{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 500.0,
  "eval_steps": 500,
  "global_step": 5000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 10.25,
      "learning_rate": 1.8e-05,
      "loss": 2.581679344177246,
      "loss_d0": 2.594497573375702,
      "step": 10
    },
    {
      "epoch": 2.0,
      "grad_norm": 4.40625,
      "learning_rate": 3.8e-05,
      "loss": 1.1221242904663087,
      "loss_d0": 1.1283691585063935,
      "step": 20
    },
    {
      "epoch": 3.0,
      "grad_norm": 3.96875,
      "learning_rate": 5.8e-05,
      "loss": 0.7012231349945068,
      "loss_d0": 0.7057632803916931,
      "step": 30
    },
    {
      "epoch": 4.0,
      "grad_norm": 3.625,
      "learning_rate": 7.800000000000001e-05,
      "loss": 0.536361837387085,
      "loss_d0": 0.5375127762556076,
      "step": 40
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.984375,
      "learning_rate": 9.8e-05,
      "loss": 0.44133820533752444,
      "loss_d0": 0.4434512883424759,
      "step": 50
    },
    {
      "epoch": 6.0,
      "grad_norm": 2.6875,
      "learning_rate": 0.000118,
      "loss": 0.39685261249542236,
      "loss_d0": 0.3982961386442184,
      "step": 60
    },
    {
      "epoch": 7.0,
      "grad_norm": 3.46875,
      "learning_rate": 0.000138,
      "loss": 0.3737038135528564,
      "loss_d0": 0.37524707019329073,
      "step": 70
    },
    {
      "epoch": 8.0,
      "grad_norm": 3.125,
      "learning_rate": 0.00015800000000000002,
      "loss": 0.35321481227874757,
      "loss_d0": 0.3535540908575058,
      "step": 80
    },
    {
      "epoch": 9.0,
      "grad_norm": 2.15625,
      "learning_rate": 0.00017800000000000002,
      "loss": 0.2637439727783203,
      "loss_d0": 0.26383873224258425,
      "step": 90
    },
    {
      "epoch": 10.0,
      "grad_norm": 1.9921875,
      "learning_rate": 0.00019800000000000002,
      "loss": 0.23801546096801757,
      "loss_d0": 0.23748010993003846,
      "step": 100
    },
    {
      "epoch": 11.0,
      "grad_norm": 1.7890625,
      "learning_rate": 0.0002,
      "loss": 0.2096700429916382,
      "loss_d0": 0.2091860368847847,
      "step": 110
    },
    {
      "epoch": 12.0,
      "grad_norm": 1.65625,
      "learning_rate": 0.0002,
      "loss": 0.1755399227142334,
      "loss_d0": 0.17519116029143333,
      "step": 120
    },
    {
      "epoch": 13.0,
      "grad_norm": 1.4140625,
      "learning_rate": 0.0002,
      "loss": 0.14668688774108887,
      "loss_d0": 0.14762124717235564,
      "step": 130
    },
    {
      "epoch": 14.0,
      "grad_norm": 1.2890625,
      "learning_rate": 0.0002,
      "loss": 0.13019635677337646,
      "loss_d0": 0.13043430745601653,
      "step": 140
    },
    {
      "epoch": 15.0,
      "grad_norm": 1.0703125,
      "learning_rate": 0.0002,
      "loss": 0.11973921060562134,
      "loss_d0": 0.12013913840055465,
      "step": 150
    },
    {
      "epoch": 16.0,
      "grad_norm": 2.421875,
      "learning_rate": 0.0002,
      "loss": 0.12330312728881836,
      "loss_d0": 0.12271309792995452,
      "step": 160
    },
    {
      "epoch": 17.0,
      "grad_norm": 0.8515625,
      "learning_rate": 0.0002,
      "loss": 0.10923677682876587,
      "loss_d0": 0.10925403907895088,
      "step": 170
    },
    {
      "epoch": 18.0,
      "grad_norm": 1.1484375,
      "learning_rate": 0.0002,
      "loss": 0.09407510757446289,
      "loss_d0": 0.09457137286663056,
      "step": 180
    },
    {
      "epoch": 19.0,
      "grad_norm": 0.92578125,
      "learning_rate": 0.0002,
      "loss": 0.09108944535255432,
      "loss_d0": 0.09049527682363986,
      "step": 190
    },
    {
      "epoch": 20.0,
      "grad_norm": 1.34375,
      "learning_rate": 0.0002,
      "loss": 0.09180540442466736,
      "loss_d0": 0.09225930981338024,
      "step": 200
    },
    {
      "epoch": 21.0,
      "grad_norm": 0.75,
      "learning_rate": 0.0002,
      "loss": 0.08098587393760681,
      "loss_d0": 0.08168995752930641,
      "step": 210
    },
    {
      "epoch": 22.0,
      "grad_norm": 1.0625,
      "learning_rate": 0.0002,
      "loss": 0.09791376590728759,
      "loss_d0": 0.09980626665055752,
      "step": 220
    },
    {
      "epoch": 23.0,
      "grad_norm": 1.0234375,
      "learning_rate": 0.0002,
      "loss": 0.08274852633476257,
      "loss_d0": 0.08421625643968582,
      "step": 230
    },
    {
      "epoch": 24.0,
      "grad_norm": 0.671875,
      "learning_rate": 0.0002,
      "loss": 0.07710103392601013,
      "loss_d0": 0.07744473703205586,
      "step": 240
    },
    {
      "epoch": 25.0,
      "grad_norm": 0.703125,
      "learning_rate": 0.0002,
      "loss": 0.07252544164657593,
      "loss_d0": 0.0727953277528286,
      "step": 250
    },
    {
      "epoch": 26.0,
      "grad_norm": 0.9140625,
      "learning_rate": 0.0002,
      "loss": 0.0800686776638031,
      "loss_d0": 0.08128003366291522,
      "step": 260
    },
    {
      "epoch": 27.0,
      "grad_norm": 0.71875,
      "learning_rate": 0.0002,
      "loss": 0.0793200671672821,
      "loss_d0": 0.07903554365038871,
      "step": 270
    },
    {
      "epoch": 28.0,
      "grad_norm": 0.6015625,
      "learning_rate": 0.0002,
      "loss": 0.07766087651252747,
      "loss_d0": 0.07749549187719822,
      "step": 280
    },
    {
      "epoch": 29.0,
      "grad_norm": 0.78515625,
      "learning_rate": 0.0002,
      "loss": 0.10222091674804687,
      "loss_d0": 0.10210502184927464,
      "step": 290
    },
    {
      "epoch": 30.0,
      "grad_norm": 0.9765625,
      "learning_rate": 0.0002,
      "loss": 0.0746152639389038,
      "loss_d0": 0.07537800036370754,
      "step": 300
    },
    {
      "epoch": 31.0,
      "grad_norm": 0.796875,
      "learning_rate": 0.0002,
      "loss": 0.07073012590408326,
      "loss_d0": 0.07163068279623985,
      "step": 310
    },
    {
      "epoch": 32.0,
      "grad_norm": 1.1953125,
      "learning_rate": 0.0002,
      "loss": 0.07479119300842285,
      "loss_d0": 0.07542065903544426,
      "step": 320
    },
    {
      "epoch": 33.0,
      "grad_norm": 0.78125,
      "learning_rate": 0.0002,
      "loss": 0.07109615206718445,
      "loss_d0": 0.07173952199518681,
      "step": 330
    },
    {
      "epoch": 34.0,
      "grad_norm": 0.48046875,
      "learning_rate": 0.0002,
      "loss": 0.07673498392105102,
      "loss_d0": 0.07673660032451153,
      "step": 340
    },
    {
      "epoch": 35.0,
      "grad_norm": 0.59375,
      "learning_rate": 0.0002,
      "loss": 0.06534476280212402,
      "loss_d0": 0.06561746634542942,
      "step": 350
    },
    {
      "epoch": 36.0,
      "grad_norm": 0.421875,
      "learning_rate": 0.0002,
      "loss": 0.05725439190864563,
      "loss_d0": 0.05825793258845806,
      "step": 360
    },
    {
      "epoch": 37.0,
      "grad_norm": 0.73046875,
      "learning_rate": 0.0002,
      "loss": 0.052153170108795166,
      "loss_d0": 0.052757271006703375,
      "step": 370
    },
    {
      "epoch": 38.0,
      "grad_norm": 0.51953125,
      "learning_rate": 0.0002,
      "loss": 0.054240524768829346,
      "loss_d0": 0.05417148284614086,
      "step": 380
    },
    {
      "epoch": 39.0,
      "grad_norm": 0.63671875,
      "learning_rate": 0.0002,
      "loss": 0.06259162425994873,
      "loss_d0": 0.06249551326036453,
      "step": 390
    },
    {
      "epoch": 40.0,
      "grad_norm": 0.390625,
      "learning_rate": 0.0002,
      "loss": 0.06685826778411866,
      "loss_d0": 0.06701613776385784,
      "step": 400
    },
    {
      "epoch": 41.0,
      "grad_norm": 0.62109375,
      "learning_rate": 0.0002,
      "loss": 0.06873815059661866,
      "loss_d0": 0.06969999745488167,
      "step": 410
    },
    {
      "epoch": 42.0,
      "grad_norm": 0.85546875,
      "learning_rate": 0.0002,
      "loss": 0.0637800931930542,
      "loss_d0": 0.06431917324662209,
      "step": 420
    },
    {
      "epoch": 43.0,
      "grad_norm": 0.64453125,
      "learning_rate": 0.0002,
      "loss": 0.06832690238952636,
      "loss_d0": 0.06909840479493141,
      "step": 430
    },
    {
      "epoch": 44.0,
      "grad_norm": 0.5078125,
      "learning_rate": 0.0002,
      "loss": 0.06170156002044678,
      "loss_d0": 0.06254102811217307,
      "step": 440
    },
    {
      "epoch": 45.0,
      "grad_norm": 0.376953125,
      "learning_rate": 0.0002,
      "loss": 0.05498591065406799,
      "loss_d0": 0.055374423414468764,
      "step": 450
    },
    {
      "epoch": 46.0,
      "grad_norm": 0.419921875,
      "learning_rate": 0.0002,
      "loss": 0.05355701446533203,
      "loss_d0": 0.05435889437794685,
      "step": 460
    },
    {
      "epoch": 47.0,
      "grad_norm": 0.494140625,
      "learning_rate": 0.0002,
      "loss": 0.04821126461029053,
      "loss_d0": 0.04932688735425472,
      "step": 470
    },
    {
      "epoch": 48.0,
      "grad_norm": 0.55078125,
      "learning_rate": 0.0002,
      "loss": 0.048481115698814393,
      "loss_d0": 0.04974230490624905,
      "step": 480
    },
    {
      "epoch": 49.0,
      "grad_norm": 0.36328125,
      "learning_rate": 0.0002,
      "loss": 0.05377357602119446,
      "loss_d0": 0.054486358910799025,
      "step": 490
    },
    {
      "epoch": 50.0,
      "grad_norm": 0.546875,
      "learning_rate": 0.0002,
      "loss": 0.0658548891544342,
      "loss_d0": 0.06611061803996562,
      "step": 500
    },
    {
      "epoch": 50.0,
      "eval_loss": 8.500927925109863,
      "eval_runtime": 0.3464,
      "eval_samples_per_second": 721.678,
      "eval_steps_per_second": 72.168,
      "step": 500
    },
    {
      "epoch": 51.0,
      "grad_norm": 0.5859375,
      "learning_rate": 0.0002,
      "loss": 0.06575464010238648,
      "loss_d0": 0.06647776179015637,
      "step": 510
    },
    {
      "epoch": 52.0,
      "grad_norm": 0.6640625,
      "learning_rate": 0.0002,
      "loss": 0.06158992648124695,
      "loss_d0": 0.06197560206055641,
      "step": 520
    },
    {
      "epoch": 53.0,
      "grad_norm": 0.58984375,
      "learning_rate": 0.0002,
      "loss": 0.06950807571411133,
      "loss_d0": 0.0693976704031229,
      "step": 530
    },
    {
      "epoch": 54.0,
      "grad_norm": 0.640625,
      "learning_rate": 0.0002,
      "loss": 0.07226678133010864,
      "loss_d0": 0.07176880724728107,
      "step": 540
    },
    {
      "epoch": 55.0,
      "grad_norm": 0.5234375,
      "learning_rate": 0.0002,
      "loss": 0.053427815437316895,
      "loss_d0": 0.053496473282575605,
      "step": 550
    },
    {
      "epoch": 56.0,
      "grad_norm": 0.42578125,
      "learning_rate": 0.0002,
      "loss": 0.05263411998748779,
      "loss_d0": 0.052844957634806636,
      "step": 560
    },
    {
      "epoch": 57.0,
      "grad_norm": 0.486328125,
      "learning_rate": 0.0002,
      "loss": 0.05049826502799988,
      "loss_d0": 0.0504428181797266,
      "step": 570
    },
    {
      "epoch": 58.0,
      "grad_norm": 0.72265625,
      "learning_rate": 0.0002,
      "loss": 0.05417170524597168,
      "loss_d0": 0.05384328365325928,
      "step": 580
    },
    {
      "epoch": 59.0,
      "grad_norm": 0.51953125,
      "learning_rate": 0.0002,
      "loss": 0.059265947341918944,
      "loss_d0": 0.059712383151054385,
      "step": 590
    },
    {
      "epoch": 60.0,
      "grad_norm": 0.443359375,
      "learning_rate": 0.0002,
      "loss": 0.051920360326766966,
      "loss_d0": 0.051929228007793427,
      "step": 600
    },
    {
      "epoch": 61.0,
      "grad_norm": 0.68359375,
      "learning_rate": 0.0002,
      "loss": 0.055924856662750246,
      "loss_d0": 0.056316133588552475,
      "step": 610
    },
    {
      "epoch": 62.0,
      "grad_norm": 0.376953125,
      "learning_rate": 0.0002,
      "loss": 0.052934885025024414,
      "loss_d0": 0.05302453897893429,
      "step": 620
    },
    {
      "epoch": 63.0,
      "grad_norm": 0.4765625,
      "learning_rate": 0.0002,
      "loss": 0.061573910713195804,
      "loss_d0": 0.06178021095693111,
      "step": 630
    },
    {
      "epoch": 64.0,
      "grad_norm": 0.45703125,
      "learning_rate": 0.0002,
      "loss": 0.05103100538253784,
      "loss_d0": 0.05124416649341583,
      "step": 640
    },
    {
      "epoch": 65.0,
      "grad_norm": 0.314453125,
      "learning_rate": 0.0002,
      "loss": 0.044987863302230834,
      "loss_d0": 0.044990910589694975,
