate": 0.0004978282073905178,
      "loss": 1.198,
      "step": 4290
    },
    {
      "epoch": 0.5162880153864647,
      "grad_norm": 0.49201318621635437,
      "learning_rate": 0.0004978219315044426,
      "loss": 1.2352,
      "step": 4295
    },
    {
      "epoch": 0.516889049164563,
      "grad_norm": 0.721458911895752,
      "learning_rate": 0.0004978156466033222,
      "loss": 0.9203,
      "step": 4300
    },
    {
      "epoch": 0.5174900829426614,
      "grad_norm": 0.5688953995704651,
      "learning_rate": 0.0004978093526873853,
      "loss": 1.1977,
      "step": 4305
    },
    {
      "epoch": 0.5180911167207597,
      "grad_norm": 0.6028774976730347,
      "learning_rate": 0.0004978030497568607,
      "loss": 1.475,
      "step": 4310
    },
    {
      "epoch": 0.5186921504988581,
      "grad_norm": 0.6064693331718445,
      "learning_rate": 0.000497796737811978,
      "loss": 1.1285,
      "step": 4315
    },
    {
      "epoch": 0.5192931842769564,
      "grad_norm": 0.47966837882995605,
      "learning_rate": 0.0004977904168529664,
      "loss": 1.2102,
      "step": 4320
    },
    {
      "epoch": 0.5198942180550546,
      "grad_norm": 0.7327610850334167,
      "learning_rate": 0.0004977840868800561,
      "loss": 1.1367,
      "step": 4325
    },
    {
      "epoch": 0.520495251833153,
      "grad_norm": 0.5901091694831848,
      "learning_rate": 0.0004977777478934774,
      "loss": 1.3352,
      "step": 4330
    },
    {
      "epoch": 0.5210962856112513,
      "grad_norm": 0.5034139752388,
      "learning_rate": 0.0004977713998934607,
      "loss": 0.9094,
      "step": 4335
    },
    {
      "epoch": 0.5216973193893497,
      "grad_norm": 0.43822070956230164,
      "learning_rate": 0.0004977650428802371,
      "loss": 1.3539,
      "step": 4340
    },
    {
      "epoch": 0.522298353167448,
      "grad_norm": 0.5016989707946777,
      "learning_rate": 0.0004977586768540377,
      "loss": 1.2398,
      "step": 4345
    },
    {
      "epoch": 0.5228993869455464,
      "grad_norm": 0.4488658607006073,
      "learning_rate": 0.0004977523018150941,
      "loss": 1.0188,
      "step": 4350
    },
    {
      "epoch": 0.5235004207236447,
      "grad_norm": 0.5360990762710571,
      "learning_rate": 0.0004977459177636384,
      "loss": 1.1133,
      "step": 4355
    },
    {
      "epoch": 0.524101454501743,
      "grad_norm": 0.41790762543678284,
      "learning_rate": 0.0004977395246999026,
      "loss": 1.0016,
      "step": 4360
    },
    {
      "epoch": 0.5247024882798413,
      "grad_norm": 0.8560382723808289,
      "learning_rate": 0.0004977331226241194,
      "loss": 1.1953,
      "step": 4365
    },
    {
      "epoch": 0.5253035220579396,
      "grad_norm": 0.5320670008659363,
      "learning_rate": 0.0004977267115365216,
      "loss": 1.0711,
      "step": 4370
    },
    {
      "epoch": 0.525904555836038,
      "grad_norm": 0.5246152877807617,
      "learning_rate": 0.0004977202914373426,
      "loss": 1.2891,
      "step": 4375
    },
    {
      "epoch": 0.5265055896141363,
      "grad_norm": 0.6537752747535706,
      "learning_rate": 0.0004977138623268156,
      "loss": 1.2719,
      "step": 4380
    },
    {
      "epoch": 0.5271066233922347,
      "grad_norm": 0.5286105275154114,
      "learning_rate": 0.0004977074242051748,
      "loss": 1.1391,
      "step": 4385
    },
    {
      "epoch": 0.527707657170333,
      "grad_norm": 0.5731997489929199,
      "learning_rate": 0.0004977009770726541,
      "loss": 1.0484,
      "step": 4390
    },
    {
      "epoch": 0.5283086909484312,
      "grad_norm": 0.6958596110343933,
      "learning_rate": 0.0004976945209294884,
      "loss": 0.9648,
      "step": 4395
    },
    {
      "epoch": 0.5289097247265296,
      "grad_norm": 0.8171592950820923,
      "learning_rate": 0.0004976880557759124,
      "loss": 1.5773,
      "step": 4400
    },
    {
      "epoch": 0.5289097247265296,
      "eval_loss": 2.049023389816284,
      "eval_model_preparation_time": 0.0053,
      "eval_runtime": 35.1838,
      "eval_samples_per_second": 4.548,
      "eval_steps_per_second": 1.137,
      "step": 4400
    },
    {
      "epoch": 0.5295107585046279,
      "grad_norm": 0.4283256530761719,
      "learning_rate": 0.000497681581612161,
      "loss": 0.9379,
      "step": 4405
    },
    {
      "epoch": 0.5301117922827263,
      "grad_norm": 0.6683317422866821,
      "learning_rate": 0.0004976750984384701,
      "loss": 1.3484,
      "step": 4410
    },
    {
      "epoch": 0.5307128260608246,
      "grad_norm": 0.7179862260818481,
      "learning_rate": 0.0004976686062550754,
      "loss": 0.7375,
      "step": 4415
    },
    {
      "epoch": 0.531313859838923,
      "grad_norm": 0.5547930002212524,
      "learning_rate": 0.000497662105062213,
      "loss": 0.932,
      "step": 4420
    },
    {
      "epoch": 0.5319148936170213,
      "grad_norm": 0.6749304533004761,
      "learning_rate": 0.0004976555948601194,
      "loss": 1.0969,
      "step": 4425
    },
    {
      "epoch": 0.5325159273951197,
      "grad_norm": 0.4598415493965149,
      "learning_rate": 0.0004976490756490316,
      "loss": 1.2563,
      "step": 4430
    },
    {
      "epoch": 0.5331169611732179,
      "grad_norm": 1.0315395593643188,
      "learning_rate": 0.0004976425474291866,
      "loss": 0.9234,
      "step": 4435
    },
    {
      "epoch": 0.5337179949513162,
      "grad_norm": 0.5202517509460449,
      "learning_rate": 0.0004976360102008219,
      "loss": 0.9219,
      "step": 4440
    },
    {
      "epoch": 0.5343190287294146,
      "grad_norm": 0.6117017269134521,
      "learning_rate": 0.0004976294639641753,
      "loss": 1.1805,
      "step": 4445
    },
    {
      "epoch": 0.5349200625075129,
      "grad_norm": 0.6244452595710754,
      "learning_rate": 0.000497622908719485,
      "loss": 1.1406,
      "step": 4450
    },
    {
      "epoch": 0.5355210962856113,
      "grad_norm": 0.6520891785621643,
      "learning_rate": 0.0004976163444669893,
      "loss": 1.2078,
      "step": 4455
    },
    {
      "epoch": 0.5361221300637096,
      "grad_norm": 0.6139251589775085,
      "learning_rate": 0.0004976097712069272,
      "loss": 1.0719,
      "step": 4460
    },
    {
      "epoch": 0.536723163841808,
      "grad_norm": 0.46594560146331787,
      "learning_rate": 0.0004976031889395376,
      "loss": 0.9531,
      "step": 4465
    },
    {
      "epoch": 0.5373241976199062,
      "grad_norm": 0.6443547606468201,
      "learning_rate": 0.0004975965976650601,
      "loss": 1.0945,
      "step": 4470
    },
    {
      "epoch": 0.5379252313980045,
      "grad_norm": 0.6175803542137146,
      "learning_rate": 0.0004975899973837344,
      "loss": 1.5594,
      "step": 4475
    },
    {
      "epoch": 0.5385262651761029,
      "grad_norm": 0.7328972220420837,
      "learning_rate": 0.0004975833880958006,
      "loss": 1.1953,
      "step": 4480
    },
    {
      "epoch": 0.5391272989542012,
      "grad_norm": 0.485140860080719,
      "learning_rate": 0.0004975767698014992,
      "loss": 1.1117,
      "step": 4485
    },
    {
      "epoch": 0.5397283327322996,
      "grad_norm": 0.5293512344360352,
      "learning_rate": 0.0004975701425010709,
      "loss": 1.025,
      "step": 4490
    },
    {
      "epoch": 0.5403293665103979,
      "grad_norm": 0.7554160952568054,
      "learning_rate": 0.0004975635061947568,
      "loss": 1.0391,
      "step": 4495
    },
    {
      "epoch": 0.5409304002884963,
      "grad_norm": 0.41734662652015686,
      "learning_rate": 0.0004975568608827982,
      "loss": 0.8695,
      "step": 4500
    },
    {
      "epoch": 0.5415314340665945,
      "grad_norm": 0.39808404445648193,
      "learning_rate": 0.0004975502065654371,
