{
  "best_metric": 6.170831738927518e-07,
  "best_model_checkpoint": "finetuned_output/adapter/checkpoint-6000",
  "epoch": 0.9497427779976256,
  "eval_steps": 200,
  "global_step": 6000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "grad_norm": 1.745740532875061,
      "learning_rate": 6.329113924050633e-07,
      "loss": 3.0374,
      "step": 1
    },
    {
      "epoch": 0.0,
      "grad_norm": 1.7303338050842285,
      "learning_rate": 6.329113924050633e-06,
      "loss": 3.0293,
      "step": 10
    },
    {
      "epoch": 0.0,
      "grad_norm": 1.6355781555175781,
      "learning_rate": 1.2658227848101267e-05,
      "loss": 2.919,
      "step": 20
    },
    {
      "epoch": 0.0,
      "grad_norm": 1.622982144355774,
      "learning_rate": 1.89873417721519e-05,
      "loss": 2.6623,
      "step": 30
    },
    {
      "epoch": 0.01,
      "grad_norm": 1.923683524131775,
      "learning_rate": 2.5316455696202533e-05,
      "loss": 2.2493,
      "step": 40
    },
    {
      "epoch": 0.01,
      "grad_norm": 3.1687545776367188,
      "learning_rate": 3.1645569620253167e-05,
      "loss": 1.5777,
      "step": 50
    },
    {
      "epoch": 0.01,
      "grad_norm": 2.44209623336792,
      "learning_rate": 3.79746835443038e-05,
      "loss": 0.6654,
      "step": 60
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.19728538393974304,
      "learning_rate": 4.430379746835443e-05,
      "loss": 0.0718,
      "step": 70
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.04344913363456726,
      "learning_rate": 5.0632911392405066e-05,
      "loss": 0.0041,
      "step": 80
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.015312756411731243,
      "learning_rate": 5.69620253164557e-05,
      "loss": 0.0015,
      "step": 90
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.007904241792857647,
      "learning_rate": 6.329113924050633e-05,
      "loss": 0.0008,
      "step": 100
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.0053505371324718,
      "learning_rate": 6.962025316455697e-05,
      "loss": 0.0005,
      "step": 110
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.004093704745173454,
      "learning_rate": 7.59493670886076e-05,
      "loss": 0.0004,
      "step": 120
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.0033361015375703573,
      "learning_rate": 8.227848101265824e-05,
      "loss": 0.0003,
      "step": 130
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.0028372355736792088,
      "learning_rate": 8.860759493670887e-05,
      "loss": 0.0003,
      "step": 140
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.002454875037074089,
      "learning_rate": 9.493670886075949e-05,
      "loss": 0.0003,
      "step": 150
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.0021501448936760426,
      "learning_rate": 0.00010126582278481013,
      "loss": 0.0002,
      "step": 160
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.0019031608244404197,
      "learning_rate": 0.00010759493670886076,
      "loss": 0.0002,
      "step": 170
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.0016916452441364527,
      "learning_rate": 0.0001139240506329114,
      "loss": 0.0002,
      "step": 180
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.0015131905674934387,
      "learning_rate": 0.00012025316455696203,
      "loss": 0.0002,
      "step": 190
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.0013695439556613564,
      "learning_rate": 0.00012658227848101267,
      "loss": 0.0001,
      "step": 200
    },
    {
      "epoch": 0.03,
      "eval_loss": 0.00013769949146080762,
      "eval_runtime": 677.7841,
      "eval_samples_per_second": 3.926,
      "eval_steps_per_second": 3.926,
      "step": 200
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.001230536145158112,
      "learning_rate": 0.0001329113924050633,
      "loss": 0.0001,
      "step": 210
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.0011144235031679273,
      "learning_rate": 0.00013924050632911395,
      "loss": 0.0001,
      "step": 220
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.0010150589514523745,
      "learning_rate": 0.00014556962025316457,
      "loss": 0.0001,
      "step": 230
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.000926435342989862,
      "learning_rate": 0.0001518987341772152,
      "loss": 0.0001,