      "step": 650
    },
    {
      "epoch": 66.0,
      "grad_norm": 0.357421875,
      "learning_rate": 0.0002,
      "loss": 0.054930204153060914,
      "loss_d0": 0.054256043210625646,
      "step": 660
    },
    {
      "epoch": 67.0,
      "grad_norm": 0.373046875,
      "learning_rate": 0.0002,
      "loss": 0.05119019150733948,
      "loss_d0": 0.05134856700897217,
      "step": 670
    },
    {
      "epoch": 68.0,
      "grad_norm": 0.462890625,
      "learning_rate": 0.0002,
      "loss": 0.054933798313140866,
      "loss_d0": 0.05563933551311493,
      "step": 680
    },
    {
      "epoch": 69.0,
      "grad_norm": 0.6015625,
      "learning_rate": 0.0002,
      "loss": 0.04948141574859619,
      "loss_d0": 0.04973691999912262,
      "step": 690
    },
    {
      "epoch": 70.0,
      "grad_norm": 0.48046875,
      "learning_rate": 0.0002,
      "loss": 0.05181562304496765,
      "loss_d0": 0.051676081120967866,
      "step": 700
    },
    {
      "epoch": 71.0,
      "grad_norm": 0.44921875,
      "learning_rate": 0.0002,
      "loss": 0.05025671720504761,
      "loss_d0": 0.05064960978925228,
      "step": 710
    },
    {
      "epoch": 72.0,
      "grad_norm": 0.5859375,
      "learning_rate": 0.0002,
      "loss": 0.052214968204498294,
      "loss_d0": 0.05240287780761719,
      "step": 720
    },
    {
      "epoch": 73.0,
      "grad_norm": 0.4140625,
      "learning_rate": 0.0002,
      "loss": 0.050692415237426756,
      "loss_d0": 0.05136246755719185,
      "step": 730
    },
    {
      "epoch": 74.0,
      "grad_norm": 0.408203125,
      "learning_rate": 0.0002,
      "loss": 0.05172141194343567,
      "loss_d0": 0.0518946498632431,
      "step": 740
    },
    {
      "epoch": 75.0,
      "grad_norm": 0.44921875,
      "learning_rate": 0.0002,
      "loss": 0.05207560658454895,
      "loss_d0": 0.05260260328650475,
      "step": 750
    },
    {
      "epoch": 76.0,
      "grad_norm": 0.703125,
      "learning_rate": 0.0002,
      "loss": 0.050001639127731326,
      "loss_d0": 0.05025416761636734,
      "step": 760
    },
    {
      "epoch": 77.0,
      "grad_norm": 0.47265625,
      "learning_rate": 0.0002,
      "loss": 0.05551525950431824,
      "loss_d0": 0.056026375666260717,
      "step": 770
    },
    {
      "epoch": 78.0,
      "grad_norm": 0.458984375,
      "learning_rate": 0.0002,
      "loss": 0.051217657327651975,
      "loss_d0": 0.052097849920392035,
      "step": 780
    },
    {
      "epoch": 79.0,
      "grad_norm": 0.546875,
      "learning_rate": 0.0002,
      "loss": 0.05090489983558655,
      "loss_d0": 0.05193852297961712,
      "step": 790
    },
    {
      "epoch": 80.0,
      "grad_norm": 0.64453125,
      "learning_rate": 0.0002,
      "loss": 0.05624193549156189,
      "loss_d0": 0.05703345984220505,
      "step": 800
    },
    {
      "epoch": 81.0,
      "grad_norm": 0.484375,
      "learning_rate": 0.0002,
      "loss": 0.05463804602622986,
      "loss_d0": 0.05500471815466881,
      "step": 810
    },
    {
      "epoch": 82.0,
      "grad_norm": 0.4921875,
      "learning_rate": 0.0002,
      "loss": 0.0497215062379837,
      "loss_d0": 0.050147075578570365,
      "step": 820
    },
    {
      "epoch": 83.0,
      "grad_norm": 0.478515625,
      "learning_rate": 0.0002,
      "loss": 0.04716762006282806,
      "loss_d0": 0.047426237538456914,
      "step": 830
    },
    {
      "epoch": 84.0,
      "grad_norm": 0.4921875,
      "learning_rate": 0.0002,
      "loss": 0.04353850483894348,
      "loss_d0": 0.04430970065295696,
      "step": 840
    },
    {
      "epoch": 85.0,
      "grad_norm": 0.41015625,
      "learning_rate": 0.0002,
      "loss": 0.047328507900238036,
      "loss_d0": 0.04820728190243244,
      "step": 850
    },
    {
      "epoch": 86.0,
      "grad_norm": 0.447265625,
      "learning_rate": 0.0002,
      "loss": 0.04739128351211548,
      "loss_d0": 0.04819867946207523,
      "step": 860
    },
    {
      "epoch": 87.0,
      "grad_norm": 0.435546875,
      "learning_rate": 0.0002,
      "loss": 0.04916658103466034,
      "loss_d0": 0.05031695365905762,
      "step": 870
    },
    {
      "epoch": 88.0,
      "grad_norm": 0.353515625,
      "learning_rate": 0.0002,
      "loss": 0.051763898134231566,
      "loss_d0": 0.05217132642865181,
      "step": 880
    },
    {
      "epoch": 89.0,
      "grad_norm": 0.50390625,
      "learning_rate": 0.0002,
      "loss": 0.048361048102378845,
      "loss_d0": 0.049080048128962515,
      "step": 890
    },
    {
      "epoch": 90.0,
      "grad_norm": 0.455078125,
      "learning_rate": 0.0002,
      "loss": 0.05283051133155823,
      "loss_d0": 0.05397628545761109,
      "step": 900
    },
    {
      "epoch": 91.0,
      "grad_norm": 0.392578125,
      "learning_rate": 0.0002,
      "loss": 0.04917954802513123,
      "loss_d0": 0.04918815195560455,
      "step": 910
    },
    {
      "epoch": 92.0,
      "grad_norm": 0.28515625,
      "learning_rate": 0.0002,
      "loss": 0.056238424777984616,
      "loss_d0": 0.057272735983133316,
      "step": 920
    },
    {
      "epoch": 93.0,
      "grad_norm": 0.431640625,
      "learning_rate": 0.0002,
      "loss": 0.05657249689102173,
      "loss_d0": 0.05702277310192585,
      "step": 930
    },
    {
      "epoch": 94.0,
      "grad_norm": 0.357421875,
      "learning_rate": 0.0002,
      "loss": 0.04906629025936127,
      "loss_d0": 0.049605177342891695,
      "step": 940
    },
    {
      "epoch": 95.0,
      "grad_norm": 0.484375,
      "learning_rate": 0.0002,
      "loss": 0.046135890483856204,
      "loss_d0": 0.04623321630060673,
      "step": 950
    },
    {
      "epoch": 96.0,
      "grad_norm": 0.37890625,
      "learning_rate": 0.0002,
      "loss": 0.047981783747673035,
      "loss_d0": 0.048135362938046454,
      "step": 960
    },
    {
      "epoch": 97.0,
      "grad_norm": 0.470703125,
      "learning_rate": 0.0002,
      "loss": 0.05710349082946777,
      "loss_d0": 0.05879225209355354,
      "step": 970
    },
    {
      "epoch": 98.0,
      "grad_norm": 0.291015625,
      "learning_rate": 0.0002,
      "loss": 0.04263607263565063,
      "loss_d0": 0.04290797896683216,
      "step": 980
    },
    {
      "epoch": 99.0,
      "grad_norm": 0.3671875,
      "learning_rate": 0.0002,
      "loss": 0.045274817943573,
      "loss_d0": 0.0448206938803196,
      "step": 990
    },
    {
      "epoch": 100.0,
      "grad_norm": 0.4140625,
      "learning_rate": 0.0002,
      "loss": 0.0417318195104599,
      "loss_d0": 0.04169005416333675,
      "step": 1000
    },
    {
      "epoch": 100.0,
      "eval_loss": 8.929769515991211,
      "eval_runtime": 0.348,
      "eval_samples_per_second": 718.488,
      "eval_steps_per_second": 71.849,
      "step": 1000
    },
    {
      "epoch": 101.0,
      "grad_norm": 0.31640625,
      "learning_rate": 0.0002,
      "loss": 0.04035484790802002,
      "loss_d0": 0.04052091836929321,
      "step": 1010
    },
    {
      "epoch": 102.0,
      "grad_norm": 0.396484375,
      "learning_rate": 0.0002,
      "loss": 0.043715986609458926,
      "loss_d0": 0.04388895146548748,
      "step": 1020
    },
    {
      "epoch": 103.0,
      "grad_norm": 0.4375,
      "learning_rate": 0.0002,
      "loss": 0.043914207816123964,
      "loss_d0": 0.044013023003935815,
      "step": 1030
    },
    {
      "epoch": 104.0,
      "grad_norm": 0.42578125,
      "learning_rate": 0.0002,
      "loss": 0.05073145627975464,
      "loss_d0": 0.051011853665113446,
      "step": 1040
    },
    {
      "epoch": 105.0,
      "grad_norm": 0.455078125,
      "learning_rate": 0.0002,
      "loss": 0.043595680594444276,
      "loss_d0": 0.04368243329226971,
      "step": 1050
    },
    {
      "epoch": 106.0,
      "grad_norm": 0.38671875,
      "learning_rate": 0.0002,
      "loss": 0.043256306648254396,
      "loss_d0": 0.04335380308330059,
      "step": 1060
    },
    {
      "epoch": 107.0,
      "grad_norm": 0.56640625,
      "learning_rate": 0.0002,
      "loss": 0.04898662865161896,
      "loss_d0": 0.04895379841327667,
      "step": 1070
    },
    {
      "epoch": 108.0,
      "grad_norm": 0.404296875,
      "learning_rate": 0.0002,
      "loss": 0.06072341799736023,
      "loss_d0": 0.059587860107421876,
      "step": 1080
    },
    {
      "epoch": 109.0,
      "grad_norm": 0.42578125,
      "learning_rate": 0.0002,
      "loss": 0.0477506160736084,
      "loss_d0": 0.04787628389894962,
      "step": 1090
    },
    {
      "epoch": 110.0,
      "grad_norm": 0.51953125,
      "learning_rate": 0.0002,
      "loss": 0.05249601602554321,
      "loss_d0": 0.0535334087908268,
      "step": 1100
    },
    {
      "epoch": 111.0,
      "grad_norm": 0.419921875,
      "learning_rate": 0.0002,
      "loss": 0.04094515442848205,
      "loss_d0": 0.04123275801539421,
      "step": 1110
    },
    {
      "epoch": 112.0,
      "grad_norm": 0.39453125,
      "learning_rate": 0.0002,
      "loss": 0.041908282041549685,
      "loss_d0": 0.04213452525436878,
      "step": 1120
    },
    {
      "epoch": 113.0,
      "grad_norm": 0.36328125,
      "learning_rate": 0.0002,
      "loss": 0.04445734322071075,
      "loss_d0": 0.04442833214998245,
      "step": 1130
    },
    {
      "epoch": 114.0,
      "grad_norm": 0.46875,
      "learning_rate": 0.0002,
      "loss": 0.05096038579940796,
      "loss_d0": 0.05110368393361568,
      "step": 1140
    },
    {
      "epoch": 115.0,
      "grad_norm": 0.54296875,
      "learning_rate": 0.0002,
      "loss": 0.044823643565177915,
      "loss_d0": 0.04530790410935879,
      "step": 1150
    },
    {
      "epoch": 116.0,
      "grad_norm": 0.47265625,
      "learning_rate": 0.0002,
      "loss": 0.04294084906578064,
      "loss_d0": 0.043658243492245674,
      "step": 1160
    },
    {
      "epoch": 117.0,
      "grad_norm": 0.462890625,
      "learning_rate": 0.0002,
      "loss": 0.048780834674835204,
      "loss_d0": 0.0490945640951395,
      "step": 1170
    },
    {
      "epoch": 118.0,
      "grad_norm": 0.38671875,
      "learning_rate": 0.0002,
      "loss": 0.04995758831501007,
      "loss_d0": 0.04992273151874542,
      "step": 1180
    },
    {
      "epoch": 119.0,
      "grad_norm": 0.3203125,
      "learning_rate": 0.0002,
      "loss": 0.04624115526676178,
      "loss_d0": 0.04665086269378662,
      "step": 1190
    },
    {
      "epoch": 120.0,
      "grad_norm": 0.365234375,
      "learning_rate": 0.0002,
      "loss": 0.04964223206043243,
      "loss_d0": 0.050650115683674814,
      "step": 1200
    },
    {
      "epoch": 121.0,
      "grad_norm": 0.404296875,
      "learning_rate": 0.0002,
      "loss": 0.04927041232585907,
      "loss_d0": 0.05039941743016243,
      "step": 1210
    },
    {
      "epoch": 122.0,
      "grad_norm": 0.33203125,
      "learning_rate": 0.0002,
      "loss": 0.044230511784553526,
      "loss_d0": 0.044757972657680514,
      "step": 1220
    },
    {
      "epoch": 123.0,
      "grad_norm": 0.373046875,
      "learning_rate": 0.0002,
      "loss": 0.04596873819828033,
      "loss_d0": 0.047029031440615654,
      "step": 1230
    },
    {
      "epoch": 124.0,
      "grad_norm": 0.443359375,
      "learning_rate": 0.0002,
      "loss": 0.04350687265396118,
      "loss_d0": 0.0439357616007328,
      "step": 1240
    },
    {
      "epoch": 125.0,
      "grad_norm": 0.6640625,
      "learning_rate": 0.0002,
      "loss": 0.0479256808757782,
      "loss_d0": 0.0486279521137476,
      "step": 1250
    },
    {
      "epoch": 126.0,
      "grad_norm": 0.3046875,
      "learning_rate": 0.0002,
      "loss": 0.04501887857913971,
      "loss_d0": 0.045443673804402354,
      "step": 1260
    },
    {
      "epoch": 127.0,
      "grad_norm": 0.33203125,
      "learning_rate": 0.0002,
      "loss": 0.04470711350440979,
      "loss_d0": 0.045029261335730555,
      "step": 1270
    },
    {
      "epoch": 128.0,
      "grad_norm": 0.455078125,
      "learning_rate": 0.0002,
      "loss": 0.04421221315860748,
      "loss_d0": 0.04455976895987988,
      "step": 1280
    },
    {
      "epoch": 129.0,
      "grad_norm": 0.306640625,
      "learning_rate": 0.0002,
      "loss": 0.040431278944015506,
      "loss_d0": 0.04085499048233032,
      "step": 1290
    },
    {
      "epoch": 130.0,
      "grad_norm": 0.3515625,
      "learning_rate": 0.0002,