      "loss": 1.1461,
      "step": 4505
    },
    {
      "epoch": 0.5421324678446928,
      "grad_norm": 0.6031619906425476,
      "learning_rate": 0.0004975435432429153,
      "loss": 1.2563,
      "step": 4510
    },
    {
      "epoch": 0.5427335016227912,
      "grad_norm": 0.7665862441062927,
      "learning_rate": 0.0004975368709154753,
      "loss": 1.2812,
      "step": 4515
    },
    {
      "epoch": 0.5433345354008895,
      "grad_norm": 0.7608307003974915,
      "learning_rate": 0.0004975301895833598,
      "loss": 1.1766,
      "step": 4520
    },
    {
      "epoch": 0.5439355691789879,
      "grad_norm": 0.6584916114807129,
      "learning_rate": 0.0004975234992468118,
      "loss": 1.0078,
      "step": 4525
    },
    {
      "epoch": 0.5445366029570862,
      "grad_norm": 0.5199896097183228,
      "learning_rate": 0.0004975167999060748,
      "loss": 1.2406,
      "step": 4530
    },
    {
      "epoch": 0.5451376367351846,
      "grad_norm": 0.5209721922874451,
      "learning_rate": 0.0004975100915613925,
      "loss": 1.0367,
      "step": 4535
    },
    {
      "epoch": 0.5457386705132828,
      "grad_norm": 0.6140447854995728,
      "learning_rate": 0.0004975033742130087,
      "loss": 0.9734,
      "step": 4540
    },
    {
      "epoch": 0.5463397042913811,
      "grad_norm": 0.5982779860496521,
      "learning_rate": 0.0004974966478611681,
      "loss": 1.182,
      "step": 4545
    },
    {
      "epoch": 0.5469407380694795,
      "grad_norm": 0.48514458537101746,
      "learning_rate": 0.0004974899125061151,
      "loss": 1.2273,
      "step": 4550
    },
    {
      "epoch": 0.5475417718475778,
      "grad_norm": 0.5001315474510193,
      "learning_rate": 0.0004974831681480949,
      "loss": 1.2789,
      "step": 4555
    },
    {
      "epoch": 0.5481428056256762,
      "grad_norm": 0.4541556239128113,
      "learning_rate": 0.0004974764147873526,
      "loss": 0.9297,
      "step": 4560
    },
    {
      "epoch": 0.5487438394037745,
      "grad_norm": 0.5090661644935608,
      "learning_rate": 0.0004974696524241342,
      "loss": 0.984,
      "step": 4565
    },
    {
      "epoch": 0.5493448731818729,
      "grad_norm": 0.49955224990844727,
      "learning_rate": 0.0004974628810586854,
      "loss": 1.0625,
      "step": 4570
    },
    {
      "epoch": 0.5499459069599711,
      "grad_norm": 0.6976388692855835,
      "learning_rate": 0.0004974561006912527,
      "loss": 1.1914,
      "step": 4575
    },
    {
      "epoch": 0.5505469407380695,
      "grad_norm": 0.4535270929336548,
      "learning_rate": 0.0004974493113220827,
      "loss": 0.7617,
      "step": 4580
    },
    {
      "epoch": 0.5511479745161678,
      "grad_norm": 0.8714334964752197,
      "learning_rate": 0.0004974425129514224,
      "loss": 0.9938,
      "step": 4585
    },
    {
      "epoch": 0.5517490082942661,
      "grad_norm": 0.49312853813171387,
      "learning_rate": 0.000497435705579519,
      "loss": 1.2945,
      "step": 4590
    },
    {
      "epoch": 0.5523500420723645,
      "grad_norm": 0.5688885450363159,
      "learning_rate": 0.0004974288892066203,
      "loss": 1.0527,
      "step": 4595
    },
    {
      "epoch": 0.5529510758504628,
      "grad_norm": 0.5999502539634705,
      "learning_rate": 0.0004974220638329741,
      "loss": 0.9891,
      "step": 4600
    },
    {
      "epoch": 0.5529510758504628,
      "eval_loss": 2.064257860183716,
      "eval_model_preparation_time": 0.0053,
      "eval_runtime": 35.2145,
      "eval_samples_per_second": 4.544,
      "eval_steps_per_second": 1.136,
      "step": 4600
    },
    {
      "epoch": 0.5535521096285612,
      "grad_norm": 0.5289875268936157,
      "learning_rate": 0.0004974152294588289,
      "loss": 1.1328,
      "step": 4605
    },
    {
      "epoch": 0.5541531434066594,
      "grad_norm": 0.7003397941589355,
      "learning_rate": 0.000497408386084433,
      "loss": 0.8477,
      "step": 4610
    },
    {
      "epoch": 0.5547541771847578,
      "grad_norm": 0.5111108422279358,
      "learning_rate": 0.0004974015337100357,
      "loss": 1.168,
      "step": 4615
    },
    {
      "epoch": 0.5553552109628561,
      "grad_norm": 0.4160408675670624,
      "learning_rate": 0.0004973946723358858,
      "loss": 1.3211,
      "step": 4620
    },
    {
      "epoch": 0.5559562447409544,
      "grad_norm": 0.5501172542572021,
      "learning_rate": 0.0004973878019622335,
      "loss": 1.3945,
      "step": 4625
    },
    {
      "epoch": 0.5565572785190528,
      "grad_norm": 0.4727821350097656,
      "learning_rate": 0.0004973809225893282,
      "loss": 1.3078,
      "step": 4630
    },
    {
      "epoch": 0.5571583122971511,
      "grad_norm": 0.5417948365211487,
      "learning_rate": 0.0004973740342174204,
      "loss": 1.1062,
      "step": 4635
    },
    {
      "epoch": 0.5577593460752495,
      "grad_norm": 0.39009690284729004,
      "learning_rate": 0.0004973671368467607,
      "loss": 0.7863,
      "step": 4640
    },
    {
      "epoch": 0.5583603798533477,
      "grad_norm": 0.6143473386764526,
      "learning_rate": 0.0004973602304776,
      "loss": 1.0836,
      "step": 4645
    },
    {
      "epoch": 0.5589614136314461,
      "grad_norm": 0.46591776609420776,
      "learning_rate": 0.0004973533151101893,
      "loss": 1.0797,
      "step": 4650
    },
    {
      "epoch": 0.5595624474095444,
      "grad_norm": 0.636924684047699,
      "learning_rate": 0.0004973463907447804,
      "loss": 0.9961,
      "step": 4655
    },
    {
      "epoch": 0.5601634811876427,
      "grad_norm": 0.7269019484519958,
      "learning_rate": 0.0004973394573816252,
      "loss": 1.0727,
      "step": 4660
    },
    {
      "epoch": 0.5607645149657411,
      "grad_norm": 0.5581966638565063,
      "learning_rate": 0.0004973325150209758,
      "loss": 0.832,
      "step": 4665
    },
    {
      "epoch": 0.5613655487438394,
      "grad_norm": 0.586113691329956,
      "learning_rate": 0.0004973255636630847,
      "loss": 1.132,
      "step": 4670
    },
    {
      "epoch": 0.5619665825219378,
      "grad_norm": 0.7046157121658325,
      "learning_rate": 0.0004973186033082049,
      "loss": 1.1008,
      "step": 4675
    },
    {
      "epoch": 0.562567616300036,
      "grad_norm": 0.6651023626327515,
      "learning_rate": 0.0004973116339565897,
      "loss": 1.1445,
      "step": 4680
    },
    {
      "epoch": 0.5631686500781344,
      "grad_norm": 0.6627675890922546,
      "learning_rate": 0.0004973046556084923,
      "loss": 1.2563,
      "step": 4685
    },
    {
      "epoch": 0.5637696838562327,
      "grad_norm": 0.4929293692111969,
      "learning_rate": 0.0004972976682641668,
      "loss": 1.1836,
      "step": 4690
    },
    {
      "epoch": 0.564370717634331,
      "grad_norm": 0.500619649887085,
      "learning_rate": 0.0004972906719238673,
      "loss": 1.1789,
      "step": 4695
    },
    {
      "epoch": 0.5649717514124294,
      "grad_norm": 0.456321120262146,
      "learning_rate": 0.0004972836665878483,
      "loss": 1.1375,
      "step": 4700
    },
    {
      "epoch": 0.5655727851905277,
      "grad_norm": 0.43033653497695923,
      "learning_rate": 0.0004972766522563648,
      "loss": 0.7367,
      "step": 4705
    },
    {
      "epoch": 0.5661738189686261,
      "grad_norm": 0.5689796805381775,
      "learning_rate": 0.0004972696289296715,
      "loss": 0.7828,
      "step": 4710
    },
    {
      "epoch": 0.5667748527467243,
      "grad_norm": 0.6046550273895264,
      "learning_rate": 0.0004972625966080244,
      "loss": 1.082,
      "step": 4715
    },
    {
      "epoch": 0.5673758865248227,
      "grad_norm": 0.6092924475669861,
      "learning_rate": 0.0004972555552916791,
      "loss": 1.0945,