      "step": 240
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.0008492307970300317,
      "learning_rate": 0.00015822784810126583,
      "loss": 0.0001,
      "step": 250
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.0007819593884050846,
      "learning_rate": 0.00016455696202531648,
      "loss": 0.0001,
      "step": 260
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.0007205572910606861,
      "learning_rate": 0.0001708860759493671,
      "loss": 0.0001,
      "step": 270
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.0006636082543991506,
      "learning_rate": 0.00017721518987341773,
      "loss": 0.0001,
      "step": 280
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.0006147650419734418,
      "learning_rate": 0.00018354430379746836,
      "loss": 0.0001,
      "step": 290
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.0005710026598535478,
      "learning_rate": 0.00018987341772151899,
      "loss": 0.0001,
      "step": 300
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.0005307784886099398,
      "learning_rate": 0.00019620253164556964,
      "loss": 0.0001,
      "step": 310
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.0004935768665745854,
      "learning_rate": 0.0001999997807486278,
      "loss": 0.0001,
      "step": 320
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.00046100508188828826,
      "learning_rate": 0.00019999731418173165,
      "loss": 0.0001,
      "step": 330
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.00043217180063948035,
      "learning_rate": 0.00019999210705154904,
      "loss": 0.0,
      "step": 340
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.0004070006834808737,
      "learning_rate": 0.00019998415950078858,
      "loss": 0.0,
      "step": 350
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.0003833304508589208,
      "learning_rate": 0.00019997347174726367,
      "loss": 0.0,
      "step": 360
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.0003614730085246265,
      "learning_rate": 0.00019996004408388688,
      "loss": 0.0,
      "step": 370
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.00034343593870289624,
      "learning_rate": 0.00019994387687866173,
      "loss": 0.0,
      "step": 380
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.0003261593810748309,
      "learning_rate": 0.00019992497057467253,
      "loss": 0.0,
      "step": 390
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.00031127542024478316,
      "learning_rate": 0.0001999033256900725,
      "loss": 0.0,
      "step": 400
    },
    {
      "epoch": 0.06,
      "eval_loss": 3.407633994356729e-05,
      "eval_runtime": 678.5668,
      "eval_samples_per_second": 3.922,
      "eval_steps_per_second": 3.922,
      "step": 400
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.00029650679789483547,
      "learning_rate": 0.00019987894281806932,
      "loss": 0.0,
      "step": 410
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.00028275910881347954,
      "learning_rate": 0.00019985182262690897,
      "loss": 0.0,
      "step": 420
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.00027138160658068955,
      "learning_rate": 0.00019982196585985738,
      "loss": 0.0,
      "step": 430
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.00025915552396327257,
      "learning_rate": 0.00019978937333518015,
      "loss": 0.0,
      "step": 440
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.0002491356572136283,
      "learning_rate": 0.00019975404594612005,
      "loss": 0.0,
      "step": 450
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.0002396474446868524,
      "learning_rate": 0.00019971598466087253,
      "loss": 0.0,
      "step": 460
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.00022946584795136005,
      "learning_rate": 0.0001996751905225592,
      "loss": 0.0,
      "step": 470
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.00022204543347470462,
      "learning_rate": 0.0001996316646491993,
      "loss": 0.0,
      "step": 480
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.0002140016295015812,
      "learning_rate": 0.00019958540823367894,
      "loss": 0.0,
      "step": 490
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.00020588333427440375,
      "learning_rate": 0.00019953642254371854,
      "loss": 0.0,
      "step": 500