      "loss": 0.04223118424415588,
      "loss_d0": 0.04227718524634838,
      "step": 1300
    },
    {
      "epoch": 131.0,
      "grad_norm": 0.4140625,
      "learning_rate": 0.0002,
      "loss": 0.04075292944908142,
      "loss_d0": 0.04081529788672924,
      "step": 1310
    },
    {
      "epoch": 132.0,
      "grad_norm": 0.40625,
      "learning_rate": 0.0002,
      "loss": 0.03815374970436096,
      "loss_d0": 0.03851891085505486,
      "step": 1320
    },
    {
      "epoch": 133.0,
      "grad_norm": 0.310546875,
      "learning_rate": 0.0002,
      "loss": 0.04056335985660553,
      "loss_d0": 0.04040697924792767,
      "step": 1330
    },
    {
      "epoch": 134.0,
      "grad_norm": 0.435546875,
      "learning_rate": 0.0002,
      "loss": 0.04462970495223999,
      "loss_d0": 0.04509739205241203,
      "step": 1340
    },
    {
      "epoch": 135.0,
      "grad_norm": 0.4921875,
      "learning_rate": 0.0002,
      "loss": 0.04322163462638855,
      "loss_d0": 0.04335412159562111,
      "step": 1350
    },
    {
      "epoch": 136.0,
      "grad_norm": 0.52734375,
      "learning_rate": 0.0002,
      "loss": 0.06365315318107605,
      "loss_d0": 0.06332227997481824,
      "step": 1360
    },
    {
      "epoch": 137.0,
      "grad_norm": 0.41796875,
      "learning_rate": 0.0002,
      "loss": 0.04837386906147003,
      "loss_d0": 0.04851404689252377,
      "step": 1370
    },
    {
      "epoch": 138.0,
      "grad_norm": 0.4140625,
      "learning_rate": 0.0002,
      "loss": 0.05851868987083435,
      "loss_d0": 0.0596237625926733,
      "step": 1380
    },
    {
      "epoch": 139.0,
      "grad_norm": 0.373046875,
      "learning_rate": 0.0002,
      "loss": 0.045863062143325806,
      "loss_d0": 0.045821748301386835,
      "step": 1390
    },
    {
      "epoch": 140.0,
      "grad_norm": 0.314453125,
      "learning_rate": 0.0002,
      "loss": 0.041532912850379945,
      "loss_d0": 0.04213090687990188,
      "step": 1400
    },
    {
      "epoch": 141.0,
      "grad_norm": 0.318359375,
      "learning_rate": 0.0002,
      "loss": 0.0425410121679306,
      "loss_d0": 0.043128667399287224,
      "step": 1410
    },
    {
      "epoch": 142.0,
      "grad_norm": 0.56640625,
      "learning_rate": 0.0002,
      "loss": 0.051248615980148314,
      "loss_d0": 0.05137891061604023,
      "step": 1420
    },
    {
      "epoch": 143.0,
      "grad_norm": 0.32421875,
      "learning_rate": 0.0002,
      "loss": 0.038969531655311584,
      "loss_d0": 0.03941834345459938,
      "step": 1430
    },
    {
      "epoch": 144.0,
      "grad_norm": 0.263671875,
      "learning_rate": 0.0002,
      "loss": 0.039830029010772705,
      "loss_d0": 0.04029415212571621,
      "step": 1440
    },
    {
      "epoch": 145.0,
      "grad_norm": 0.349609375,
      "learning_rate": 0.0002,
      "loss": 0.039839529991149904,
      "loss_d0": 0.03990244455635548,
      "step": 1450
    },
    {
      "epoch": 146.0,
      "grad_norm": 0.68359375,
      "learning_rate": 0.0002,
      "loss": 0.04552115201950073,
      "loss_d0": 0.04556415602564812,
      "step": 1460
    },
    {
      "epoch": 147.0,
      "grad_norm": 0.78125,
      "learning_rate": 0.0002,
      "loss": 0.04971630275249481,
      "loss_d0": 0.04951465502381325,
      "step": 1470
    },
    {
      "epoch": 148.0,
      "grad_norm": 0.365234375,
      "learning_rate": 0.0002,
      "loss": 0.046517756581306455,
      "loss_d0": 0.04670452438294888,
      "step": 1480
    },
    {
      "epoch": 149.0,
      "grad_norm": 0.27734375,
      "learning_rate": 0.0002,
      "loss": 0.04213971495628357,
      "loss_d0": 0.04255228489637375,
      "step": 1490
    },
    {
      "epoch": 150.0,
      "grad_norm": 0.40625,
      "learning_rate": 0.0002,
      "loss": 0.044486671686172485,
      "loss_d0": 0.045111048594117166,
      "step": 1500
    },
    {
      "epoch": 150.0,
      "eval_loss": 9.255965232849121,
      "eval_runtime": 0.3455,
      "eval_samples_per_second": 723.579,
      "eval_steps_per_second": 72.358,
      "step": 1500
    },
    {
      "epoch": 151.0,
      "grad_norm": 0.609375,
      "learning_rate": 0.0002,
      "loss": 0.04134907722473145,
      "loss_d0": 0.04163525775074959,
      "step": 1510
    },
    {
      "epoch": 152.0,
      "grad_norm": 0.375,
      "learning_rate": 0.0002,
      "loss": 0.04379453361034393,
      "loss_d0": 0.044176562875509265,
      "step": 1520
    },
    {
      "epoch": 153.0,
      "grad_norm": 0.353515625,
      "learning_rate": 0.0002,
      "loss": 0.04323039650917053,
      "loss_d0": 0.04346066489815712,
      "step": 1530
    },
    {
      "epoch": 154.0,
      "grad_norm": 0.37109375,
      "learning_rate": 0.0002,
      "loss": 0.04052256941795349,
      "loss_d0": 0.04064999632537365,
      "step": 1540
    },
    {
      "epoch": 155.0,
      "grad_norm": 0.37109375,
      "learning_rate": 0.0002,
      "loss": 0.04049890339374542,
      "loss_d0": 0.04063104912638664,
      "step": 1550
    },
    {
      "epoch": 156.0,
      "grad_norm": 0.53125,
      "learning_rate": 0.0002,
      "loss": 0.03896942138671875,
      "loss_d0": 0.03932310901582241,
      "step": 1560
    },
    {
      "epoch": 157.0,
      "grad_norm": 0.3046875,
      "learning_rate": 0.0002,
      "loss": 0.04199636280536652,
      "loss_d0": 0.042282988876104356,
      "step": 1570
    },
    {
      "epoch": 158.0,
      "grad_norm": 0.365234375,
      "learning_rate": 0.0002,
      "loss": 0.041889071464538574,
      "loss_d0": 0.04293041825294495,
      "step": 1580
    },
    {
      "epoch": 159.0,
      "grad_norm": 0.37109375,
      "learning_rate": 0.0002,
      "loss": 0.04387465715408325,
      "loss_d0": 0.044143912941217424,
      "step": 1590
    },
    {
      "epoch": 160.0,
      "grad_norm": 0.357421875,
      "learning_rate": 0.0002,
      "loss": 0.048598700761795045,
      "loss_d0": 0.04977954216301441,
      "step": 1600
    },
    {
      "epoch": 161.0,
      "grad_norm": 0.4140625,
      "learning_rate": 0.0002,
      "loss": 0.050293147563934326,
      "loss_d0": 0.05142668709158897,
      "step": 1610
    },
    {
      "epoch": 162.0,
      "grad_norm": 0.38671875,
      "learning_rate": 0.0002,
      "loss": 0.052027159929275514,
      "loss_d0": 0.05179822221398354,
      "step": 1620
    },
    {
      "epoch": 163.0,
      "grad_norm": 0.388671875,
      "learning_rate": 0.0002,
      "loss": 0.046048501133918764,
      "loss_d0": 0.04631427824497223,
      "step": 1630
    },
    {
      "epoch": 164.0,
      "grad_norm": 0.5078125,
      "learning_rate": 0.0002,
      "loss": 0.04881615936756134,
      "loss_d0": 0.04941669814288616,
      "step": 1640
    },
    {
      "epoch": 165.0,
      "grad_norm": 0.486328125,
      "learning_rate": 0.0002,
      "loss": 0.04277309775352478,
      "loss_d0": 0.04293011873960495,
      "step": 1650
    },
    {
      "epoch": 166.0,
      "grad_norm": 0.341796875,
      "learning_rate": 0.0002,
      "loss": 0.03941013514995575,
      "loss_d0": 0.03964241743087769,
      "step": 1660
    },
    {
      "epoch": 167.0,
      "grad_norm": 0.369140625,
      "learning_rate": 0.0002,
      "loss": 0.037150949239730835,
      "loss_d0": 0.037612902373075484,
      "step": 1670
    },
    {
      "epoch": 168.0,
      "grad_norm": 0.3203125,
      "learning_rate": 0.0002,
      "loss": 0.035756093263626096,
      "loss_d0": 0.03606977947056293,
      "step": 1680
    },
    {
      "epoch": 169.0,
      "grad_norm": 0.326171875,
      "learning_rate": 0.0002,
      "loss": 0.03544518649578095,
      "loss_d0": 0.03580256886780262,
      "step": 1690
    },
    {
      "epoch": 170.0,
      "grad_norm": 0.232421875,
      "learning_rate": 0.0002,
      "loss": 0.035137712955474854,
      "loss_d0": 0.035555777698755266,
      "step": 1700
    },
    {
      "epoch": 171.0,
      "grad_norm": 0.2333984375,
      "learning_rate": 0.0002,
      "loss": 0.035359475016593936,
      "loss_d0": 0.03564097918570042,
      "step": 1710
    },
    {
      "epoch": 172.0,
      "grad_norm": 0.40234375,
      "learning_rate": 0.0002,
      "loss": 0.03545220494270325,
      "loss_d0": 0.035841194540262224,
      "step": 1720
    },
    {
      "epoch": 173.0,
      "grad_norm": 0.3046875,
      "learning_rate": 0.0002,
      "loss": 0.03551629185676575,
      "loss_d0": 0.03590298071503639,
      "step": 1730
    },
    {
      "epoch": 174.0,
      "grad_norm": 0.259765625,
      "learning_rate": 0.0002,
      "loss": 0.03612012267112732,
      "loss_d0": 0.03644734881818294,
      "step": 1740
    },
    {
      "epoch": 175.0,
      "grad_norm": 0.380859375,
      "learning_rate": 0.0002,
      "loss": 0.037238198518753055,
      "loss_d0": 0.037554935365915296,
      "step": 1750
    },
    {
      "epoch": 176.0,
      "grad_norm": 0.28125,
      "learning_rate": 0.0002,
      "loss": 0.03618746995925903,
      "loss_d0": 0.036390235647559166,
      "step": 1760
    },
    {
      "epoch": 177.0,
      "grad_norm": 0.3671875,
      "learning_rate": 0.0002,
      "loss": 0.05430065989494324,
      "loss_d0": 0.054887108877301215,
      "step": 1770
    },
    {
      "epoch": 178.0,
      "grad_norm": 0.34375,
      "learning_rate": 0.0002,
      "loss": 0.044845637679100034,
      "loss_d0": 0.0448883879929781,
      "step": 1780
    },
    {
      "epoch": 179.0,
      "grad_norm": 0.28515625,
      "learning_rate": 0.0002,
      "loss": 0.04127602577209473,
      "loss_d0": 0.041632624343037605,
      "step": 1790
    },
    {
      "epoch": 180.0,
      "grad_norm": 0.40234375,
      "learning_rate": 0.0002,
      "loss": 0.044097939133644105,
      "loss_d0": 0.043751441314816474,
      "step": 1800
    },
    {
      "epoch": 181.0,
      "grad_norm": 0.3828125,
      "learning_rate": 0.0002,
      "loss": 0.041216480731964114,
      "loss_d0": 0.04157479107379913,
      "step": 1810
    },
    {
      "epoch": 182.0,
      "grad_norm": 0.427734375,
      "learning_rate": 0.0002,
      "loss": 0.04994258582592011,
      "loss_d0": 0.05002143010497093,
      "step": 1820
    },
    {
      "epoch": 183.0,
      "grad_norm": 0.3828125,
      "learning_rate": 0.0002,
      "loss": 0.04401806592941284,
      "loss_d0": 0.04452627077698708,
      "step": 1830
    },
    {
      "epoch": 184.0,
      "grad_norm": 0.48828125,
      "learning_rate": 0.0002,
      "loss": 0.05316650867462158,
      "loss_d0": 0.05357714407145977,
      "step": 1840
    },
    {
      "epoch": 185.0,
      "grad_norm": 0.43359375,
      "learning_rate": 0.0002,
      "loss": 0.0466656357049942,
      "loss_d0": 0.04696873389184475,
      "step": 1850
    },
    {
      "epoch": 186.0,
      "grad_norm": 0.37109375,
      "learning_rate": 0.0002,
      "loss": 0.040113252401351926,
      "loss_d0": 0.04068625308573246,
      "step": 1860
    },
    {
      "epoch": 187.0,
      "grad_norm": 0.333984375,
      "learning_rate": 0.0002,
      "loss": 0.03879719078540802,
      "loss_d0": 0.03932934701442718,
      "step": 1870
    },
    {
      "epoch": 188.0,
      "grad_norm": 0.275390625,
      "learning_rate": 0.0002,
      "loss": 0.04475597441196442,
      "loss_d0": 0.045417624339461324,
      "step": 1880
    },
    {
      "epoch": 189.0,
      "grad_norm": 0.384765625,
      "learning_rate": 0.0002,
      "loss": 0.04336912930011749,
      "loss_d0": 0.043632737919688226,
      "step": 1890
    },
    {
      "epoch": 190.0,
      "grad_norm": 0.296875,
      "learning_rate": 0.0002,
      "loss": 0.04053463935852051,
      "loss_d0": 0.0409706324338913,
      "step": 1900
    },
    {
      "epoch": 191.0,
      "grad_norm": 0.41015625,
      "learning_rate": 0.0002,
      "loss": 0.05158035755157471,
      "loss_d0": 0.05149606578052044,
      "step": 1910
    },
    {
      "epoch": 192.0,
      "grad_norm": 0.341796875,
      "learning_rate": 0.0002,
      "loss": 0.0414920300245285,
      "loss_d0": 0.04197779670357704,
      "step": 1920
    },
    {
      "epoch": 193.0,
      "grad_norm": 0.322265625,
      "learning_rate": 0.0002,
      "loss": 0.03931396007537842,
      "loss_d0": 0.039707991480827334,
      "step": 1930
    },
    {
      "epoch": 194.0,
      "grad_norm": 0.35546875,
      "learning_rate": 0.0002,
      "loss": 0.03731209933757782,
      "loss_d0": 0.03755508027970791,