      "step": 4720
    },
    {
      "epoch": 0.567976920302921,
      "grad_norm": 0.6022126078605652,
      "learning_rate": 0.0004972485049808918,
      "loss": 1.0648,
      "step": 4725
    },
    {
      "epoch": 0.5685779540810193,
      "grad_norm": 0.6475672721862793,
      "learning_rate": 0.0004972414456759189,
      "loss": 1.4375,
      "step": 4730
    },
    {
      "epoch": 0.5691789878591177,
      "grad_norm": 0.5474737882614136,
      "learning_rate": 0.0004972343773770172,
      "loss": 1.2367,
      "step": 4735
    },
    {
      "epoch": 0.569780021637216,
      "grad_norm": 0.5624226331710815,
      "learning_rate": 0.0004972273000844439,
      "loss": 0.9305,
      "step": 4740
    },
    {
      "epoch": 0.5703810554153144,
      "grad_norm": 0.5779100656509399,
      "learning_rate": 0.0004972202137984564,
      "loss": 0.9156,
      "step": 4745
    },
    {
      "epoch": 0.5709820891934126,
      "grad_norm": 0.8340283036231995,
      "learning_rate": 0.0004972131185193123,
      "loss": 1.3273,
      "step": 4750
    },
    {
      "epoch": 0.571583122971511,
      "grad_norm": 0.5648449063301086,
      "learning_rate": 0.0004972060142472702,
      "loss": 1.1727,
      "step": 4755
    },
    {
      "epoch": 0.5721841567496093,
      "grad_norm": 0.6421749591827393,
      "learning_rate": 0.0004971989009825879,
      "loss": 1.0184,
      "step": 4760
    },
    {
      "epoch": 0.5727851905277077,
      "grad_norm": 0.5583392381668091,
      "learning_rate": 0.0004971917787255247,
      "loss": 1.3945,
      "step": 4765
    },
    {
      "epoch": 0.573386224305806,
      "grad_norm": 0.6780814528465271,
      "learning_rate": 0.0004971846474763394,
      "loss": 1.2617,
      "step": 4770
    },
    {
      "epoch": 0.5739872580839043,
      "grad_norm": 0.6801932454109192,
      "learning_rate": 0.0004971775072352914,
      "loss": 1.127,
      "step": 4775
    },
    {
      "epoch": 0.5745882918620027,
      "grad_norm": 0.7682610154151917,
      "learning_rate": 0.0004971703580026407,
      "loss": 1.093,
      "step": 4780
    },
    {
      "epoch": 0.575189325640101,
      "grad_norm": 0.5586374998092651,
      "learning_rate": 0.000497163199778647,
      "loss": 1.1852,
      "step": 4785
    },
    {
      "epoch": 0.5757903594181993,
      "grad_norm": 0.574213445186615,
      "learning_rate": 0.000497156032563571,
      "loss": 1.257,
      "step": 4790
    },
    {
      "epoch": 0.5763913931962976,
      "grad_norm": 0.4758005738258362,
      "learning_rate": 0.0004971488563576732,
      "loss": 0.7699,
      "step": 4795
    },
    {
      "epoch": 0.576992426974396,
      "grad_norm": 0.7012805938720703,
      "learning_rate": 0.0004971416711612149,
      "loss": 0.9437,
      "step": 4800
    },
    {
      "epoch": 0.576992426974396,
      "eval_loss": 2.10546875,
      "eval_model_preparation_time": 0.0053,
      "eval_runtime": 35.1867,
      "eval_samples_per_second": 4.547,
      "eval_steps_per_second": 1.137,
      "step": 4800
    },
    {
      "epoch": 0.5775934607524943,
      "grad_norm": 0.5702618360519409,
      "learning_rate": 0.0004971344769744572,
      "loss": 1.4156,
      "step": 4805
    },
    {
      "epoch": 0.5781944945305926,
      "grad_norm": 0.5781665444374084,
      "learning_rate": 0.000497127273797662,
      "loss": 1.1836,
      "step": 4810
    },
    {
      "epoch": 0.578795528308691,
      "grad_norm": 0.6150456070899963,
      "learning_rate": 0.0004971200616310914,
      "loss": 1.0434,
      "step": 4815
    },
    {
      "epoch": 0.5793965620867892,
      "grad_norm": 0.49010351300239563,
      "learning_rate": 0.0004971128404750075,
      "loss": 1.0312,
      "step": 4820
    },
    {
      "epoch": 0.5799975958648876,
      "grad_norm": 0.5255641341209412,
      "learning_rate": 0.000497105610329673,
      "loss": 1.1898,
      "step": 4825
    },
    {
      "epoch": 0.5805986296429859,
      "grad_norm": 0.5797913670539856,
      "learning_rate": 0.0004970983711953512,
      "loss": 0.9797,
      "step": 4830
    },
    {
      "epoch": 0.5811996634210843,
      "grad_norm": 0.3937300741672516,
      "learning_rate": 0.0004970911230723052,
      "loss": 1.1031,
      "step": 4835
    },
    {
      "epoch": 0.5818006971991826,
      "grad_norm": 0.5963497161865234,
      "learning_rate": 0.0004970838659607987,
      "loss": 1.0164,
      "step": 4840
    },
    {
      "epoch": 0.5824017309772809,
      "grad_norm": 0.7100831866264343,
      "learning_rate": 0.0004970765998610957,
      "loss": 0.943,
      "step": 4845
    },
    {
      "epoch": 0.5830027647553793,
      "grad_norm": 0.5534042119979858,
      "learning_rate": 0.0004970693247734606,
      "loss": 1.143,
      "step": 4850
    },
    {
      "epoch": 0.5836037985334775,
      "grad_norm": 0.8651660084724426,
      "learning_rate": 0.000497062040698158,
      "loss": 0.8562,
      "step": 4855
    },
    {
      "epoch": 0.5842048323115759,
      "grad_norm": 0.5462768077850342,
      "learning_rate": 0.0004970547476354528,
      "loss": 1.118,
      "step": 4860
    },
    {
      "epoch": 0.5848058660896742,
      "grad_norm": 0.5735900402069092,
      "learning_rate": 0.0004970474455856103,
      "loss": 1.0475,
      "step": 4865
    },
    {
      "epoch": 0.5854068998677726,
      "grad_norm": 0.7304772138595581,
      "learning_rate": 0.0004970401345488962,
      "loss": 0.9406,
      "step": 4870
    },
    {
      "epoch": 0.5860079336458709,
      "grad_norm": 0.8543664813041687,
      "learning_rate": 0.0004970328145255767,
      "loss": 1.1461,
      "step": 4875
    },
    {
      "epoch": 0.5866089674239692,
      "grad_norm": 0.47991877794265747,
      "learning_rate": 0.0004970254855159176,
      "loss": 1.2852,
      "step": 4880
    },
    {
      "epoch": 0.5872100012020676,
      "grad_norm": 0.5912867188453674,
      "learning_rate": 0.0004970181475201857,
      "loss": 1.0023,
      "step": 4885
    },
    {
      "epoch": 0.5878110349801658,
      "grad_norm": 0.598426878452301,
      "learning_rate": 0.0004970108005386482,
      "loss": 1.3953,
      "step": 4890
    },
    {
      "epoch": 0.5884120687582642,
      "grad_norm": 0.40192949771881104,
      "learning_rate": 0.0004970034445715719,
      "loss": 1.3156,
      "step": 4895
    },
    {
      "epoch": 0.5890131025363625,
      "grad_norm": 0.663162112236023,
      "learning_rate": 0.0004969960796192246,
      "loss": 1.0555,
      "step": 4900
    },
    {
      "epoch": 0.5896141363144609,
      "grad_norm": 0.5530306696891785,
      "learning_rate": 0.0004969887056818743,
      "loss": 1.3445,
      "step": 4905
    },
    {
      "epoch": 0.5902151700925592,
      "grad_norm": 0.5497463941574097,
      "learning_rate": 0.0004969813227597892,
      "loss": 1.3188,
      "step": 4910
    },
    {
      "epoch": 0.5908162038706576,
      "grad_norm": 0.40855851769447327,
      "learning_rate": 0.0004969739308532379,
      "loss": 1.1141,
      "step": 4915
    },
    {
      "epoch": 0.5914172376487559,
      "grad_norm": 0.6550401449203491,
      "learning_rate": 0.0004969665299624891,
      "loss": 1.0875,
      "step": 4920
    },
    {
      "epoch": 0.5920182714268541,
      "grad_norm": 0.6358505487442017,
      "learning_rate": 0.0004969591200878122,
      "loss": 1.4852,
      "step": 4925
    },
    {
      "epoch": 0.5926193052049525,
      "grad_norm": 0.45583033561706543,
      "learning_rate": 0.0004969517012294768,
      "loss": 1.1062,
      "step": 4930
    },
    {
      "epoch": 0.5932203389830508,
      "grad_norm": 0.6163007020950317,
      "learning_rate": 0.0004969442733877526,
      "loss": 1.1102,
      "step": 4935
    },
    {