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.00019794062245637178,
      "learning_rate": 0.00019948470892183797,
      "loss": 0.0,
      "step": 510
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.00019234861247241497,
      "learning_rate": 0.00019943026878531983,
      "loss": 0.0,
      "step": 520
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.00018528260989114642,
      "learning_rate": 0.00019937310362617058,
      "loss": 0.0,
      "step": 530
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.00017959771503228694,
      "learning_rate": 0.0001993132150110797,
      "loss": 0.0,
      "step": 540
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.0001739515137160197,
      "learning_rate": 0.00019925060458137663,
      "loss": 0.0,
      "step": 550
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.00016869328101165593,
      "learning_rate": 0.00019918527405298592,
      "loss": 0.0,
      "step": 560
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.0001634912332519889,
      "learning_rate": 0.00019911722521638017,
      "loss": 0.0,
      "step": 570
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.00015854141383897513,
      "learning_rate": 0.00019904645993653084,
      "loss": 0.0,
      "step": 580
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.000153573855641298,
      "learning_rate": 0.00019897298015285735,
      "loss": 0.0,
      "step": 590
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.00014929994358681142,
      "learning_rate": 0.00019889678787917371,
      "loss": 0.0,
      "step": 600
    },
    {
      "epoch": 0.09,
      "eval_loss": 1.6852183762239292e-05,
      "eval_runtime": 678.4175,
      "eval_samples_per_second": 3.922,
      "eval_steps_per_second": 3.922,
      "step": 600
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.00014471146278083324,
      "learning_rate": 0.00019881788520363354,
      "loss": 0.0,
      "step": 610
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.0001407177624059841,
      "learning_rate": 0.00019873627428867264,
      "loss": 0.0,
      "step": 620
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.0001374084677081555,
      "learning_rate": 0.00019865195737094987,
      "loss": 0.0,
      "step": 630
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.00013354024849832058,
      "learning_rate": 0.00019856493676128582,
      "loss": 0.0,
      "step": 640
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.00013027052045799792,
      "learning_rate": 0.00019847521484459932,
      "loss": 0.0,
      "step": 650
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.00012684000830631703,
      "learning_rate": 0.00019838279407984245,
      "loss": 0.0,
      "step": 660
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.00012365348811727017,
      "learning_rate": 0.00019828767699993268,
      "loss": 0.0,
      "step": 670
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.00012069966032868251,
      "learning_rate": 0.00019818986621168388,
      "loss": 0.0,
      "step": 680
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.00011781425564549863,
      "learning_rate": 0.00019808936439573454,
      "loss": 0.0,
      "step": 690
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.00011501591507112607,
      "learning_rate": 0.00019798617430647454,
      "loss": 0.0,
      "step": 700
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.00011201005690963939,
      "learning_rate": 0.0001978802987719695,
      "loss": 0.0,
      "step": 710
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.0001095582265406847,
      "learning_rate": 0.00019777174069388344,
      "loss": 0.0,
      "step": 720
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.00010714546078816056,
      "learning_rate": 0.00019766050304739907,
      "loss": 0.0,
      "step": 730
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.0001043137745000422,
      "learning_rate": 0.0001975465888811364,
      "loss": 0.0,
      "step": 740
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.0001020901690935716,
      "learning_rate": 0.00019743000131706902,
      "loss": 0.0,
      "step": 750
    },
    {
      "epoch": 0.12,
      "grad_norm": 9.944423800334334e-05,
      "learning_rate": 0.0001973107435504388,
      "loss": 0.0,
      "step": 760
    },
    {