      "step": 1940
    },
    {
      "epoch": 195.0,
      "grad_norm": 0.3359375,
      "learning_rate": 0.0002,
      "loss": 0.03920927345752716,
      "loss_d0": 0.03930322341620922,
      "step": 1950
    },
    {
      "epoch": 196.0,
      "grad_norm": 0.287109375,
      "learning_rate": 0.0002,
      "loss": 0.035973536968231204,
      "loss_d0": 0.0363770566880703,
      "step": 1960
    },
    {
      "epoch": 197.0,
      "grad_norm": 0.267578125,
      "learning_rate": 0.0002,
      "loss": 0.03653622269630432,
      "loss_d0": 0.03696334585547447,
      "step": 1970
    },
    {
      "epoch": 198.0,
      "grad_norm": 0.53515625,
      "learning_rate": 0.0002,
      "loss": 0.04161535501480103,
      "loss_d0": 0.041733665764331816,
      "step": 1980
    },
    {
      "epoch": 199.0,
      "grad_norm": 0.37890625,
      "learning_rate": 0.0002,
      "loss": 0.04091428816318512,
      "loss_d0": 0.04112983606755734,
      "step": 1990
    },
    {
      "epoch": 200.0,
      "grad_norm": 0.392578125,
      "learning_rate": 0.0002,
      "loss": 0.04442690908908844,
      "loss_d0": 0.044637547433376314,
      "step": 2000
    },
    {
      "epoch": 200.0,
      "eval_loss": 10.370587348937988,
      "eval_runtime": 0.3453,
      "eval_samples_per_second": 723.928,
      "eval_steps_per_second": 72.393,
      "step": 2000
    },
    {
      "epoch": 201.0,
      "grad_norm": 0.3828125,
      "learning_rate": 0.0002,
      "loss": 0.04656229317188263,
      "loss_d0": 0.046998518705368045,
      "step": 2010
    },
    {
      "epoch": 202.0,
      "grad_norm": 0.34375,
      "learning_rate": 0.0002,
      "loss": 0.04548130929470062,
      "loss_d0": 0.045519032329320905,
      "step": 2020
    },
    {
      "epoch": 203.0,
      "grad_norm": 0.322265625,
      "learning_rate": 0.0002,
      "loss": 0.04562685787677765,
      "loss_d0": 0.046041432023048404,
      "step": 2030
    },
    {
      "epoch": 204.0,
      "grad_norm": 0.408203125,
      "learning_rate": 0.0002,
      "loss": 0.03905513286590576,
      "loss_d0": 0.03923566862940788,
      "step": 2040
    },
    {
      "epoch": 205.0,
      "grad_norm": 0.30859375,
      "learning_rate": 0.0002,
      "loss": 0.056406271457672116,
      "loss_d0": 0.056651026383042334,
      "step": 2050
    },
    {
      "epoch": 206.0,
      "grad_norm": 0.298828125,
      "learning_rate": 0.0002,
      "loss": 0.04273752272129059,
      "loss_d0": 0.04327204711735248,
      "step": 2060
    },
    {
      "epoch": 207.0,
      "grad_norm": 0.3359375,
      "learning_rate": 0.0002,
      "loss": 0.04044491946697235,
      "loss_d0": 0.04109087809920311,
      "step": 2070
    },
    {
      "epoch": 208.0,
      "grad_norm": 0.34375,
      "learning_rate": 0.0002,
      "loss": 0.03677527606487274,
      "loss_d0": 0.037081345170736316,
      "step": 2080
    },
    {
      "epoch": 209.0,
      "grad_norm": 0.30859375,
      "learning_rate": 0.0002,
      "loss": 0.035640382766723634,
      "loss_d0": 0.03597054369747639,
      "step": 2090
    },
    {
      "epoch": 210.0,
      "grad_norm": 0.32421875,
      "learning_rate": 0.0002,
      "loss": 0.03491756021976471,
      "loss_d0": 0.03526601791381836,
      "step": 2100
    },
    {
      "epoch": 211.0,
      "grad_norm": 0.302734375,
      "learning_rate": 0.0002,
      "loss": 0.03535093665122986,
      "loss_d0": 0.035667216032743455,
      "step": 2110
    },
    {
      "epoch": 212.0,
      "grad_norm": 0.373046875,
      "learning_rate": 0.0002,
      "loss": 0.03561493158340454,
      "loss_d0": 0.03596670255064964,
      "step": 2120
    },
    {
      "epoch": 213.0,
      "grad_norm": 0.369140625,
      "learning_rate": 0.0002,
      "loss": 0.03607224822044373,
      "loss_d0": 0.036514755338430405,
      "step": 2130
    },
    {
      "epoch": 214.0,
      "grad_norm": 0.30859375,
      "learning_rate": 0.0002,
      "loss": 0.043662580847740176,
      "loss_d0": 0.04414476417005062,
      "step": 2140
    },
    {
      "epoch": 215.0,
      "grad_norm": 0.314453125,
      "learning_rate": 0.0002,
      "loss": 0.04338793754577637,
      "loss_d0": 0.0440433070063591,
      "step": 2150
    },
    {
      "epoch": 216.0,
      "grad_norm": 0.3984375,
      "learning_rate": 0.0002,
      "loss": 0.046759802103042605,
      "loss_d0": 0.04691650569438934,
      "step": 2160
    },
    {
      "epoch": 217.0,
      "grad_norm": 0.390625,
      "learning_rate": 0.0002,
      "loss": 0.043434417247772215,
      "loss_d0": 0.04425071477890015,
      "step": 2170
    },
    {
      "epoch": 218.0,
      "grad_norm": 0.435546875,
      "learning_rate": 0.0002,
      "loss": 0.042248526215553285,
      "loss_d0": 0.04283340498805046,
      "step": 2180
    },
    {
      "epoch": 219.0,
      "grad_norm": 0.37109375,
      "learning_rate": 0.0002,
      "loss": 0.04337837100028992,
      "loss_d0": 0.04396646395325661,
      "step": 2190
    },
    {
      "epoch": 220.0,
      "grad_norm": 0.3203125,
      "learning_rate": 0.0002,
      "loss": 0.03833777606487274,
      "loss_d0": 0.03898017816245556,
      "step": 2200
    },
    {
      "epoch": 221.0,
      "grad_norm": 0.330078125,
      "learning_rate": 0.0002,
      "loss": 0.040862342715263365,
      "loss_d0": 0.041422368586063386,
      "step": 2210
    },
    {
      "epoch": 222.0,
      "grad_norm": 0.36328125,
      "learning_rate": 0.0002,
      "loss": 0.03841360509395599,
      "loss_d0": 0.0387649767100811,
      "step": 2220
    },
    {
      "epoch": 223.0,
      "grad_norm": 0.33984375,
      "learning_rate": 0.0002,
      "loss": 0.04090321660041809,
      "loss_d0": 0.04126690775156021,
      "step": 2230
    },
    {
      "epoch": 224.0,
      "grad_norm": 0.388671875,
      "learning_rate": 0.0002,
      "loss": 0.04010078608989716,
      "loss_d0": 0.04049705974757671,
      "step": 2240
    },
    {
      "epoch": 225.0,
      "grad_norm": 0.494140625,
      "learning_rate": 0.0002,
      "loss": 0.043158155679702756,
      "loss_d0": 0.04369219690561295,
      "step": 2250
    },
    {
      "epoch": 226.0,
      "grad_norm": 0.322265625,
      "learning_rate": 0.0002,
      "loss": 0.04452347755432129,
      "loss_d0": 0.04531903751194477,
      "step": 2260
    },
    {
      "epoch": 227.0,
      "grad_norm": 0.244140625,
      "learning_rate": 0.0002,
      "loss": 0.03993140459060669,
      "loss_d0": 0.04027389250695705,
      "step": 2270
    },
    {
      "epoch": 228.0,
      "grad_norm": 0.3359375,
      "learning_rate": 0.0002,
      "loss": 0.04052254259586334,
      "loss_d0": 0.0407680869102478,
      "step": 2280
    },
    {
      "epoch": 229.0,
      "grad_norm": 0.33984375,
      "learning_rate": 0.0002,
      "loss": 0.04180628061294556,
      "loss_d0": 0.041768335923552516,
      "step": 2290
    },
    {
      "epoch": 230.0,
      "grad_norm": 0.33984375,
      "learning_rate": 0.0002,
      "loss": 0.041827014088630675,
      "loss_d0": 0.04191113151609897,
      "step": 2300
    },
    {
      "epoch": 231.0,
      "grad_norm": 0.3125,
      "learning_rate": 0.0002,
      "loss": 0.038024112582206726,
      "loss_d0": 0.03836959525942803,
      "step": 2310
    },
    {
      "epoch": 232.0,
      "grad_norm": 0.28125,
      "learning_rate": 0.0002,
      "loss": 0.03627745509147644,
      "loss_d0": 0.036746613681316376,
      "step": 2320
    },
    {
      "epoch": 233.0,
      "grad_norm": 0.287109375,
      "learning_rate": 0.0002,
      "loss": 0.03757380247116089,
      "loss_d0": 0.0378836490213871,
      "step": 2330
    },
    {
      "epoch": 234.0,
      "grad_norm": 0.29296875,
      "learning_rate": 0.0002,
      "loss": 0.03801417648792267,
      "loss_d0": 0.038370263576507566,
      "step": 2340
    },
    {
      "epoch": 235.0,
      "grad_norm": 0.287109375,
      "learning_rate": 0.0002,
      "loss": 0.03692303895950318,
      "loss_d0": 0.03739718943834305,
      "step": 2350
    },
    {
      "epoch": 236.0,
      "grad_norm": 0.263671875,
      "learning_rate": 0.0002,
      "loss": 0.036152446269989015,
      "loss_d0": 0.03652026541531086,
      "step": 2360
    },
    {
      "epoch": 237.0,
      "grad_norm": 0.302734375,
      "learning_rate": 0.0002,
      "loss": 0.03825657963752747,
      "loss_d0": 0.03856762647628784,
      "step": 2370
    },
    {
      "epoch": 238.0,
      "grad_norm": 0.416015625,
      "learning_rate": 0.0002,
      "loss": 0.043493041396141054,
      "loss_d0": 0.04372312910854816,
      "step": 2380
    },
    {
      "epoch": 239.0,
      "grad_norm": 0.447265625,
      "learning_rate": 0.0002,
      "loss": 0.04036698043346405,
      "loss_d0": 0.04092743098735809,
      "step": 2390
    },
    {
      "epoch": 240.0,
      "grad_norm": 0.326171875,
      "learning_rate": 0.0002,
      "loss": 0.04108090996742249,
      "loss_d0": 0.041520928218960765,
      "step": 2400
    },
    {
      "epoch": 241.0,
      "grad_norm": 0.37890625,
      "learning_rate": 0.0002,
      "loss": 0.045140665769577024,
      "loss_d0": 0.04539525546133518,
      "step": 2410
    },
    {
      "epoch": 242.0,
      "grad_norm": 0.5546875,
      "learning_rate": 0.0002,
      "loss": 0.04377622902393341,
      "loss_d0": 0.044199855998158455,
      "step": 2420
    },
    {
      "epoch": 243.0,
      "grad_norm": 0.369140625,
      "learning_rate": 0.0002,
      "loss": 0.03707056045532227,
      "loss_d0": 0.03749646060168743,
      "step": 2430
    },
    {
      "epoch": 244.0,
      "grad_norm": 0.36328125,
      "learning_rate": 0.0002,
      "loss": 0.03947383463382721,
      "loss_d0": 0.03957220241427421,
      "step": 2440
    },
    {
      "epoch": 245.0,
      "grad_norm": 0.310546875,
      "learning_rate": 0.0002,
      "loss": 0.03732470870018005,
      "loss_d0": 0.03755390048027039,
      "step": 2450
    },
    {
      "epoch": 246.0,
      "grad_norm": 0.462890625,
      "learning_rate": 0.0002,
      "loss": 0.03801247477531433,
      "loss_d0": 0.038219067454338077,
      "step": 2460
    },
    {
      "epoch": 247.0,
      "grad_norm": 0.3515625,
      "learning_rate": 0.0002,
      "loss": 0.03672387003898621,
      "loss_d0": 0.036860080808401106,
      "step": 2470
    },
    {
      "epoch": 248.0,
      "grad_norm": 0.447265625,
      "learning_rate": 0.0002,
      "loss": 0.03827687501907349,
      "loss_d0": 0.03846933878958225,
      "step": 2480
    },
    {
      "epoch": 249.0,
      "grad_norm": 0.34375,
      "learning_rate": 0.0002,
      "loss": 0.038004118204116824,
      "loss_d0": 0.03824130594730377,
      "step": 2490
    },
    {
      "epoch": 250.0,
      "grad_norm": 0.36328125,
      "learning_rate": 0.0002,
      "loss": 0.037803095579147336,
      "loss_d0": 0.03784290738403797,
      "step": 2500
    },
    {
      "epoch": 250.0,
      "eval_loss": 9.545710563659668,
      "eval_runtime": 0.3459,
      "eval_samples_per_second": 722.648,
      "eval_steps_per_second": 72.265,
      "step": 2500
    },
    {
      "epoch": 251.0,
      "grad_norm": 0.3125,
      "learning_rate": 0.0002,
      "loss": 0.038292118906974794,
      "loss_d0": 0.038580729812383655,
      "step": 2510
    },
    {
      "epoch": 252.0,
      "grad_norm": 0.392578125,
      "learning_rate": 0.0002,
      "loss": 0.037233129143714905,
      "loss_d0": 0.0375215470790863,
      "step": 2520
    },
    {
      "epoch": 253.0,
      "grad_norm": 0.314453125,
      "learning_rate": 0.0002,
      "loss": 0.044102489948272705,
      "loss_d0": 0.04566294178366661,
      "step": 2530
    },
    {
      "epoch": 254.0,
      "grad_norm": 0.275390625,
      "learning_rate": 0.0002,
      "loss": 0.037907305359840396,
      "loss_d0": 0.03828296475112438,
      "step": 2540
    },
    {
      "epoch": 255.0,
      "grad_norm": 0.28125,
      "learning_rate": 0.0002,
      "loss": 0.04824570715427399,
      "loss_d0": 0.04789377823472023,
      "step": 2550
    },
    {
      "epoch": 256.0,
      "grad_norm": 0.380859375,
      "learning_rate": 0.0002,
      "loss": 0.04258241653442383,
      "loss_d0": 0.04261836186051369,
      "step": 2560
    },
    {
      "epoch": 257.0,
      "grad_norm": 0.34765625,
      "learning_rate": 0.0002,
      "loss": 0.0418660581111908,
      "loss_d0": 0.04167159348726272,
      "step": 2570
    },
    {