      "epoch": 0.5938213727611492,
      "grad_norm": 0.7980563640594482,
      "learning_rate": 0.00049693683656291,
      "loss": 1.1547,
      "step": 4940
    },
    {
      "epoch": 0.5944224065392475,
      "grad_norm": 0.371162086725235,
      "learning_rate": 0.0004969293907552193,
      "loss": 1.2648,
      "step": 4945
    },
    {
      "epoch": 0.5950234403173459,
      "grad_norm": 0.4477928578853607,
      "learning_rate": 0.0004969219359649516,
      "loss": 1.193,
      "step": 4950
    },
    {
      "epoch": 0.5956244740954442,
      "grad_norm": 0.5350117683410645,
      "learning_rate": 0.0004969144721923779,
      "loss": 0.7547,
      "step": 4955
    },
    {
      "epoch": 0.5962255078735424,
      "grad_norm": 0.7265620231628418,
      "learning_rate": 0.0004969069994377697,
      "loss": 1.1359,
      "step": 4960
    },
    {
      "epoch": 0.5968265416516408,
      "grad_norm": 0.4857397973537445,
      "learning_rate": 0.0004968995177013991,
      "loss": 1.2906,
      "step": 4965
    },
    {
      "epoch": 0.5974275754297391,
      "grad_norm": 0.43424108624458313,
      "learning_rate": 0.000496892026983538,
      "loss": 1.1633,
      "step": 4970
    },
    {
      "epoch": 0.5980286092078375,
      "grad_norm": 0.5249903202056885,
      "learning_rate": 0.0004968845272844589,
      "loss": 0.8766,
      "step": 4975
    },
    {
      "epoch": 0.5986296429859358,
      "grad_norm": 0.591312825679779,
      "learning_rate": 0.0004968770186044347,
      "loss": 1.107,
      "step": 4980
    },
    {
      "epoch": 0.5992306767640342,
      "grad_norm": 0.5988960266113281,
      "learning_rate": 0.0004968695009437385,
      "loss": 1.0281,
      "step": 4985
    },
    {
      "epoch": 0.5998317105421325,
      "grad_norm": 0.5821179151535034,
      "learning_rate": 0.0004968619743026439,
      "loss": 1.0852,
      "step": 4990
    },
    {
      "epoch": 0.6004327443202307,
      "grad_norm": 0.5867944359779358,
      "learning_rate": 0.0004968544386814245,
      "loss": 1.1008,
      "step": 4995
    },
    {
      "epoch": 0.6010337780983291,
      "grad_norm": 0.461322158575058,
      "learning_rate": 0.0004968468940803546,
      "loss": 0.9305,
      "step": 5000
    },
    {
      "epoch": 0.6010337780983291,
      "eval_loss": 2.072265625,
      "eval_model_preparation_time": 0.0053,
      "eval_runtime": 35.2054,
      "eval_samples_per_second": 4.545,
      "eval_steps_per_second": 1.136,
      "step": 5000
    },
    {
      "epoch": 0.6016348118764274,
      "grad_norm": 0.47805655002593994,
      "learning_rate": 0.0004968393404997085,
      "loss": 0.8844,
      "step": 5005
    },
    {
      "epoch": 0.6022358456545258,
      "grad_norm": 0.6164557933807373,
      "learning_rate": 0.0004968317779397611,
      "loss": 1.0875,
      "step": 5010
    },
    {
      "epoch": 0.6028368794326241,
      "grad_norm": 0.6458466053009033,
      "learning_rate": 0.0004968242064007875,
      "loss": 0.8664,
      "step": 5015
    },
    {
      "epoch": 0.6034379132107225,
      "grad_norm": 0.5833812355995178,
      "learning_rate": 0.000496816625883063,
      "loss": 0.7797,
      "step": 5020
    },
    {
      "epoch": 0.6040389469888208,
      "grad_norm": 0.6610152721405029,
      "learning_rate": 0.0004968090363868634,
      "loss": 0.8711,
      "step": 5025
    },
    {
      "epoch": 0.604639980766919,
      "grad_norm": 0.5444918274879456,
      "learning_rate": 0.0004968014379124649,
      "loss": 1.1195,
      "step": 5030
    },
    {
      "epoch": 0.6052410145450174,
      "grad_norm": 0.7095007300376892,
      "learning_rate": 0.0004967938304601438,
      "loss": 1.2203,
      "step": 5035
    },
    {
      "epoch": 0.6058420483231157,
      "grad_norm": 0.5106276869773865,
      "learning_rate": 0.000496786214030177,
      "loss": 0.9242,
      "step": 5040
    },
    {
      "epoch": 0.6064430821012141,
      "grad_norm": 0.7541503310203552,
      "learning_rate": 0.0004967785886228414,
      "loss": 1.1953,
      "step": 5045
    },
    {
      "epoch": 0.6070441158793124,
      "grad_norm": 0.6698989868164062,
      "learning_rate": 0.0004967709542384142,
      "loss": 1.0977,
      "step": 5050
    },
    {
      "epoch": 0.6076451496574108,
      "grad_norm": 0.65958172082901,
      "learning_rate": 0.0004967633108771735,
      "loss": 1.1953,
      "step": 5055
    },
    {
      "epoch": 0.6082461834355091,
      "grad_norm": 0.4623226225376129,
      "learning_rate": 0.0004967556585393972,
      "loss": 1.0875,
      "step": 5060
    },
    {
      "epoch": 0.6088472172136074,
      "grad_norm": 0.5534173846244812,
      "learning_rate": 0.0004967479972253637,
      "loss": 0.8086,
      "step": 5065
    },
    {
      "epoch": 0.6094482509917057,
      "grad_norm": 0.5159333944320679,
      "learning_rate": 0.0004967403269353516,
      "loss": 1.0781,
      "step": 5070
    },
    {
      "epoch": 0.610049284769804,
      "grad_norm": 0.9555249810218811,
      "learning_rate": 0.00049673264766964,
      "loss": 1.3039,
      "step": 5075
    },
    {
      "epoch": 0.6106503185479024,
      "grad_norm": 0.5126526951789856,
      "learning_rate": 0.0004967249594285081,
      "loss": 1.3594,
      "step": 5080
    },
    {
      "epoch": 0.6112513523260007,
      "grad_norm": 0.4960355758666992,
      "learning_rate": 0.0004967172622122358,
      "loss": 1.0219,
      "step": 5085
    },
    {
      "epoch": 0.6118523861040991,
      "grad_norm": 0.6622259020805359,
      "learning_rate": 0.000496709556021103,
      "loss": 0.9633,
      "step": 5090
    },
    {
      "epoch": 0.6124534198821974,
      "grad_norm": 0.5355902910232544,
      "learning_rate": 0.0004967018408553901,
      "loss": 1.2641,
      "step": 5095
    },
    {
      "epoch": 0.6130544536602958,
      "grad_norm": 0.5535646080970764,
      "learning_rate": 0.0004966941167153776,
      "loss": 1.125,
      "step": 5100
    },
    {
      "epoch": 0.613655487438394,
      "grad_norm": 0.40969082713127136,
      "learning_rate": 0.0004966863836013465,
      "loss": 1.5328,
      "step": 5105
    },
    {
      "epoch": 0.6142565212164923,
      "grad_norm": 0.7018454074859619,
      "learning_rate": 0.0004966786415135783,
      "loss": 1.009,
      "step": 5110
    },
    {
      "epoch": 0.6148575549945907,
      "grad_norm": 0.7394018769264221,
      "learning_rate": 0.0004966708904523546,
      "loss": 1.4625,
      "step": 5115
    },
    {
      "epoch": 0.615458588772689,
      "grad_norm": 0.43013033270835876,
      "learning_rate": 0.0004966631304179571,
      "loss": 0.7812,
      "step": 5120
    },
    {
      "epoch": 0.6160596225507874,
      "grad_norm": 0.6355203986167908,
      "learning_rate": 0.0004966553614106684,
      "loss": 0.8352,
      "step": 5125
    },
    {
      "epoch": 0.6166606563288857,
      "grad_norm": 0.4331319332122803,
      "learning_rate": 0.0004966475834307708,
      "loss": 1.0063,
      "step": 5130
    },
    {
      "epoch": 0.6172616901069841,
      "grad_norm": 0.47516930103302,
      "learning_rate": 0.0004966397964785475,
      "loss": 1.1465,
      "step": 5135
    },
    {
      "epoch": 0.6178627238850823,
      "grad_norm": 0.46409985423088074,
      "learning_rate": 0.0004966320005542817,
      "loss": 1.3172,
      "step": 5140
    },
    {
      "epoch": 0.6184637576631806,
      "grad_norm": 0.5062604546546936,
      "learning_rate": 0.000496624195658257,
      "loss": 1.0352,
      "step": 5145
    },
    {
      "epoch": 0.619064791441279,
      "grad_norm": 0.5797644257545471,
      "learning_rate": 0.0004966163817907573,
      "loss": 1.043,
      "step": 5150
    },
    {
      "epoch": 0.6196658252193773,