      "epoch": 258.0,
      "grad_norm": 0.310546875,
      "learning_rate": 0.0002,
      "loss": 0.04080710113048554,
      "loss_d0": 0.04094872139394283,
      "step": 2580
    },
    {
      "epoch": 259.0,
      "grad_norm": 0.52734375,
      "learning_rate": 0.0002,
      "loss": 0.041302931308746335,
      "loss_d0": 0.04151376113295555,
      "step": 2590
    },
    {
      "epoch": 260.0,
      "grad_norm": 0.25,
      "learning_rate": 0.0002,
      "loss": 0.040556898713111876,
      "loss_d0": 0.0407320111989975,
      "step": 2600
    },
    {
      "epoch": 261.0,
      "grad_norm": 0.39453125,
      "learning_rate": 0.0002,
      "loss": 0.041184079647064206,
      "loss_d0": 0.041263216733932497,
      "step": 2610
    },
    {
      "epoch": 262.0,
      "grad_norm": 0.4609375,
      "learning_rate": 0.0002,
      "loss": 0.03699159026145935,
      "loss_d0": 0.037303267419338225,
      "step": 2620
    },
    {
      "epoch": 263.0,
      "grad_norm": 0.44921875,
      "learning_rate": 0.0002,
      "loss": 0.03918337523937225,
      "loss_d0": 0.03950182273983956,
      "step": 2630
    },
    {
      "epoch": 264.0,
      "grad_norm": 0.310546875,
      "learning_rate": 0.0002,
      "loss": 0.03995798230171203,
      "loss_d0": 0.04008491151034832,
      "step": 2640
    },
    {
      "epoch": 265.0,
      "grad_norm": 0.396484375,
      "learning_rate": 0.0002,
      "loss": 0.041723164916038516,
      "loss_d0": 0.041936520114541055,
      "step": 2650
    },
    {
      "epoch": 266.0,
      "grad_norm": 0.369140625,
      "learning_rate": 0.0002,
      "loss": 0.04007298946380615,
      "loss_d0": 0.039936954900622365,
      "step": 2660
    },
    {
      "epoch": 267.0,
      "grad_norm": 0.609375,
      "learning_rate": 0.0002,
      "loss": 0.04925413727760315,
      "loss_d0": 0.048130762204527854,
      "step": 2670
    },
    {
      "epoch": 268.0,
      "grad_norm": 0.28125,
      "learning_rate": 0.0002,
      "loss": 0.03803152143955231,
      "loss_d0": 0.0383452296257019,
      "step": 2680
    },
    {
      "epoch": 269.0,
      "grad_norm": 0.240234375,
      "learning_rate": 0.0002,
      "loss": 0.036663559079170224,
      "loss_d0": 0.03692438825964928,
      "step": 2690
    },
    {
      "epoch": 270.0,
      "grad_norm": 0.248046875,
      "learning_rate": 0.0002,
      "loss": 0.035174107551574706,
      "loss_d0": 0.03558996543288231,
      "step": 2700
    },
    {
      "epoch": 271.0,
      "grad_norm": 0.2099609375,
      "learning_rate": 0.0002,
      "loss": 0.035706958174705504,
      "loss_d0": 0.03603263683617115,
      "step": 2710
    },
    {
      "epoch": 272.0,
      "grad_norm": 0.291015625,
      "learning_rate": 0.0002,
      "loss": 0.03466865718364716,
      "loss_d0": 0.03504250831902027,
      "step": 2720
    },
    {
      "epoch": 273.0,
      "grad_norm": 0.322265625,
      "learning_rate": 0.0002,
      "loss": 0.03593834340572357,
      "loss_d0": 0.036205967888236046,
      "step": 2730
    },
    {
      "epoch": 274.0,
      "grad_norm": 0.275390625,
      "learning_rate": 0.0002,
      "loss": 0.036024799942970274,
      "loss_d0": 0.03630461469292641,
      "step": 2740
    },
    {
      "epoch": 275.0,
      "grad_norm": 0.384765625,
      "learning_rate": 0.0002,
      "loss": 0.039151713252067566,
      "loss_d0": 0.039199193567037584,
      "step": 2750
    },
    {
      "epoch": 276.0,
      "grad_norm": 0.298828125,
      "learning_rate": 0.0002,
      "loss": 0.036157673597335814,
      "loss_d0": 0.036374952644109726,
      "step": 2760
    },
    {
      "epoch": 277.0,
      "grad_norm": 0.359375,
      "learning_rate": 0.0002,
      "loss": 0.035508188605308535,
      "loss_d0": 0.03584931716322899,
      "step": 2770
    },
    {
      "epoch": 278.0,
      "grad_norm": 0.2470703125,
      "learning_rate": 0.0002,
      "loss": 0.03636508285999298,
      "loss_d0": 0.03663929682224989,
      "step": 2780
    },
    {
      "epoch": 279.0,
      "grad_norm": 0.494140625,
      "learning_rate": 0.0002,
      "loss": 0.040090903639793396,
      "loss_d0": 0.04021787643432617,
      "step": 2790
    },
    {
      "epoch": 280.0,
      "grad_norm": 0.328125,
      "learning_rate": 0.0002,
      "loss": 0.04161113798618317,
      "loss_d0": 0.04178265742957592,
      "step": 2800
    },
    {
      "epoch": 281.0,
      "grad_norm": 0.322265625,
      "learning_rate": 0.0002,
      "loss": 0.041626724600791934,
      "loss_d0": 0.04190756045281887,
      "step": 2810
    },
    {
      "epoch": 282.0,
      "grad_norm": 0.3125,
      "learning_rate": 0.0002,
      "loss": 0.036975172162055966,
      "loss_d0": 0.0373986829072237,
      "step": 2820
    },
    {
      "epoch": 283.0,
      "grad_norm": 0.462890625,
      "learning_rate": 0.0002,
      "loss": 0.039800339937210084,
      "loss_d0": 0.04016715399920941,
      "step": 2830
    },
    {
      "epoch": 284.0,
      "grad_norm": 0.30859375,
      "learning_rate": 0.0002,
      "loss": 0.035621476173400876,
      "loss_d0": 0.03597722910344601,
      "step": 2840
    },
    {
      "epoch": 285.0,
      "grad_norm": 0.291015625,
      "learning_rate": 0.0002,
      "loss": 0.03560549020767212,
      "loss_d0": 0.03596907928586006,
      "step": 2850
    },
    {
      "epoch": 286.0,
      "grad_norm": 0.265625,
      "learning_rate": 0.0002,
      "loss": 0.03517293334007263,
      "loss_d0": 0.03547440096735954,
      "step": 2860
    },
    {
      "epoch": 287.0,
      "grad_norm": 0.37109375,
      "learning_rate": 0.0002,
      "loss": 0.0386651873588562,
      "loss_d0": 0.039219272881746294,
      "step": 2870
    },
    {
      "epoch": 288.0,
      "grad_norm": 0.30859375,
      "learning_rate": 0.0002,
      "loss": 0.03703486919403076,
      "loss_d0": 0.03765572048723698,
      "step": 2880
    },
    {
      "epoch": 289.0,
      "grad_norm": 0.2490234375,
      "learning_rate": 0.0002,
      "loss": 0.036672037839889524,
      "loss_d0": 0.037157762050628665,
      "step": 2890
    },
    {
      "epoch": 290.0,
      "grad_norm": 0.3203125,
      "learning_rate": 0.0002,
      "loss": 0.03655970096588135,
      "loss_d0": 0.03675979822874069,
      "step": 2900
    },
    {
      "epoch": 291.0,
      "grad_norm": 0.33203125,
      "learning_rate": 0.0002,
      "loss": 0.03832969963550568,
      "loss_d0": 0.039013080298900604,
      "step": 2910
    },
    {
      "epoch": 292.0,
      "grad_norm": 0.55859375,
      "learning_rate": 0.0002,
      "loss": 0.04422642886638641,
      "loss_d0": 0.04533644244074821,
      "step": 2920
    },
    {
      "epoch": 293.0,
      "grad_norm": 0.462890625,
      "learning_rate": 0.0002,
      "loss": 0.04290195107460022,
      "loss_d0": 0.04374877475202084,
      "step": 2930
    },
    {
      "epoch": 294.0,
      "grad_norm": 0.2890625,
      "learning_rate": 0.0002,
      "loss": 0.042818278074264526,
      "loss_d0": 0.04318895749747753,
      "step": 2940
    },
    {
      "epoch": 295.0,
      "grad_norm": 0.345703125,
      "learning_rate": 0.0002,
      "loss": 0.040858533978462216,
      "loss_d0": 0.040828976407647134,
      "step": 2950
    },
    {
      "epoch": 296.0,
      "grad_norm": 0.34375,
      "learning_rate": 0.0002,
      "loss": 0.04414619207382202,
      "loss_d0": 0.04415134936571121,
      "step": 2960
    },
    {
      "epoch": 297.0,
      "grad_norm": 0.396484375,
      "learning_rate": 0.0002,
      "loss": 0.03988617062568665,
      "loss_d0": 0.040220906212925914,
      "step": 2970
    },
    {
      "epoch": 298.0,
      "grad_norm": 0.40234375,
      "learning_rate": 0.0002,
      "loss": 0.03869087994098663,
      "loss_d0": 0.038922371342778206,
      "step": 2980
    },
    {
      "epoch": 299.0,
      "grad_norm": 0.34375,
      "learning_rate": 0.0002,
      "loss": 0.04153143167495728,
      "loss_d0": 0.041573012620210646,
      "step": 2990
    },
    {
      "epoch": 300.0,
      "grad_norm": 0.302734375,
      "learning_rate": 0.0002,
      "loss": 0.036962729692459104,
      "loss_d0": 0.037265943735837935,
      "step": 3000
    },
    {
      "epoch": 300.0,
      "eval_loss": 10.115313529968262,
      "eval_runtime": 0.3457,
      "eval_samples_per_second": 723.217,
      "eval_steps_per_second": 72.322,
      "step": 3000
    },
    {
      "epoch": 301.0,
      "grad_norm": 0.251953125,
      "learning_rate": 0.0002,
      "loss": 0.037201502919197084,
      "loss_d0": 0.037610159069299695,
      "step": 3010
    },
    {
      "epoch": 302.0,
      "grad_norm": 0.310546875,
      "learning_rate": 0.0002,
      "loss": 0.03739930093288422,
      "loss_d0": 0.0379462331533432,
      "step": 3020
    },
    {
      "epoch": 303.0,
      "grad_norm": 0.328125,
      "learning_rate": 0.0002,
      "loss": 0.03508531451225281,
      "loss_d0": 0.03542482852935791,
      "step": 3030
    },
    {
      "epoch": 304.0,
      "grad_norm": 0.365234375,
      "learning_rate": 0.0002,
      "loss": 0.034917259216308595,
      "loss_d0": 0.03526957035064697,
      "step": 3040
    },
    {
      "epoch": 305.0,
      "grad_norm": 0.275390625,
      "learning_rate": 0.0002,
      "loss": 0.035196778178215024,
      "loss_d0": 0.03563609905540943,
      "step": 3050
    },
    {
      "epoch": 306.0,
      "grad_norm": 0.30859375,
      "learning_rate": 0.0002,
      "loss": 0.03536902666091919,
      "loss_d0": 0.035680387169122696,
      "step": 3060
    },
    {
      "epoch": 307.0,
      "grad_norm": 0.25,
      "learning_rate": 0.0002,
      "loss": 0.03461025953292847,
      "loss_d0": 0.03496658094227314,
      "step": 3070
    },
    {
      "epoch": 308.0,
      "grad_norm": 0.302734375,
      "learning_rate": 0.0002,
      "loss": 0.03487989902496338,
      "loss_d0": 0.03524426631629467,
      "step": 3080
    },
    {
      "epoch": 309.0,
      "grad_norm": 0.359375,
      "learning_rate": 0.0002,
      "loss": 0.03497541546821594,
      "loss_d0": 0.03536835759878158,
      "step": 3090
    },
    {
      "epoch": 310.0,
      "grad_norm": 0.27734375,
      "learning_rate": 0.0002,
      "loss": 0.03510336279869079,
      "loss_d0": 0.035488955304026605,
      "step": 3100
    },
    {
      "epoch": 311.0,
      "grad_norm": 0.333984375,
      "learning_rate": 0.0002,
      "loss": 0.03497164249420166,
      "loss_d0": 0.03530625440180302,
      "step": 3110
    },
    {
      "epoch": 312.0,
      "grad_norm": 0.359375,
      "learning_rate": 0.0002,
      "loss": 0.03485918045043945,
      "loss_d0": 0.03521176762878895,
      "step": 3120
    },
    {
      "epoch": 313.0,
      "grad_norm": 0.29296875,
      "learning_rate": 0.0002,
      "loss": 0.03485134243965149,
      "loss_d0": 0.035247354209423064,
      "step": 3130
    },
    {
      "epoch": 314.0,
      "grad_norm": 0.310546875,
      "learning_rate": 0.0002,
      "loss": 0.03514372706413269,
      "loss_d0": 0.03544243313372135,
      "step": 3140
    },
    {
      "epoch": 315.0,
      "grad_norm": 0.318359375,
      "learning_rate": 0.0002,
      "loss": 0.034876453876495364,
      "loss_d0": 0.0352023221552372,
      "step": 3150
    },
    {
      "epoch": 316.0,
      "grad_norm": 0.33203125,
      "learning_rate": 0.0002,
      "loss": 0.03504140079021454,
      "loss_d0": 0.03542985171079636,
      "step": 3160
    },
    {
      "epoch": 317.0,
      "grad_norm": 0.44140625,
      "learning_rate": 0.0002,
      "loss": 0.03640889525413513,
      "loss_d0": 0.036533677205443385,
      "step": 3170
    },
    {
      "epoch": 318.0,
      "grad_norm": 0.318359375,
      "learning_rate": 0.0002,
      "loss": 0.036491012573242186,
      "loss_d0": 0.036695629730820654,
      "step": 3180
    },
    {
      "epoch": 319.0,
      "grad_norm": 0.267578125,
      "learning_rate": 0.0002,
      "loss": 0.03531668484210968,
      "loss_d0": 0.035679520666599275,
      "step": 3190
    },
    {
      "epoch": 320.0,
      "grad_norm": 0.30859375,
      "learning_rate": 0.0002,
      "loss": 0.03540306091308594,
      "loss_d0": 0.03574192002415657,
      "step": 3200
    },
    {
      "epoch": 321.0,
      "grad_norm": 0.27734375,
      "learning_rate": 0.0002,
      "loss": 0.03663922846317291,
      "loss_d0": 0.03695997782051563,
      "step": 3210
    },
    {
      "epoch": 322.0,
      "grad_norm": 0.291015625,
      "learning_rate": 0.0002,
      "loss": 0.037931612133979796,
      "loss_d0": 0.03835091814398765,
      "step": 3220
    },
    {
      "epoch": 323.0,
      "grad_norm": 0.3046875,
      "learning_rate": 0.0002,
      "loss": 0.03923901319503784,
      "loss_d0": 0.039649711921811104,
      "step": 3230
    },
    {
      "epoch": 324.0,
      "grad_norm": 0.357421875,
      "learning_rate": 0.0002,
      "loss": 0.04315422773361206,
      "loss_d0": 0.043376489728689197,
      "step": 3240
    },
    {
      "epoch": 325.0,
      "grad_norm": 0.390625,
      "learning_rate": 0.0002,
      "loss": 0.046264243125915525,
      "loss_d0": 0.04667617976665497,
      "step": 3250
    },
    {
      "epoch": 326.0,
      "grad_norm": 0.447265625,
      "learning_rate": 0.0002,
      "loss": 0.04075667560100556,
      "loss_d0": 0.041156957671046254,
      "step": 3260
    },
    {
      "epoch": 327.0,
      "grad_norm": 0.26953125,
      "learning_rate": 0.0002,
      "loss": 0.037901556491851805,
      "loss_d0": 0.03819670677185059,
      "step": 3270
    },
    {
      "epoch": 328.0,
      "grad_norm": 0.333984375,
      "learning_rate": 0.0002,
      "loss": 0.04078449308872223,
      "loss_d0": 0.04096781946718693,
      "step": 3280
    },
    {
      "epoch": 329.0,
      "grad_norm": 0.306640625,
      "learning_rate": 0.0002,
      "loss": 0.04341503381729126,
      "loss_d0": 0.044286961853504184,
      "step": 3290
    },
    {
      "epoch": 330.0,
      "grad_norm": 0.3515625,
      "learning_rate": 0.0002,
      "loss": 0.03675068318843842,
      "loss_d0": 0.03723678104579449,
      "step": 3300
    },
    {
      "epoch": 331.0,
      "grad_norm": 0.2490234375,
      "learning_rate": 0.0002,
      "loss": 0.04029195308685303,
      "loss_d0": 0.040922777354717256,
      "step": 3310
    },
    {
      "epoch": 332.0,
      "grad_norm": 0.3125,
      "learning_rate": 0.0002,
      "loss": 0.03685526847839356,
      "loss_d0": 0.037366196513175964,
      "step": 3320
    },
    {
      "epoch": 333.0,
      "grad_norm": 0.28125,
      "learning_rate": 0.0002,
      "loss": 0.038150754570960996,
      "loss_d0": 0.03864995911717415,
      "step": 3330
    },
    {
      "epoch": 334.0,
      "grad_norm": 0.259765625,
      "learning_rate": 0.0002,
      "loss": 0.038439178466796876,
      "loss_d0": 0.038616427779197694,
      "step": 3340
    },
    {
      "epoch": 335.0,
      "grad_norm": 0.53515625,
      "learning_rate": 0.0002,
      "loss": 0.040246385335922244,
      "loss_d0": 0.041175222024321556,
      "step": 3350
    },
    {
      "epoch": 336.0,
      "grad_norm": 0.3671875,
      "learning_rate": 0.0002,
      "loss": 0.04247841536998749,
      "loss_d0": 0.04267943426966667,
      "step": 3360
    },
    {
      "epoch": 337.0,
      "grad_norm": 1.046875,
      "learning_rate": 0.0002,
      "loss": 0.046033608913421634,
      "loss_d0": 0.04620472937822342,
      "step": 3370
    },
    {
      "epoch": 338.0,
      "grad_norm": 0.435546875,
      "learning_rate": 0.0002,
      "loss": 0.04217771589756012,
      "loss_d0": 0.04241746999323368,
      "step": 3380
    },
    {
      "epoch": 339.0,
      "grad_norm": 0.314453125,
      "learning_rate": 0.0002,
      "loss": 0.04194062054157257,
      "loss_d0": 0.042676547169685365,
      "step": 3390
    },
    {
      "epoch": 340.0,
      "grad_norm": 0.35546875,
      "learning_rate": 0.0002,
      "loss": 0.03749172687530518,
      "loss_d0": 0.03779852800071239,
      "step": 3400
    },
    {
      "epoch": 341.0,
      "grad_norm": 0.443359375,
      "learning_rate": 0.0002,
      "loss": 0.03817381858825684,
      "loss_d0": 0.038390916958451274,
      "step": 3410
    },
    {
      "epoch": 342.0,
      "grad_norm": 0.337890625,
      "learning_rate": 0.0002,
      "loss": 0.03953771889209747,
      "loss_d0": 0.03967275694012642,
      "step": 3420
    },
    {
      "epoch": 343.0,
      "grad_norm": 0.298828125,
      "learning_rate": 0.0002,
      "loss": 0.03631482720375061,
      "loss_d0": 0.03663896918296814,
      "step": 3430
    },
    {
      "epoch": 344.0,
      "grad_norm": 0.380859375,
      "learning_rate": 0.0002,
      "loss": 0.035741984844207764,
      "loss_d0": 0.036156880855560306,
      "step": 3440
    },
    {
      "epoch": 345.0,
      "grad_norm": 0.279296875,
      "learning_rate": 0.0002,
      "loss": 0.03471076786518097,
      "loss_d0": 0.035087333992123604,
      "step": 3450
    },
    {
      "epoch": 346.0,
      "grad_norm": 0.31640625,
      "learning_rate": 0.0002,
      "loss": 0.03551497161388397,
      "loss_d0": 0.035923638567328456,
      "step": 3460
    },
    {
      "epoch": 347.0,
      "grad_norm": 0.33203125,
      "learning_rate": 0.0002,
      "loss": 0.04135353863239288,
      "loss_d0": 0.04165848158299923,
      "step": 3470
    },
    {
      "epoch": 348.0,
      "grad_norm": 0.3515625,
      "learning_rate": 0.0002,
      "loss": 0.03566417694091797,
      "loss_d0": 0.03601394034922123,
      "step": 3480
    },
    {
      "epoch": 349.0,
      "grad_norm": 0.322265625,
      "learning_rate": 0.0002,
      "loss": 0.039245966076850894,
      "loss_d0": 0.039576159045100215,
      "step": 3490
    },
    {
      "epoch": 350.0,
      "grad_norm": 0.384765625,
      "learning_rate": 0.0002,
      "loss": 0.037160205841064456,
      "loss_d0": 0.037364447861909865,
      "step": 3500
    },
    {
      "epoch": 350.0,
      "eval_loss": 9.983804702758789,
      "eval_runtime": 0.3465,
      "eval_samples_per_second": 721.5,
      "eval_steps_per_second": 72.15,
      "step": 3500
    },
    {
      "epoch": 351.0,
      "grad_norm": 0.380859375,
      "learning_rate": 0.0002,
      "loss": 0.03493095934391022,
      "loss_d0": 0.035223822295665744,
      "step": 3510
    },
    {
      "epoch": 352.0,
      "grad_norm": 0.349609375,
      "learning_rate": 0.0002,
      "loss": 0.034974557161331174,
      "loss_d0": 0.03528392426669598,
      "step": 3520
    },
    {
      "epoch": 353.0,
      "grad_norm": 0.359375,
      "learning_rate": 0.0002,
      "loss": 0.03885301649570465,
      "loss_d0": 0.03938428051769734,
      "step": 3530
    },
    {
      "epoch": 354.0,
      "grad_norm": 0.328125,
      "learning_rate": 0.0002,
      "loss": 0.03921197950839996,
      "loss_d0": 0.03949593603610992,
      "step": 3540
    },
    {
      "epoch": 355.0,
      "grad_norm": 0.302734375,
      "learning_rate": 0.0002,
      "loss": 0.03750967383384705,
      "loss_d0": 0.037822453677654265,
      "step": 3550
    },
    {
      "epoch": 356.0,
      "grad_norm": 0.416015625,
      "learning_rate": 0.0002,
      "loss": 0.03778342306613922,
      "loss_d0": 0.03798616342246532,
      "step": 3560
    },
    {
      "epoch": 357.0,
      "grad_norm": 0.294921875,
      "learning_rate": 0.0002,
      "loss": 0.03592687249183655,
      "loss_d0": 0.036273398995399476,
      "step": 3570
    },
    {
      "epoch": 358.0,
      "grad_norm": 0.451171875,
      "learning_rate": 0.0002,
      "loss": 0.04027557075023651,
      "loss_d0": 0.040579644963145256,
      "step": 3580
    },
    {
      "epoch": 359.0,
      "grad_norm": 0.37890625,
      "learning_rate": 0.0002,
      "loss": 0.045068970322608946,
      "loss_d0": 0.04435764774680138,
      "step": 3590
    },
    {
      "epoch": 360.0,
      "grad_norm": 0.31640625,
      "learning_rate": 0.0002,
      "loss": 0.04428997039794922,
      "loss_d0": 0.04421007707715034,
      "step": 3600
    },
    {
      "epoch": 361.0,
      "grad_norm": 0.306640625,
      "learning_rate": 0.00019998164618097514,
      "loss": 0.039733481407165525,
      "loss_d0": 0.03993206955492497,
      "step": 3610
    },
    {
      "epoch": 362.0,
      "grad_norm": 0.33203125,
      "learning_rate": 0.0001999182104912672,
      "loss": 0.04485340416431427,
      "loss_d0": 0.04463536143302917,
      "step": 3620
    },
    {
      "epoch": 363.0,
      "grad_norm": 0.29296875,
      "learning_rate": 0.00019980949827469119,
      "loss": 0.04066751599311828,
      "loss_d0": 0.04052218459546566,
      "step": 3630
    },
    {
      "epoch": 364.0,
      "grad_norm": 0.3515625,
      "learning_rate": 0.00019965556427112201,
      "loss": 0.038836562633514406,
      "loss_d0": 0.03895148001611233,
      "step": 3640
    },
    {
      "epoch": 365.0,
      "grad_norm": 0.283203125,
      "learning_rate": 0.0001994564859909662,
      "loss": 0.03510969579219818,
      "loss_d0": 0.03547583632171154,
      "step": 3650
    },
    {
      "epoch": 366.0,
      "grad_norm": 0.275390625,
      "learning_rate": 0.000199212363676133,
      "loss": 0.03486792743206024,
      "loss_d0": 0.03522332645952701,
      "step": 3660
    },
    {
      "epoch": 367.0,
      "grad_norm": 0.357421875,
      "learning_rate": 0.00019892332024955975,
      "loss": 0.035192292928695676,
      "loss_d0": 0.035578233748674394,
      "step": 3670
    },
    {
      "epoch": 368.0,
      "grad_norm": 0.255859375,
      "learning_rate": 0.00019858950125331616,
      "loss": 0.03473028242588043,
      "loss_d0": 0.03515429347753525,
      "step": 3680
    },
    {
      "epoch": 369.0,
      "grad_norm": 0.27734375,
      "learning_rate": 0.00019821107477531995,
      "loss": 0.03524012267589569,
      "loss_d0": 0.03551997728645802,
      "step": 3690
    },
    {
      "epoch": 370.0,
      "grad_norm": 0.3984375,
      "learning_rate": 0.00019778823136469895,
      "loss": 0.036434537172317503,
      "loss_d0": 0.036938946694135666,
      "step": 3700
    },
    {
      "epoch": 371.0,
      "grad_norm": 0.361328125,
      "learning_rate": 0.0001973211839358444,
      "loss": 0.03875879943370819,
      "loss_d0": 0.039299290254712106,
      "step": 3710
    },
    {
      "epoch": 372.0,
      "grad_norm": 0.310546875,
      "learning_rate": 0.00019681016766120182,
      "loss": 0.040684208273887634,
      "loss_d0": 0.04135290943086147,
      "step": 3720
    },
    {
      "epoch": 373.0,
      "grad_norm": 0.29296875,
      "learning_rate": 0.00019625543985285482,
      "loss": 0.038448911905288694,
      "loss_d0": 0.03910197205841541,
      "step": 3730
    },
    {
      "epoch": 374.0,
      "grad_norm": 0.44140625,
      "learning_rate": 0.00019565727983296045,
      "loss": 0.043462514877319336,
      "loss_d0": 0.0441141027957201,
      "step": 3740
    },
    {
      "epoch": 375.0,
      "grad_norm": 0.408203125,
      "learning_rate": 0.00019501598879310232,
      "loss": 0.039922356605529785,
      "loss_d0": 0.04066242426633835,
      "step": 3750
    },
    {
      "epoch": 376.0,
      "grad_norm": 0.271484375,
      "learning_rate": 0.0001943318896426315,
      "loss": 0.03755326569080353,
      "loss_d0": 0.03808530755341053,
      "step": 3760
    },
    {
      "epoch": 377.0,
      "grad_norm": 0.353515625,
      "learning_rate": 0.00019360532684607184,
      "loss": 0.03615553677082062,
      "loss_d0": 0.036697663366794586,
      "step": 3770
    },
    {
      "epoch": 378.0,
      "grad_norm": 0.431640625,
      "learning_rate": 0.00019283666624967188,
      "loss": 0.034849488735198976,
      "loss_d0": 0.035189850628376006,
      "step": 3780
    },
    {
      "epoch": 379.0,
      "grad_norm": 0.279296875,
      "learning_rate": 0.00019202629489719006,
      "loss": 0.034756675362586975,
      "loss_d0": 0.03515330813825131,
      "step": 3790
    },
    {
      "epoch": 380.0,
      "grad_norm": 0.34765625,
      "learning_rate": 0.00019117462083500676,
      "loss": 0.03513258397579193,
      "loss_d0": 0.035471546649932864,
      "step": 3800
    },
    {
      "epoch": 381.0,
      "grad_norm": 0.279296875,
      "learning_rate": 0.00019028207290666037,
      "loss": 0.034799987077713014,
      "loss_d0": 0.035196726024150846,
      "step": 3810
    },
    {
      "epoch": 382.0,
      "grad_norm": 0.3671875,
      "learning_rate": 0.00018934910053691195,
      "loss": 0.03699601292610168,
      "loss_d0": 0.037290577217936514,
      "step": 3820
    },
    {
      "epoch": 383.0,
      "grad_norm": 0.2890625,
      "learning_rate": 0.00018837617350544604,
      "loss": 0.03611961305141449,
      "loss_d0": 0.03667362257838249,
      "step": 3830
    },
    {
      "epoch": 384.0,
      "grad_norm": 0.318359375,
      "learning_rate": 0.00018736378171032268,