      "grad_norm": 0.6194161176681519,
      "learning_rate": 0.0004966085589520668,
      "loss": 0.9633,
      "step": 5155
    },
    {
      "epoch": 0.6202668589974757,
      "grad_norm": 0.570642352104187,
      "learning_rate": 0.0004966007271424701,
      "loss": 1.6047,
      "step": 5160
    },
    {
      "epoch": 0.620867892775574,
      "grad_norm": 0.6968045830726624,
      "learning_rate": 0.0004965928863622522,
      "loss": 1.2914,
      "step": 5165
    },
    {
      "epoch": 0.6214689265536724,
      "grad_norm": 0.4688586890697479,
      "learning_rate": 0.0004965850366116982,
      "loss": 1.4031,
      "step": 5170
    },
    {
      "epoch": 0.6220699603317706,
      "grad_norm": 0.6257545351982117,
      "learning_rate": 0.0004965771778910936,
      "loss": 1.1641,
      "step": 5175
    },
    {
      "epoch": 0.6226709941098689,
      "grad_norm": 0.5326806902885437,
      "learning_rate": 0.0004965693102007245,
      "loss": 0.973,
      "step": 5180
    },
    {
      "epoch": 0.6232720278879673,
      "grad_norm": 0.5571873188018799,
      "learning_rate": 0.0004965614335408769,
      "loss": 0.9023,
      "step": 5185
    },
    {
      "epoch": 0.6238730616660656,
      "grad_norm": 0.67780601978302,
      "learning_rate": 0.0004965535479118374,
      "loss": 1.1531,
      "step": 5190
    },
    {
      "epoch": 0.624474095444164,
      "grad_norm": 0.5032584071159363,
      "learning_rate": 0.0004965456533138928,
      "loss": 1.3109,
      "step": 5195
    },
    {
      "epoch": 0.6250751292222623,
      "grad_norm": 0.5984988808631897,
      "learning_rate": 0.0004965377497473304,
      "loss": 1.3664,
      "step": 5200
    },
    {
      "epoch": 0.6250751292222623,
      "eval_loss": 1.9845702648162842,
      "eval_model_preparation_time": 0.0053,
      "eval_runtime": 35.2079,
      "eval_samples_per_second": 4.544,
      "eval_steps_per_second": 1.136,
      "step": 5200
    },
    {
      "epoch": 0.6256761630003607,
      "grad_norm": 0.6601026654243469,
      "learning_rate": 0.0004965298372124376,
      "loss": 1.0969,
      "step": 5205
    },
    {
      "epoch": 0.6262771967784589,
      "grad_norm": 0.6069896817207336,
      "learning_rate": 0.0004965219157095024,
      "loss": 0.7707,
      "step": 5210
    },
    {
      "epoch": 0.6268782305565572,
      "grad_norm": 0.5757645964622498,
      "learning_rate": 0.0004965139852388127,
      "loss": 0.9836,
      "step": 5215
    },
    {
      "epoch": 0.6274792643346556,
      "grad_norm": 0.658743143081665,
      "learning_rate": 0.0004965060458006573,
      "loss": 0.7297,
      "step": 5220
    },
    {
      "epoch": 0.6280802981127539,
      "grad_norm": 0.6893699765205383,
      "learning_rate": 0.0004964980973953247,
      "loss": 1.0586,
      "step": 5225
    },
    {
      "epoch": 0.6286813318908523,
      "grad_norm": 0.6678071022033691,
      "learning_rate": 0.0004964901400231043,
      "loss": 0.9922,
      "step": 5230
    },
    {
      "epoch": 0.6292823656689506,
      "grad_norm": 0.8199008703231812,
      "learning_rate": 0.0004964821736842854,
      "loss": 1.0664,
      "step": 5235
    },
    {
      "epoch": 0.629883399447049,
      "grad_norm": 0.516659140586853,
      "learning_rate": 0.0004964741983791578,
      "loss": 1.0359,
      "step": 5240
    },
    {
      "epoch": 0.6304844332251472,
      "grad_norm": 0.5256388187408447,
      "learning_rate": 0.0004964662141080117,
      "loss": 0.8906,
      "step": 5245
    },
    {
      "epoch": 0.6310854670032456,
      "grad_norm": 0.6450999975204468,
      "learning_rate": 0.0004964582208711375,
      "loss": 0.8363,
      "step": 5250
    },
    {
      "epoch": 0.6316865007813439,
      "grad_norm": 0.7471349239349365,
      "learning_rate": 0.000496450218668826,
      "loss": 1.198,
      "step": 5255
    },
    {
      "epoch": 0.6322875345594422,
      "grad_norm": 0.5109617114067078,
      "learning_rate": 0.0004964422075013682,
      "loss": 0.9227,
      "step": 5260
    },
    {
      "epoch": 0.6328885683375406,
      "grad_norm": 0.48837974667549133,
      "learning_rate": 0.0004964341873690557,
      "loss": 0.9469,
      "step": 5265
    },
    {
      "epoch": 0.6334896021156389,
      "grad_norm": 0.7013422250747681,
      "learning_rate": 0.0004964261582721801,
      "loss": 1.1594,
      "step": 5270
    },
    {
      "epoch": 0.6340906358937373,
      "grad_norm": 0.3042369484901428,
      "learning_rate": 0.0004964181202110335,
      "loss": 1.3625,
      "step": 5275
    },
    {
      "epoch": 0.6346916696718355,
      "grad_norm": 0.8467724919319153,
      "learning_rate": 0.0004964100731859084,
      "loss": 0.9328,
      "step": 5280
    },
    {
      "epoch": 0.6352927034499339,
      "grad_norm": 0.7622195482254028,
      "learning_rate": 0.0004964020171970974,
      "loss": 1.0672,
      "step": 5285
    },
    {
      "epoch": 0.6358937372280322,
      "grad_norm": 0.5378862023353577,
      "learning_rate": 0.0004963939522448936,
      "loss": 1.0277,
      "step": 5290
    },
    {
      "epoch": 0.6364947710061305,
      "grad_norm": 0.5819463133811951,
      "learning_rate": 0.0004963858783295905,
      "loss": 1.1039,
      "step": 5295
    },
    {
      "epoch": 0.6370958047842289,
      "grad_norm": 0.5175995826721191,
      "learning_rate": 0.0004963777954514816,
      "loss": 1.3875,
      "step": 5300
    },
    {
      "epoch": 0.6376968385623272,
      "grad_norm": 0.6846615076065063,
      "learning_rate": 0.000496369703610861,
      "loss": 0.8723,
      "step": 5305
    },
    {
      "epoch": 0.6382978723404256,
      "grad_norm": 0.6325473189353943,
      "learning_rate": 0.000496361602808023,
      "loss": 1.2469,
      "step": 5310
    },
    {
      "epoch": 0.6388989061185238,
      "grad_norm": 0.49080929160118103,
      "learning_rate": 0.0004963534930432625,
      "loss": 1.2977,
      "step": 5315
    },
    {
      "epoch": 0.6394999398966222,
      "grad_norm": 0.4832301437854767,
      "learning_rate": 0.0004963453743168743,
      "loss": 1.1719,
      "step": 5320
    },
    {
      "epoch": 0.6401009736747205,
      "grad_norm": 0.7145053744316101,
      "learning_rate": 0.000496337246629154,
      "loss": 1.2859,
      "step": 5325
    },
    {
      "epoch": 0.6407020074528188,
      "grad_norm": 0.5795300602912903,
      "learning_rate": 0.0004963291099803969,
      "loss": 0.9617,
      "step": 5330
    },
    {
      "epoch": 0.6413030412309172,
      "grad_norm": 0.9390627145767212,
      "learning_rate": 0.0004963209643708991,
      "loss": 1.1703,
      "step": 5335
    },
    {
      "epoch": 0.6419040750090155,
      "grad_norm": 0.7437942624092102,
      "learning_rate": 0.000496312809800957,
      "loss": 1.0711,
      "step": 5340
    },
    {
      "epoch": 0.6425051087871139,
      "grad_norm": 0.5893829464912415,
      "learning_rate": 0.0004963046462708673,
      "loss": 1.2203,
      "step": 5345
    },
    {
      "epoch": 0.6431061425652121,
      "grad_norm": 0.430411696434021,
      "learning_rate": 0.0004962964737809268,
      "loss": 1.068,
      "step": 5350
    },
    {
      "epoch": 0.6437071763433105,
      "grad_norm": 0.4798862040042877,
      "learning_rate": 0.0004962882923314329,
      "loss": 1.0195,
      "step": 5355
    },
    {
      "epoch": 0.6443082101214088,
      "grad_norm": 0.7981201410293579,
      "learning_rate": 0.0004962801019226833,
      "loss": 0.9211,
      "step": 5360
    },
    {
      "epoch": 0.6449092438995071,
      "grad_norm": 0.4801620841026306,
      "learning_rate": 0.0004962719025549757,
      "loss": 1.2961,
      "step": 5365
    },
    {
      "epoch": 0.6455102776776055,
      "grad_norm": 0.4748445451259613,