      "loss": 0.035090076923370364,
      "loss_d0": 0.03538455814123154,
      "step": 3840
    },
    {
      "epoch": 385.0,
      "grad_norm": 0.30859375,
      "learning_rate": 0.00018631243492129893,
      "loss": 0.0343282163143158,
      "loss_d0": 0.0347426351159811,
      "step": 3850
    },
    {
      "epoch": 386.0,
      "grad_norm": 0.2890625,
      "learning_rate": 0.00018522266252314433,
      "loss": 0.03467389941215515,
      "loss_d0": 0.035047439858317375,
      "step": 3860
    },
    {
      "epoch": 387.0,
      "grad_norm": 0.265625,
      "learning_rate": 0.00018409501324907984,
      "loss": 0.03439282476902008,
      "loss_d0": 0.034719069674611094,
      "step": 3870
    },
    {
      "epoch": 388.0,
      "grad_norm": 0.2578125,
      "learning_rate": 0.0001829300549044741,
      "loss": 0.034527868032455444,
      "loss_d0": 0.03482235707342625,
      "step": 3880
    },
    {
      "epoch": 389.0,
      "grad_norm": 0.2734375,
      "learning_rate": 0.00018172837408093626,
      "loss": 0.03430909514427185,
      "loss_d0": 0.03469268716871739,
      "step": 3890
    },
    {
      "epoch": 390.0,
      "grad_norm": 0.255859375,
      "learning_rate": 0.00018049057586094967,
      "loss": 0.03444467484951019,
      "loss_d0": 0.03479441031813622,
      "step": 3900
    },
    {
      "epoch": 391.0,
      "grad_norm": 0.306640625,
      "learning_rate": 0.0001792172835131943,
      "loss": 0.03463874459266662,
      "loss_d0": 0.03487694039940834,
      "step": 3910
    },
    {
      "epoch": 392.0,
      "grad_norm": 0.2490234375,
      "learning_rate": 0.00017790913817871247,
      "loss": 0.034380468726158145,
      "loss_d0": 0.03472727090120316,
      "step": 3920
    },
    {
      "epoch": 393.0,
      "grad_norm": 0.3046875,
      "learning_rate": 0.00017656679854807488,
      "loss": 0.03448335528373718,
      "loss_d0": 0.034803297743201254,
      "step": 3930
    },
    {
      "epoch": 394.0,
      "grad_norm": 0.3984375,
      "learning_rate": 0.00017519094052971006,
      "loss": 0.03478938937187195,
      "loss_d0": 0.03513539023697376,
      "step": 3940
    },
    {
      "epoch": 395.0,
      "grad_norm": 0.44140625,
      "learning_rate": 0.0001737822569095642,
      "loss": 0.03497895896434784,
      "loss_d0": 0.03534998074173927,
      "step": 3950
    },
    {
      "epoch": 396.0,
      "grad_norm": 0.2470703125,
      "learning_rate": 0.00017234145700226254,
      "loss": 0.034469103813171385,
      "loss_d0": 0.03478188551962376,
      "step": 3960
    },
    {
      "epoch": 397.0,
      "grad_norm": 0.322265625,
      "learning_rate": 0.0001708692662939481,
      "loss": 0.03481238782405853,
      "loss_d0": 0.03517717607319355,
      "step": 3970
    },
    {
      "epoch": 398.0,
      "grad_norm": 0.271484375,
      "learning_rate": 0.0001693664260769775,
      "loss": 0.03424340188503265,
      "loss_d0": 0.0345684889703989,
      "step": 3980
    },
    {
      "epoch": 399.0,
      "grad_norm": 0.328125,
      "learning_rate": 0.00016783369307665796,
      "loss": 0.03449050784111023,
      "loss_d0": 0.034836267307400706,
      "step": 3990
    },
    {
      "epoch": 400.0,
      "grad_norm": 0.328125,
      "learning_rate": 0.0001662718390702135,
      "loss": 0.034349840879440305,
      "loss_d0": 0.03471621163189411,
      "step": 4000
    },
    {
      "epoch": 400.0,
      "eval_loss": 10.383418083190918,
      "eval_runtime": 0.3452,
      "eval_samples_per_second": 724.235,
      "eval_steps_per_second": 72.423,
      "step": 4000
    },
    {
      "epoch": 401.0,
      "grad_norm": 0.283203125,
      "learning_rate": 0.0001646816504981714,
      "loss": 0.03403621315956116,
      "loss_d0": 0.03437307737767696,
      "step": 4010
    },
    {
      "epoch": 402.0,
      "grad_norm": 0.232421875,
      "learning_rate": 0.00016306392806836635,
      "loss": 0.03419321179389954,
      "loss_d0": 0.03459935560822487,
      "step": 4020
    },
    {
      "epoch": 403.0,
      "grad_norm": 0.37890625,
      "learning_rate": 0.00016141948635275936,
      "loss": 0.03453047871589661,
      "loss_d0": 0.03483349122107029,
      "step": 4030
    },
    {
      "epoch": 404.0,
      "grad_norm": 0.2294921875,
      "learning_rate": 0.0001597491533772766,
      "loss": 0.0340300053358078,
      "loss_d0": 0.03438904657959938,
      "step": 4040
    },
    {
      "epoch": 405.0,
      "grad_norm": 0.28125,
      "learning_rate": 0.00015805377020487328,
      "loss": 0.03427405655384064,
      "loss_d0": 0.034636927768588066,
      "step": 4050
    },
    {
      "epoch": 406.0,
      "grad_norm": 0.337890625,
      "learning_rate": 0.00015633419051203315,
      "loss": 0.034300050139427184,
      "loss_d0": 0.03467072732746601,
      "step": 4060
    },
    {
      "epoch": 407.0,
      "grad_norm": 0.2255859375,
      "learning_rate": 0.00015459128015891672,
      "loss": 0.03416513502597809,
      "loss_d0": 0.03450297638773918,
      "step": 4070
    },
    {
      "epoch": 408.0,
      "grad_norm": 0.27734375,
      "learning_rate": 0.00015282591675337437,
      "loss": 0.03396352529525757,
      "loss_d0": 0.03434969037771225,
      "step": 4080
    },
    {
      "epoch": 409.0,
      "grad_norm": 0.3515625,
      "learning_rate": 0.00015103898920904485,
      "loss": 0.03412180244922638,
      "loss_d0": 0.03446668200194836,
      "step": 4090
    },
    {
      "epoch": 410.0,
      "grad_norm": 0.291015625,
      "learning_rate": 0.00014923139729775993,
      "loss": 0.03411206305027008,
      "loss_d0": 0.034487524256110194,
      "step": 4100
    },
    {
      "epoch": 411.0,
      "grad_norm": 0.3125,
      "learning_rate": 0.00014740405119648256,
      "loss": 0.03426051139831543,
      "loss_d0": 0.03457261025905609,
      "step": 4110
    },
    {
      "epoch": 412.0,
      "grad_norm": 0.35546875,
      "learning_rate": 0.00014555787102900526,
      "loss": 0.0340737909078598,
      "loss_d0": 0.03446582928299904,
      "step": 4120
    },
    {
      "epoch": 413.0,
      "grad_norm": 0.2490234375,
      "learning_rate": 0.00014369378640264005,
      "loss": 0.033984017372131345,
      "loss_d0": 0.03432649001479149,
      "step": 4130
    },
    {
      "epoch": 414.0,
      "grad_norm": 0.396484375,
      "learning_rate": 0.00014181273594013316,
      "loss": 0.03396552503108978,
      "loss_d0": 0.034310311824083325,
      "step": 4140
    },
    {
      "epoch": 415.0,
      "grad_norm": 0.33203125,
      "learning_rate": 0.00013991566680704028,
      "loss": 0.03403688073158264,
      "loss_d0": 0.03440046086907387,
      "step": 4150
    },
    {
      "epoch": 416.0,
      "grad_norm": 0.25,
      "learning_rate": 0.00013800353423480018,
      "loss": 0.03403026461601257,
      "loss_d0": 0.03437587544322014,
      "step": 4160
    },
    {
      "epoch": 417.0,
      "grad_norm": 0.27734375,
      "learning_rate": 0.00013607730103974675,
      "loss": 0.03386533558368683,
      "loss_d0": 0.03420776091516018,
      "step": 4170
    },
    {
      "epoch": 418.0,
      "grad_norm": 0.255859375,
      "learning_rate": 0.00013413793713830226,
      "loss": 0.03379028737545013,
      "loss_d0": 0.034149153530597685,
      "step": 4180
    },
    {
      "epoch": 419.0,
      "grad_norm": 0.25390625,
      "learning_rate": 0.00013218641905859497,
      "loss": 0.03388975858688355,
      "loss_d0": 0.034247222170233724,
      "step": 4190
    },
    {
      "epoch": 420.0,
      "grad_norm": 0.306640625,
      "learning_rate": 0.00013022372944874794,
      "loss": 0.033803710341453554,
      "loss_d0": 0.03417713083326816,
      "step": 4200
    },
    {
      "epoch": 421.0,
      "grad_norm": 0.271484375,
      "learning_rate": 0.00012825085658208613,
      "loss": 0.03380556702613831,
      "loss_d0": 0.034181630611419676,
      "step": 4210
    },
    {
      "epoch": 422.0,
      "grad_norm": 0.326171875,
      "learning_rate": 0.00012626879385951094,
      "loss": 0.033895039558410646,
      "loss_d0": 0.03424294553697109,
      "step": 4220
    },
    {
      "epoch": 423.0,
      "grad_norm": 0.2578125,
      "learning_rate": 0.00012427853930929296,
      "loss": 0.03377222716808319,
      "loss_d0": 0.03410911038517952,
      "step": 4230
    },
    {
      "epoch": 424.0,
      "grad_norm": 0.34375,
      "learning_rate": 0.00012228109508453477,
      "loss": 0.033803755044937135,
      "loss_d0": 0.03419635631144047,
      "step": 4240
    },
    {
      "epoch": 425.0,
      "grad_norm": 0.3046875,
      "learning_rate": 0.00012027746695855644,
      "loss": 0.033851489424705505,
      "loss_d0": 0.034207236021757126,
      "step": 4250
    },
    {
      "epoch": 426.0,
      "grad_norm": 0.306640625,
      "learning_rate": 0.00011826866381845871,
      "loss": 0.03399422466754913,
      "loss_d0": 0.03440755046904087,
      "step": 4260
    },
    {
      "epoch": 427.0,
      "grad_norm": 0.2197265625,
      "learning_rate": 0.00011625569715711771,
      "loss": 0.033890485763549805,
      "loss_d0": 0.03422976918518543,
      "step": 4270
    },
    {
      "epoch": 428.0,
      "grad_norm": 0.27734375,
      "learning_rate": 0.00011423958056386783,
      "loss": 0.03372063934803009,
      "loss_d0": 0.03409699872136116,
      "step": 4280
    },
    {
      "epoch": 429.0,
      "grad_norm": 0.2734375,
      "learning_rate": 0.00011222132921412918,
      "loss": 0.03375758528709412,
      "loss_d0": 0.03410529233515262,
      "step": 4290
    },
    {
      "epoch": 430.0,
      "grad_norm": 0.265625,
      "learning_rate": 0.00011020195935823588,
      "loss": 0.033726030588150026,
      "loss_d0": 0.03407609239220619,
      "step": 4300
    },
    {
      "epoch": 431.0,
      "grad_norm": 0.265625,
      "learning_rate": 0.00010818248780972371,
      "loss": 0.03368253111839294,
      "loss_d0": 0.03405856639146805,
      "step": 4310
    },
    {
      "epoch": 432.0,
      "grad_norm": 0.248046875,
      "learning_rate": 0.0001061639314333337,
      "loss": 0.03376383483409882,
      "loss_d0": 0.03409139961004257,
      "step": 4320
    },
    {
      "epoch": 433.0,
      "grad_norm": 0.271484375,
      "learning_rate": 0.00010414730663299018,
      "loss": 0.033721867203712466,
      "loss_d0": 0.0340782780200243,
      "step": 4330
    },
    {
      "epoch": 434.0,
      "grad_norm": 0.298828125,
      "learning_rate": 0.00010213362884001068,
      "loss": 0.03363073468208313,
      "loss_d0": 0.034021326526999476,
      "step": 4340
    },
    {
      "epoch": 435.0,
      "grad_norm": 0.26953125,
      "learning_rate": 0.00010012391200180593,
      "loss": 0.03361994624137878,
      "loss_d0": 0.033954566717147826,
      "step": 4350
    },
    {
      "epoch": 436.0,
      "grad_norm": 0.28515625,
      "learning_rate": 9.811916807132654e-05,
      "loss": 0.03361427187919617,
      "loss_d0": 0.033991168811917306,
      "step": 4360
    },
    {
      "epoch": 437.0,
      "grad_norm": 0.333984375,
      "learning_rate": 9.612040649751434e-05,
      "loss": 0.03355664610862732,
      "loss_d0": 0.03390999883413315,
      "step": 4370
    },
    {
      "epoch": 438.0,
      "grad_norm": 0.353515625,
      "learning_rate": 9.412863371701439e-05,
      "loss": 0.033678284287452696,
      "loss_d0": 0.03404993377625942,
      "step": 4380
    },
    {
      "epoch": 439.0,
      "grad_norm": 0.3203125,
      "learning_rate": 9.214485264740393e-05,
      "loss": 0.033527278900146486,
      "loss_d0": 0.03388219699263573,
      "step": 4390
    },
    {
      "epoch": 440.0,
      "grad_norm": 0.34765625,
      "learning_rate": 9.017006218219303e-05,
      "loss": 0.03374922573566437,
      "loss_d0": 0.03406397998332977,
      "step": 4400
    },
    {
      "epoch": 441.0,
      "grad_norm": 0.259765625,
      "learning_rate": 8.82052566878517e-05,
      "loss": 0.033463144302368165,
      "loss_d0": 0.03382991291582584,
      "step": 4410
    },
    {
      "epoch": 442.0,
      "grad_norm": 0.287109375,
      "learning_rate": 8.625142550311645e-05,
      "loss": 0.03350736498832703,
      "loss_d0": 0.033871978893876074,
      "step": 4420
    },
    {
      "epoch": 443.0,
      "grad_norm": 0.279296875,
      "learning_rate": 8.430955244082839e-05,
      "loss": 0.03351675570011139,
      "loss_d0": 0.033879267796874044,