      "learning_rate": 0.0004962636942286086,
      "loss": 1.207,
      "step": 5370
    },
    {
      "epoch": 0.6461113114557038,
      "grad_norm": 0.5984495282173157,
      "learning_rate": 0.0004962554769438802,
      "loss": 0.9734,
      "step": 5375
    },
    {
      "epoch": 0.6467123452338022,
      "grad_norm": 0.6702393889427185,
      "learning_rate": 0.0004962472507010901,
      "loss": 1.0977,
      "step": 5380
    },
    {
      "epoch": 0.6473133790119004,
      "grad_norm": 0.5429012775421143,
      "learning_rate": 0.0004962390155005369,
      "loss": 1.0641,
      "step": 5385
    },
    {
      "epoch": 0.6479144127899988,
      "grad_norm": 0.39091694355010986,
      "learning_rate": 0.0004962307713425206,
      "loss": 0.9906,
      "step": 5390
    },
    {
      "epoch": 0.6485154465680971,
      "grad_norm": 0.4490283131599426,
      "learning_rate": 0.0004962225182273409,
      "loss": 1.1297,
      "step": 5395
    },
    {
      "epoch": 0.6491164803461954,
      "grad_norm": 0.6250773072242737,
      "learning_rate": 0.0004962142561552981,
      "loss": 0.8102,
      "step": 5400
    },
    {
      "epoch": 0.6491164803461954,
      "eval_loss": 2.007031202316284,
      "eval_model_preparation_time": 0.0053,
      "eval_runtime": 35.2151,
      "eval_samples_per_second": 4.544,
      "eval_steps_per_second": 1.136,
      "step": 5400
    },
    {
      "epoch": 0.6497175141242938,
      "grad_norm": 0.4846701920032501,
      "learning_rate": 0.0004962059851266926,
      "loss": 1.4289,
      "step": 5405
    },
    {
      "epoch": 0.6503185479023921,
      "grad_norm": 0.5710358023643494,
      "learning_rate": 0.0004961977051418253,
      "loss": 1.0961,
      "step": 5410
    },
    {
      "epoch": 0.6509195816804905,
      "grad_norm": 0.46397438645362854,
      "learning_rate": 0.0004961894162009977,
      "loss": 0.9297,
      "step": 5415
    },
    {
      "epoch": 0.6515206154585887,
      "grad_norm": 0.5259897708892822,
      "learning_rate": 0.0004961811183045111,
      "loss": 1.0891,
      "step": 5420
    },
    {
      "epoch": 0.6521216492366871,
      "grad_norm": 0.44706469774246216,
      "learning_rate": 0.0004961728114526672,
      "loss": 1.175,
      "step": 5425
    },
    {
      "epoch": 0.6527226830147854,
      "grad_norm": 0.6101322174072266,
      "learning_rate": 0.0004961644956457685,
      "loss": 1.143,
      "step": 5430
    },
    {
      "epoch": 0.6533237167928838,
      "grad_norm": 0.544548511505127,
      "learning_rate": 0.0004961561708841173,
      "loss": 0.9246,
      "step": 5435
    },
    {
      "epoch": 0.6539247505709821,
      "grad_norm": 0.7637187242507935,
      "learning_rate": 0.0004961478371680165,
      "loss": 1.3031,
      "step": 5440
    },
    {
      "epoch": 0.6545257843490804,
      "grad_norm": 0.5239999294281006,
      "learning_rate": 0.0004961394944977692,
      "loss": 1.0242,
      "step": 5445
    },
    {
      "epoch": 0.6551268181271788,
      "grad_norm": 0.6028966307640076,
      "learning_rate": 0.000496131142873679,
      "loss": 1.2961,
      "step": 5450
    },
    {
      "epoch": 0.655727851905277,
      "grad_norm": 0.9021722674369812,
      "learning_rate": 0.0004961227822960495,
      "loss": 1.0543,
      "step": 5455
    },
    {
      "epoch": 0.6563288856833754,
      "grad_norm": 0.8960506916046143,
      "learning_rate": 0.0004961144127651851,
      "loss": 1.4555,
      "step": 5460
    },
    {
      "epoch": 0.6569299194614737,
      "grad_norm": 0.6960546374320984,
      "learning_rate": 0.0004961060342813901,
      "loss": 0.7801,
      "step": 5465
    },
    {
      "epoch": 0.6575309532395721,
      "grad_norm": 0.7192292213439941,
      "learning_rate": 0.0004960976468449692,
      "loss": 0.9641,
      "step": 5470
    },
    {
      "epoch": 0.6581319870176704,
      "grad_norm": 0.605060875415802,
      "learning_rate": 0.0004960892504562277,
      "loss": 1.6898,
      "step": 5475
    },
    {
      "epoch": 0.6587330207957687,
      "grad_norm": 0.5543851256370544,
      "learning_rate": 0.000496080845115471,
      "loss": 1.4141,
      "step": 5480
    },
    {
      "epoch": 0.6593340545738671,
      "grad_norm": 0.6614383459091187,
      "learning_rate": 0.0004960724308230047,
      "loss": 1.0016,
      "step": 5485
    },
    {
      "epoch": 0.6599350883519653,
      "grad_norm": 0.5792709589004517,
      "learning_rate": 0.0004960640075791351,
      "loss": 1.2906,
      "step": 5490
    },
    {
      "epoch": 0.6605361221300637,
      "grad_norm": 0.509619951248169,
      "learning_rate": 0.0004960555753841685,
      "loss": 0.9914,
      "step": 5495
    },
    {
      "epoch": 0.661137155908162,
      "grad_norm": 0.6596083045005798,
      "learning_rate": 0.0004960471342384116,
      "loss": 0.8969,
      "step": 5500
    },
    {
      "epoch": 0.6617381896862604,
      "grad_norm": 0.43772363662719727,
      "learning_rate": 0.0004960386841421716,
      "loss": 1.3281,
      "step": 5505
    },
    {
      "epoch": 0.6623392234643587,
      "grad_norm": 0.551668643951416,
      "learning_rate": 0.0004960302250957558,
      "loss": 0.9844,
      "step": 5510
    },
    {
      "epoch": 0.662940257242457,
      "grad_norm": 0.5039249658584595,
      "learning_rate": 0.0004960217570994719,
      "loss": 1.1273,
      "step": 5515
    },
    {
      "epoch": 0.6635412910205554,
      "grad_norm": 0.48881009221076965,
      "learning_rate": 0.0004960132801536281,
      "loss": 1.3953,
      "step": 5520
    },
    {
      "epoch": 0.6641423247986536,
      "grad_norm": 0.7355161905288696,
      "learning_rate": 0.0004960047942585324,
      "loss": 0.868,
      "step": 5525
    },
    {
      "epoch": 0.664743358576752,
      "grad_norm": 0.38211506605148315,
      "learning_rate": 0.0004959962994144939,
      "loss": 1.1547,
      "step": 5530
    },
    {
      "epoch": 0.6653443923548503,
      "grad_norm": 0.5978872776031494,
      "learning_rate": 0.0004959877956218213,
      "loss": 1.1145,
      "step": 5535
    },
    {
      "epoch": 0.6659454261329487,
      "grad_norm": 0.6254252791404724,
      "learning_rate": 0.0004959792828808241,
      "loss": 1.1023,
      "step": 5540
    },
    {
      "epoch": 0.666546459911047,
      "grad_norm": 0.5093739032745361,
      "learning_rate": 0.0004959707611918121,
      "loss": 1.1742,
      "step": 5545
    },
    {
      "epoch": 0.6671474936891453,
      "grad_norm": 0.5325815677642822,
      "learning_rate": 0.0004959622305550951,
      "loss": 1.1641,
      "step": 5550
    },
    {
      "epoch": 0.6677485274672437,
      "grad_norm": 0.5326328873634338,
      "learning_rate": 0.0004959536909709834,
      "loss": 0.8969,
      "step": 5555
    },
    {
      "epoch": 0.668349561245342,
      "grad_norm": 0.45378610491752625,
      "learning_rate": 0.0004959451424397879,
      "loss": 1.0051,
      "step": 5560
    },
    {
      "epoch": 0.6689505950234403,
      "grad_norm": 0.6395484209060669,
      "learning_rate": 0.0004959365849618192,
      "loss": 1.0594,
      "step": 5565
    },
    {
      "epoch": 0.6695516288015386,
      "grad_norm": 0.6629273891448975,
      "learning_rate": 0.0004959280185373888,
      "loss": 1.182,
      "step": 5570
    },
    {
      "epoch": 0.670152662579637,
      "grad_norm": 0.6001206040382385,
      "learning_rate": 0.0004959194431668084,
      "loss": 0.825,
      "step": 5575
    },
    {
      "epoch": 0.6707536963577353,
      "grad_norm": 0.6217541098594666,
      "learning_rate": 0.0004959108588503898,
      "loss": 1.0355,
      "step": 5580
    },
    {
      "epoch": 0.6713547301358337,
      "grad_norm": 0.5813589096069336,