      "step": 4430
    },
    {
      "epoch": 444.0,
      "grad_norm": 0.28125,
      "learning_rate": 8.238061529255388e-05,
      "loss": 0.03351102471351623,
      "loss_d0": 0.0338929008692503,
      "step": 4440
    },
    {
      "epoch": 445.0,
      "grad_norm": 0.296875,
      "learning_rate": 8.046558533623694e-05,
      "loss": 0.033553725481033324,
      "loss_d0": 0.03388892859220505,
      "step": 4450
    },
    {
      "epoch": 446.0,
      "grad_norm": 0.2392578125,
      "learning_rate": 7.856542684713174e-05,
      "loss": 0.033469372987747194,
      "loss_d0": 0.03383687436580658,
      "step": 4460
    },
    {
      "epoch": 447.0,
      "grad_norm": 0.326171875,
      "learning_rate": 7.668109661226076e-05,
      "loss": 0.03343449532985687,
      "loss_d0": 0.03379834964871407,
      "step": 4470
    },
    {
      "epoch": 448.0,
      "grad_norm": 0.265625,
      "learning_rate": 7.481354344864386e-05,
      "loss": 0.03345822095870972,
      "loss_d0": 0.03380800560116768,
      "step": 4480
    },
    {
      "epoch": 449.0,
      "grad_norm": 0.3046875,
      "learning_rate": 7.296370772554021e-05,
      "loss": 0.03347650170326233,
      "loss_d0": 0.03381855022162199,
      "step": 4490
    },
    {
      "epoch": 450.0,
      "grad_norm": 0.244140625,
      "learning_rate": 7.113252089094421e-05,
      "loss": 0.033470362424850464,
      "loss_d0": 0.033837277069687845,
      "step": 4500
    },
    {
      "epoch": 450.0,
      "eval_loss": 11.09699821472168,
      "eval_runtime": 0.3456,
      "eval_samples_per_second": 723.471,
      "eval_steps_per_second": 72.347,
      "step": 4500
    },
    {
      "epoch": 451.0,
      "grad_norm": 0.279296875,
      "learning_rate": 6.932090500257327e-05,
      "loss": 0.03349886238574982,
      "loss_d0": 0.03383576162159443,
      "step": 4510
    },
    {
      "epoch": 452.0,
      "grad_norm": 0.283203125,
      "learning_rate": 6.752977226358397e-05,
      "loss": 0.03333920836448669,
      "loss_d0": 0.033707871288061145,
      "step": 4520
    },
    {
      "epoch": 453.0,
      "grad_norm": 0.271484375,
      "learning_rate": 6.576002456325026e-05,
      "loss": 0.03335910141468048,
      "loss_d0": 0.033707983791828156,
      "step": 4530
    },
    {
      "epoch": 454.0,
      "grad_norm": 0.29296875,
      "learning_rate": 6.40125530228352e-05,
      "loss": 0.033454349637031554,
      "loss_d0": 0.03382563963532448,
      "step": 4540
    },
    {
      "epoch": 455.0,
      "grad_norm": 0.267578125,
      "learning_rate": 6.228823754688459e-05,
      "loss": 0.033449777960777284,
      "loss_d0": 0.03374672494828701,
      "step": 4550
    },
    {
      "epoch": 456.0,
      "grad_norm": 0.306640625,
      "learning_rate": 6.058794638016817e-05,
      "loss": 0.033386945724487305,
      "loss_d0": 0.03375318460166454,
      "step": 4560
    },
    {
      "epoch": 457.0,
      "grad_norm": 0.279296875,
      "learning_rate": 5.8912535670492685e-05,
      "loss": 0.03335491418838501,
      "loss_d0": 0.0337275929749012,
      "step": 4570
    },
    {
      "epoch": 458.0,
      "grad_norm": 0.291015625,
      "learning_rate": 5.7262849037605394e-05,
      "loss": 0.033367255330085756,
      "loss_d0": 0.0337124053388834,
      "step": 4580
    },
    {
      "epoch": 459.0,
      "grad_norm": 0.2890625,
      "learning_rate": 5.5639717148406334e-05,
      "loss": 0.03335196375846863,
      "loss_d0": 0.03371399827301502,
      "step": 4590
    },
    {
      "epoch": 460.0,
      "grad_norm": 0.265625,
      "learning_rate": 5.404395729868224e-05,
      "loss": 0.033342966437339784,
      "loss_d0": 0.03371122777462006,
      "step": 4600
    },
    {
      "epoch": 461.0,
      "grad_norm": 0.296875,
      "learning_rate": 5.247637300157372e-05,
      "loss": 0.03331690728664398,
      "loss_d0": 0.03370426036417484,
      "step": 4610
    },
    {
      "epoch": 462.0,
      "grad_norm": 0.28515625,
      "learning_rate": 5.0937753582982186e-05,
      "loss": 0.033305788040161134,
      "loss_d0": 0.033711410313844684,
      "step": 4620
    },
    {
      "epoch": 463.0,
      "grad_norm": 0.341796875,
      "learning_rate": 4.9428873784120425e-05,
      "loss": 0.033309575915336606,
      "loss_d0": 0.03368793986737728,
      "step": 4630
    },
    {
      "epoch": 464.0,
      "grad_norm": 0.3359375,
      "learning_rate": 4.7950493371407e-05,
      "loss": 0.03336468935012817,
      "loss_d0": 0.033730581775307654,
      "step": 4640
    },
    {
      "epoch": 465.0,
      "grad_norm": 0.25,
      "learning_rate": 4.650335675390113e-05,
      "loss": 0.033278393745422366,
      "loss_d0": 0.033654334396123885,
      "step": 4650
    },
    {
      "epoch": 466.0,
      "grad_norm": 0.220703125,
      "learning_rate": 4.508819260847027e-05,
      "loss": 0.033287009596824645,
      "loss_d0": 0.03361189253628254,
      "step": 4660
    },
    {
      "epoch": 467.0,
      "grad_norm": 0.2470703125,
      "learning_rate": 4.37057135128793e-05,
      "loss": 0.03328198194503784,
      "loss_d0": 0.03362148329615593,
      "step": 4670
    },
    {
      "epoch": 468.0,
      "grad_norm": 0.275390625,
      "learning_rate": 4.235661558698614e-05,
      "loss": 0.033373472094535825,
      "loss_d0": 0.033687862753868106,
      "step": 4680
    },
    {
      "epoch": 469.0,
      "grad_norm": 0.275390625,
      "learning_rate": 4.104157814222454e-05,
      "loss": 0.0332866907119751,
      "loss_d0": 0.033633769303560254,
      "step": 4690
    },
    {
      "epoch": 470.0,
      "grad_norm": 0.33984375,
      "learning_rate": 3.976126333955032e-05,
      "loss": 0.033249574899673465,
      "loss_d0": 0.03360855244100094,
      "step": 4700
    },
    {
      "epoch": 471.0,
      "grad_norm": 0.28125,
      "learning_rate": 3.8516315856023284e-05,
      "loss": 0.03323942124843597,
      "loss_d0": 0.033591284230351445,
      "step": 4710
    },
    {
      "epoch": 472.0,
      "grad_norm": 0.244140625,
      "learning_rate": 3.7307362560192935e-05,
      "loss": 0.03331650197505951,
      "loss_d0": 0.03367946818470955,
      "step": 4720
    },
    {
      "epoch": 473.0,
      "grad_norm": 0.2578125,
      "learning_rate": 3.613501219645132e-05,
      "loss": 0.03328208923339844,
      "loss_d0": 0.03366287276148796,
      "step": 4730
    },
    {
      "epoch": 474.0,
      "grad_norm": 0.302734375,
      "learning_rate": 3.4999855078511965e-05,
      "loss": 0.03330889940261841,
      "loss_d0": 0.03368374221026897,
      "step": 4740
    },
    {
      "epoch": 475.0,
      "grad_norm": 0.30078125,
      "learning_rate": 3.3902462792168955e-05,
      "loss": 0.033204281330108644,
      "loss_d0": 0.033571257069706914,
      "step": 4750
    },
    {
      "epoch": 476.0,
      "grad_norm": 0.259765625,
      "learning_rate": 3.284338790748641e-05,
      "loss": 0.03327343761920929,
      "loss_d0": 0.033621998876333235,
      "step": 4760
    },
    {
      "epoch": 477.0,
      "grad_norm": 0.291015625,
      "learning_rate": 3.1823163700562786e-05,
      "loss": 0.033239158987998965,
      "loss_d0": 0.03362444527447224,
      "step": 4770
    },
    {
      "epoch": 478.0,
      "grad_norm": 0.2314453125,
      "learning_rate": 3.084230388501036e-05,
      "loss": 0.033210772275924685,
      "loss_d0": 0.03361221067607403,
      "step": 4780
    },
    {
      "epoch": 479.0,
      "grad_norm": 0.2490234375,
      "learning_rate": 2.9901302353284794e-05,
      "loss": 0.03327825665473938,
      "loss_d0": 0.03364139422774315,
      "step": 4790
    },
    {
      "epoch": 480.0,
      "grad_norm": 0.265625,
      "learning_rate": 2.9000632927995554e-05,
      "loss": 0.03325657844543457,
      "loss_d0": 0.03365279622375965,
      "step": 4800
    },
    {
      "epoch": 481.0,
      "grad_norm": 0.2392578125,
      "learning_rate": 2.814074912332197e-05,
      "loss": 0.03324391841888428,
      "loss_d0": 0.03360977917909622,
      "step": 4810
    },
    {
      "epoch": 482.0,
      "grad_norm": 0.328125,
      "learning_rate": 2.732208391665507e-05,
      "loss": 0.0332427978515625,
      "loss_d0": 0.03361234851181507,
      "step": 4820
    },
    {
      "epoch": 483.0,
      "grad_norm": 0.294921875,
      "learning_rate": 2.6545049530580464e-05,
      "loss": 0.033264780044555665,
      "loss_d0": 0.03362235967069864,
      "step": 4830
    },
    {
      "epoch": 484.0,
      "grad_norm": 0.283203125,
      "learning_rate": 2.5810037225311933e-05,
      "loss": 0.033229345083236696,
      "loss_d0": 0.033553915470838545,
      "step": 4840
    },
    {
      "epoch": 485.0,
      "grad_norm": 0.35546875,
      "learning_rate": 2.5117417101680106e-05,
      "loss": 0.03322125971317291,
      "loss_d0": 0.03357453048229218,
      "step": 4850
    },
    {
      "epoch": 486.0,
      "grad_norm": 0.27734375,
      "learning_rate": 2.446753791477545e-05,
      "loss": 0.033195909857749936,
      "loss_d0": 0.0335524931550026,
      "step": 4860
    },
    {
      "epoch": 487.0,
      "grad_norm": 0.259765625,
      "learning_rate": 2.386072689833968e-05,
      "loss": 0.03325251340866089,
      "loss_d0": 0.03359146863222122,
      "step": 4870
    },
    {
      "epoch": 488.0,
      "grad_norm": 0.25,
      "learning_rate": 2.3297289599993645e-05,
      "loss": 0.033237877488136294,
      "loss_d0": 0.03359217718243599,
      "step": 4880
    },
    {
      "epoch": 489.0,
      "grad_norm": 0.267578125,
      "learning_rate": 2.277750972738503e-05,
      "loss": 0.03322227001190185,
      "loss_d0": 0.03359748162329197,
      "step": 4890
    },
    {
      "epoch": 490.0,
      "grad_norm": 0.35546875,
      "learning_rate": 2.2301649005332805e-05,
      "loss": 0.03325247764587402,
      "loss_d0": 0.033624636754393575,
      "step": 4900
    },
    {
      "epoch": 491.0,
      "grad_norm": 0.26171875,
      "learning_rate": 2.1869947044041086e-05,
      "loss": 0.03326715230941772,
      "loss_d0": 0.03362556584179401,
      "step": 4910
    },
    {
      "epoch": 492.0,
      "grad_norm": 0.244140625,
      "learning_rate": 2.148262121844807e-05,
      "loss": 0.03318788707256317,
      "loss_d0": 0.033572882786393164,
      "step": 4920
    },
    {
      "epoch": 493.0,
      "grad_norm": 0.29296875,
      "learning_rate": 2.113986655877128e-05,
      "loss": 0.03322984576225281,
      "loss_d0": 0.03360966518521309,
      "step": 4930
    },
    {
      "epoch": 494.0,
      "grad_norm": 0.2373046875,
      "learning_rate": 2.0841855652303886e-05,
      "loss": 0.03327394425868988,
      "loss_d0": 0.03363410085439682,
      "step": 4940
    },
    {
      "epoch": 495.0,
      "grad_norm": 0.29296875,
      "learning_rate": 2.0588738556511914e-05,
      "loss": 0.03323635458946228,
      "loss_d0": 0.03359650447964668,
      "step": 4950
    },
    {
      "epoch": 496.0,
      "grad_norm": 0.314453125,
      "learning_rate": 2.0380642723475683e-05,
      "loss": 0.033234477043151855,
      "loss_d0": 0.033594860881567004,
      "step": 4960
    },
    {
      "epoch": 497.0,
      "grad_norm": 0.234375,
      "learning_rate": 2.0217672935713957e-05,
      "loss": 0.033152416348457336,
      "loss_d0": 0.03353768773376942,
      "step": 4970
    },
    {
      "epoch": 498.0,
      "grad_norm": 0.31640625,
      "learning_rate": 2.009991125342271e-05,
      "loss": 0.03321554660797119,
      "loss_d0": 0.033586297556757926,
      "step": 4980
    },
    {
      "epoch": 499.0,
      "grad_norm": 0.298828125,
      "learning_rate": 2.002741697315534e-05,
      "loss": 0.03322638869285584,
      "loss_d0": 0.03359231799840927,
      "step": 4990
    },
    {
      "epoch": 500.0,
      "grad_norm": 0.3359375,
      "learning_rate": 2.0000226597965142e-05,
      "loss": 0.03330298364162445,
      "loss_d0": 0.03363474495708942,
      "step": 5000
    },
    {
      "epoch": 500.0,
      "eval_loss": 11.294451713562012,
      "eval_runtime": 0.3449,
      "eval_samples_per_second": 724.744,
      "eval_steps_per_second": 72.474,
      "step": 5000
    }
  ],
  "logging_steps": 10,
  "max_steps": 5000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 500,
  "save_steps": 1000,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 6.368413009575936e+16,
  "train_batch_size": 10,
  "trial_name": null,
  "trial_params": null
}