      "learning_rate": 0.0004959022655884453,
      "loss": 0.9965,
      "step": 5585
    },
    {
      "epoch": 0.671955763913932,
      "grad_norm": 0.4870590269565582,
      "learning_rate": 0.0004958936633812876,
      "loss": 1.1906,
      "step": 5590
    },
    {
      "epoch": 0.6725567976920303,
      "grad_norm": 0.46450790762901306,
      "learning_rate": 0.0004958850522292295,
      "loss": 0.9266,
      "step": 5595
    },
    {
      "epoch": 0.6731578314701286,
      "grad_norm": 0.7803342938423157,
      "learning_rate": 0.0004958764321325843,
      "loss": 1.1867,
      "step": 5600
    },
    {
      "epoch": 0.6731578314701286,
      "eval_loss": 1.960351586341858,
      "eval_model_preparation_time": 0.0053,
      "eval_runtime": 35.1953,
      "eval_samples_per_second": 4.546,
      "eval_steps_per_second": 1.137,
      "step": 5600
    },
    {
      "epoch": 0.6737588652482269,
      "grad_norm": 0.39002323150634766,
      "learning_rate": 0.0004958678030916655,
      "loss": 0.9242,
      "step": 5605
    },
    {
      "epoch": 0.6743598990263253,
      "grad_norm": 0.6863328814506531,
      "learning_rate": 0.0004958591651067872,
      "loss": 0.9203,
      "step": 5610
    },
    {
      "epoch": 0.6749609328044236,
      "grad_norm": 0.5728829503059387,
      "learning_rate": 0.0004958505181782635,
      "loss": 1.1797,
      "step": 5615
    },
    {
      "epoch": 0.675561966582522,
      "grad_norm": 0.5186361074447632,
      "learning_rate": 0.0004958418623064088,
      "loss": 0.7125,
      "step": 5620
    },
    {
      "epoch": 0.6761630003606203,
      "grad_norm": 0.5343161225318909,
      "learning_rate": 0.0004958331974915382,
      "loss": 1.0008,
      "step": 5625
    },
    {
      "epoch": 0.6767640341387186,
      "grad_norm": 0.6222845315933228,
      "learning_rate": 0.0004958245237339669,
      "loss": 1.0906,
      "step": 5630
    },
    {
      "epoch": 0.6773650679168169,
      "grad_norm": 0.5673485398292542,
      "learning_rate": 0.0004958158410340103,
      "loss": 1.0516,
      "step": 5635
    },
    {
      "epoch": 0.6779661016949152,
      "grad_norm": 0.6129093170166016,
      "learning_rate": 0.0004958071493919842,
      "loss": 0.932,
      "step": 5640
    },
    {
      "epoch": 0.6785671354730136,
      "grad_norm": 0.43740183115005493,
      "learning_rate": 0.0004957984488082049,
      "loss": 1.4109,
      "step": 5645
    },
    {
      "epoch": 0.6791681692511119,
      "grad_norm": 0.632262647151947,
      "learning_rate": 0.0004957897392829889,
      "loss": 1.3547,
      "step": 5650
    },
    {
      "epoch": 0.6797692030292103,
      "grad_norm": 0.5351587533950806,
      "learning_rate": 0.0004957810208166531,
      "loss": 1.1938,
      "step": 5655
    },
    {
      "epoch": 0.6803702368073086,
      "grad_norm": 0.6918035745620728,
      "learning_rate": 0.0004957722934095145,
      "loss": 0.9172,
      "step": 5660
    },
    {
      "epoch": 0.6809712705854069,
      "grad_norm": 0.5031920671463013,
      "learning_rate": 0.0004957635570618906,
      "loss": 0.9242,
      "step": 5665
    },
    {
      "epoch": 0.6815723043635052,
      "grad_norm": 0.8015581965446472,
      "learning_rate": 0.0004957548117740993,
      "loss": 0.884,
      "step": 5670
    },
    {
      "epoch": 0.6821733381416035,
      "grad_norm": 0.5075511336326599,
      "learning_rate": 0.0004957460575464586,
      "loss": 1.0656,
      "step": 5675
    },
    {
      "epoch": 0.6827743719197019,
      "grad_norm": 0.43553194403648376,
      "learning_rate": 0.000495737294379287,
      "loss": 1.1762,
      "step": 5680
    },
    {
      "epoch": 0.6833754056978002,
      "grad_norm": 0.48835834860801697,
      "learning_rate": 0.0004957285222729034,
      "loss": 1.0547,
      "step": 5685
    },
    {
      "epoch": 0.6839764394758986,
      "grad_norm": 0.5027529001235962,
      "learning_rate": 0.0004957197412276267,
      "loss": 1.1742,
      "step": 5690
    },
    {
      "epoch": 0.6845774732539969,
      "grad_norm": 0.581287682056427,
      "learning_rate": 0.0004957109512437766,
      "loss": 0.9242,
      "step": 5695
    },
    {
      "epoch": 0.6851785070320952,
      "grad_norm": 0.5454129576683044,
      "learning_rate": 0.0004957021523216725,
      "loss": 1.1492,
      "step": 5700
    },
    {
      "epoch": 0.6857795408101935,
      "grad_norm": 0.5323740839958191,
      "learning_rate": 0.0004956933444616347,
      "loss": 0.8586,
      "step": 5705
    },
    {
      "epoch": 0.6863805745882918,
      "grad_norm": 0.8149203658103943,
      "learning_rate": 0.0004956845276639836,
      "loss": 1.15,
      "step": 5710
    },
    {
      "epoch": 0.6869816083663902,
      "grad_norm": 0.6589764952659607,
      "learning_rate": 0.00049567570192904,
      "loss": 1.132,
      "step": 5715
    },
    {
      "epoch": 0.6875826421444885,
      "grad_norm": 0.5598886609077454,
      "learning_rate": 0.0004956668672571247,
      "loss": 0.9125,
      "step": 5720
    },
    {
      "epoch": 0.6881836759225869,
      "grad_norm": 0.4387410581111908,
      "learning_rate": 0.0004956580236485593,
      "loss": 0.8129,
      "step": 5725
    },
    {
      "epoch": 0.6887847097006852,
      "grad_norm": 1.346904993057251,
      "learning_rate": 0.0004956491711036654,
      "loss": 1.3328,
      "step": 5730
    },
    {
      "epoch": 0.6893857434787835,
      "grad_norm": 0.597332775592804,
      "learning_rate": 0.0004956403096227651,
      "loss": 1.2812,
      "step": 5735
    },
    {
      "epoch": 0.6899867772568818,
      "grad_norm": 0.4443131685256958,
      "learning_rate": 0.0004956314392061808,
      "loss": 0.9059,
      "step": 5740
    },
    {
      "epoch": 0.6905878110349801,
      "grad_norm": 0.547560453414917,
      "learning_rate": 0.000495622559854235,
      "loss": 1.1344,
      "step": 5745
    },
    {
      "epoch": 0.6911888448130785,
      "grad_norm": 0.5721062421798706,
      "learning_rate": 0.0004956136715672509,
      "loss": 1.1805,
      "step": 5750
    },
    {
      "epoch": 0.6917898785911768,
      "grad_norm": 0.5476487278938293,
      "learning_rate": 0.0004956047743455517,
      "loss": 1.2406,
      "step": 5755
    },
    {
      "epoch": 0.6923909123692752,
      "grad_norm": 0.5772802829742432,
      "learning_rate": 0.0004955958681894611,
      "loss": 1.1797,
      "step": 5760
    },
    {
      "epoch": 0.6929919461473735,
      "grad_norm": 0.5622345805168152,
      "learning_rate": 0.000495586953099303,
      "loss": 1.0258,
      "step": 5765
    },
    {
      "epoch": 0.6935929799254719,
      "grad_norm": 0.3974035680294037,
      "learning_rate": 0.0004955780290754018,
      "loss": 0.9348,
      "step": 5770
    },
    {
      "epoch": 0.6941940137035701,
      "grad_norm": 0.7566075921058655,
      "learning_rate": 0.0004955690961180822,
      "loss": 1.2188,
      "step": 5775
    },
    {
      "epoch": 0.6947950474816684,
      "grad_norm": 0.5623620748519897,
      "learning_rate": 0.0004955601542276691,
      "loss": 1.0195,
      "step": 5780
    },
    {
      "epoch": 0.6953960812597668,
      "grad_norm": 0.5877131819725037,
      "learning_rate": 0.0004955512034044876,
      "loss": 0.7508,
      "step": 5785
    },
    {
      "epoch": 0.6959971150378651,
      "grad_norm": 0.35780635476112366,
      "learning_rate": 0.0004955422436488635,
      "loss": 1.0418,
      "step": 5790
    },
    {
      "epoch": 0.6965981488159635,
      "grad_norm": 0.6338667869567871,
      "learning_rate": 0.0004955332749611227,
      "loss": 0.866,
      "step": 5795
    },
    {
      "epoch": 0.6971991825940618,
      "grad_norm": 0.6346995830535889,
      "learning_rate": 0.0004955242973415915,
      "loss": 0.9914,
      "step": 5800
    },
    {
      "epoch": 0.6971991825940618,
      "eval_loss": 1.979101538658142,
      "eval_model_preparation_time": 0.0053,
      "eval_runtime": 35.1923,
      "eval_samples_per_second": 4.546,
      "eval_steps_per_second": 1.137,
      "step": 5800
    },
    {
      "epoch": 0.6978002163721602,
      "grad_norm": 0.6727893948554993,
      "learning_rate": 0.0004955153107905964,
      "loss": 1.1859,
      "step": 5805
    },
    {
      "epoch": 0.6984012501502584,
      "grad_norm": 0.4857117235660553,
      "learning_rate": 0.0004955063153084642,
      "loss": 1.5766,
      "step": 5810
    },
    {
      "epoch": 0.6990022839283567,
      "grad_norm": 0.6355534791946411,
      "learning_rate": 0.0004954973108955223,
      "loss": 1.1781,
      "step": 5815
    },
    {
      "epoch": 0.6996033177064551,
      "grad_norm": 0.5181488990783691,
      "learning_rate": 0.0004954882975520983,
      "loss": 1.2094,
      "step": 5820
    },
    {
      "epoch": 0.7002043514845534,
      "grad_norm": 0.3698882460594177,
      "learning_rate": 0.0004954792752785198,
      "loss": 1.0766,
      "step": 5825
    },
    {
      "epoch": 0.7008053852626518,
      "grad_norm": 0.6260622143745422,
      "learning_rate": 0.0004954702440751153,
      "loss": 0.9172,
      "step": 5830
    },
    {
      "epoch": 0.7014064190407501,
      "grad_norm": 0.5361654162406921,
      "learning_rate": 0.0004954612039422132,
      "loss": 1.0484,
      "step": 5835
    },
    {
      "epoch": 0.7020074528188485,
      "grad_norm": 0.4724903404712677,
      "learning_rate": 0.0004954521548801424,
      "loss": 1.225,
      "step": 5840
    },
    {
      "epoch": 0.7026084865969467,
      "grad_norm": 0.5695177912712097,
      "learning_rate": 0.000495443096889232,
      "loss": 0.9734,
      "step": 5845
    },
    {
      "epoch": 0.703209520375045,
      "grad_norm": 0.8220880627632141,
      "learning_rate": 0.0004954340299698116,
      "loss": 0.9305,
      "step": 5850
    },
    {
      "epoch": 0.7038105541531434,
      "grad_norm": 0.46409645676612854,
      "learning_rate": 0.000495424954122211,
      "loss": 1.5328,
      "step": 5855
    },
    {
      "epoch": 0.7044115879312417,
      "grad_norm": 0.5940005779266357,
      "learning_rate": 0.0004954158693467603,
      "loss": 0.9586,
      "step": 5860
    },
    {
      "epoch": 0.7050126217093401,
      "grad_norm": 0.6850906014442444,
      "learning_rate": 0.00049540677564379,
      "loss": 1.2406,
      "step": 5865
    },
    {
      "epoch": 0.7056136554874384,
      "grad_norm": 0.6025176048278809,
      "learning_rate": 0.0004953976730136309,
      "loss": 1.0621,
      "step": 5870
    },
    {
      "epoch": 0.7062146892655368,
      "grad_norm": 0.5903462767601013,
      "learning_rate": 0.0004953885614566142,
      "loss": 0.8078,
      "step": 5875
    },
    {
      "epoch": 0.706815723043635,
      "grad_norm": 0.8402755856513977,
      "learning_rate": 0.0004953794409730713,
      "loss": 0.8309,
      "step": 5880
    },
    {
      "epoch": 0.7074167568217333,
      "grad_norm": 0.525725245475769,
      "learning_rate": 0.0004953703115633339,
      "loss": 1.1781,
      "step": 5885
    },
    {
      "epoch": 0.7080177905998317,
      "grad_norm": 0.5770952701568604,
      "learning_rate": 0.0004953611732277342,
      "loss": 1.3102,
      "step": 5890
    },
    {
      "epoch": 0.70861882437793,
      "grad_norm": 0.5640763640403748,
      "learning_rate": 0.0004953520259666046,
      "loss": 1.1016,
      "step": 5895
    },
    {
      "epoch": 0.7092198581560284,
      "grad_norm": 0.6585139036178589,
      "learning_rate": 0.0004953428697802778,
      "loss": 0.9719,
      "step": 5900
    },
    {
      "epoch": 0.7098208919341267,
      "grad_norm": 0.8663554191589355,
      "learning_rate": 0.0004953337046690871,
      "loss": 0.9508,
      "step": 5905
    },
    {
      "epoch": 0.7104219257122251,
      "grad_norm": 0.7086400985717773,
      "learning_rate": 0.0004953245306333656,
      "loss": 0.8703,
      "step": 5910
    },
    {
      "epoch": 0.7110229594903233,
      "grad_norm": 0.36062702536582947,
      "learning_rate": 0.0004953153476734472,
      "loss": 0.8773,
      "step": 5915
    },
    {
      "epoch": 0.7116239932684217,
      "grad_norm": 0.5696423053741455,
      "learning_rate": 0.0004953061557896658,
      "loss": 1.15,
      "step": 5920
    },
    {
      "epoch": 0.71222502704652,
      "grad_norm": 0.3699018955230713,
      "learning_rate": 0.000495296954982356,
      "loss": 1.0984,
      "step": 5925
    },
    {
      "epoch": 0.7128260608246183,
      "grad_norm": 0.5031812787055969,
      "learning_rate": 0.0004952877452518523,
      "loss": 0.802,
      "step": 5930
    },
    {
      "epoch": 0.7134270946027167,
      "grad_norm": 0.5342919230461121,
      "learning_rate": 0.0004952785265984898,
      "loss": 1.0727,
      "step": 5935
    },
    {
      "epoch": 0.714028128380815,
      "grad_norm": 0.7044137120246887,
      "learning_rate": 0.0004952692990226039,
      "loss": 0.7984,
      "step": 5940
    },
    {
      "epoch": 0.7146291621589134,
      "grad_norm": 0.5580726861953735,
      "learning_rate": 0.0004952600625245301,
      "loss": 1.1875,
      "step": 5945
    },
    {
      "epoch": 0.7152301959370116,
      "grad_norm": 0.5382014513015747,
      "learning_rate": 0.0004952508171046046,
      "loss": 0.9633,
      "step": 5950
    },
    {
      "epoch": 0.71583122971511,
      "grad_norm": 0.6365616321563721,
      "learning_rate": 0.0004952415627631636,
      "loss": 1.0867,
      "step": 5955
    },
    {
      "epoch": 0.7164322634932083,
      "grad_norm": 0.572903573513031,
      "learning_rate": 0.0004952322995005438,
      "loss": 0.9461,
      "step": 5960
    },
    {
      "epoch": 0.7170332972713066,
      "grad_norm": 0.5480600595474243,
      "learning_rate": 0.0004952230273170822,
      "loss": 1.0547,
      "step": 5965
    },
    {
      "epoch": 0.717634331049405,
      "grad_norm": 0.6455062031745911,
      "learning_rate": 0.000495213746213116,
      "loss": 0.8516,
      "step": 5970
    },
    {
      "epoch": 0.7182353648275033,
      "grad_norm": 0.4712236225605011,
      "learning_rate": 0.0004952044561889829,
      "loss": 0.8742,
      "step": 5975
    },
    {
      "epoch": 0.7188363986056017,
      "grad_norm": 0.4829672873020172,
      "learning_rate": 0.0004951951572450207,
      "loss": 0.9465,
      "step": 5980
    },
    {
      "epoch": 0.7194374323837,
      "grad_norm": 0.5550187826156616,
      "learning_rate": 0.000495185849381568,
      "loss": 0.8234,
      "step": 5985
    },
    {
      "epoch": 0.7200384661617983,
      "grad_norm": 0.4478211998939514,
      "learning_rate": 0.000495176532598963,
      "loss": 1.3477,
      "step": 5990
    },
    {
      "epoch": 0.7206394999398966,
      "grad_norm": 0.5679304599761963,
      "learning_rate": 0.0004951672068975448,
      "loss": 1.0531,
      "step": 5995
    },
    {
      "epoch": 0.7212405337179949,
      "grad_norm": 0.4058151841163635,
      "learning_rate": 0.0004951578722776526,
      "loss": 0.9563,
      "step": 6000
    },
    {
      "epoch": 0.7212405337179949,
      "eval_loss": 1.9480469226837158,
      "eval_model_preparation_time": 0.0053,
      "eval_runtime": 35.1671,
      "eval_samples_per_second": 4.55,
      "eval_steps_per_second": 1.137,
      "step": 6000
    }
  ],
  "logging_steps": 5,
  "max_steps": 83190,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 3000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 1.307687484653568e+18,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}