      "epoch": 0.12,
      "grad_norm": 9.746244177222252e-05,
      "learning_rate": 0.00019718881884966805,
      "loss": 0.0,
      "step": 770
    },
    {
      "epoch": 0.12,
      "grad_norm": 9.531319665256888e-05,
      "learning_rate": 0.00019706423055627014,
      "loss": 0.0,
      "step": 780
    },
    {
      "epoch": 0.13,
      "grad_norm": 9.33794435695745e-05,
      "learning_rate": 0.0001969369820847578,
      "loss": 0.0,
      "step": 790
    },
    {
      "epoch": 0.13,
      "grad_norm": 9.126246004598215e-05,
      "learning_rate": 0.00019680707692254954,
      "loss": 0.0,
      "step": 800
    },
    {
      "epoch": 0.13,
      "eval_loss": 1.0555763765296433e-05,
      "eval_runtime": 678.7777,
      "eval_samples_per_second": 3.92,
      "eval_steps_per_second": 3.92,
      "step": 800
    },
    {
      "epoch": 0.13,
      "grad_norm": 8.943810826167464e-05,
      "learning_rate": 0.00019667451862987422,
      "loss": 0.0,
      "step": 810
    },
    {
      "epoch": 0.13,
      "grad_norm": 8.747832180233672e-05,
      "learning_rate": 0.0001965393108396733,
      "loss": 0.0,
      "step": 820
    },
    {
      "epoch": 0.13,
      "grad_norm": 8.603781316196546e-05,
      "learning_rate": 0.00019640145725750136,
      "loss": 0.0,
      "step": 830
    },
    {
      "epoch": 0.13,
      "grad_norm": 8.413767500314862e-05,
      "learning_rate": 0.00019626096166142453,
      "loss": 0.0,
      "step": 840
    },
    {
      "epoch": 0.13,
      "grad_norm": 8.279470057459548e-05,
      "learning_rate": 0.00019611782790191696,
      "loss": 0.0,
      "step": 850
    },
    {
      "epoch": 0.14,
      "grad_norm": 8.096309466054663e-05,
      "learning_rate": 0.00019597205990175525,
      "loss": 0.0,
      "step": 860
    },
    {
      "epoch": 0.14,
      "grad_norm": 7.947826816234738e-05,
      "learning_rate": 0.00019582366165591105,
      "loss": 0.0,
      "step": 870
    },
    {
      "epoch": 0.14,
      "grad_norm": 7.810323586454615e-05,
      "learning_rate": 0.00019567263723144137,
      "loss": 0.0,
      "step": 880
    },
    {
      "epoch": 0.14,
      "grad_norm": 7.679883856326342e-05,
      "learning_rate": 0.00019551899076737734,
      "loss": 0.0,
      "step": 890
    },
    {
      "epoch": 0.14,
      "grad_norm": 7.530328002758324e-05,
      "learning_rate": 0.00019536272647461064,
      "loss": 0.0,
      "step": 900
    },
    {
      "epoch": 0.14,
      "grad_norm": 7.370118692051619e-05,
      "learning_rate": 0.00019520384863577813,
      "loss": 0.0,
      "step": 910
    },
    {
      "epoch": 0.15,
      "grad_norm": 7.2304523200728e-05,
      "learning_rate": 0.0001950423616051445,
      "loss": 0.0,
      "step": 920
    },
    {
      "epoch": 0.15,
      "grad_norm": 7.084510434651747e-05,
      "learning_rate": 0.00019487826980848282,
      "loss": 0.0,
      "step": 930
    },
    {
      "epoch": 0.15,
      "grad_norm": 6.962777115404606e-05,
      "learning_rate": 0.00019471157774295347,
      "loss": 0.0,
      "step": 940
    },
    {
      "epoch": 0.15,
      "grad_norm": 6.825547461630777e-05,
      "learning_rate": 0.0001945422899769807,
      "loss": 0.0,
      "step": 950
    },
    {
      "epoch": 0.15,
      "grad_norm": 6.71844682074152e-05,
      "learning_rate": 0.0001943704111501275,
      "loss": 0.0,
      "step": 960
    },
    {
      "epoch": 0.15,
      "grad_norm": 6.606526585528627e-05,
      "learning_rate": 0.00019419594597296834,
      "loss": 0.0,
      "step": 970
    },
    {
      "epoch": 0.16,
      "grad_norm": 6.487029895652086e-05,
      "learning_rate": 0.00019401889922696035,
      "loss": 0.0,
      "step": 980
    },
    {
      "epoch": 0.16,
      "grad_norm": 6.384370499290526e-05,
      "learning_rate": 0.00019383927576431184,
      "loss": 0.0,
      "step": 990
    },
    {
      "epoch": 0.16,
      "grad_norm": 6.286577990977094e-05,
      "learning_rate": 0.00019365708050784978,
      "loss": 0.0,
      "step": 1000
    },
    {
      "epoch": 0.16,
      "eval_loss": 7.393263331323396e-06,
      "eval_runtime": 678.1441,
      "eval_samples_per_second": 3.924,
      "eval_steps_per_second": 3.924,
      "step": 1000
    },
    {
      "epoch": 0.16,
      "grad_norm": 6.17443656665273e-05,
      "learning_rate": 0.0001934723184508846,
      "loss": 0.0,
      "step": 1010
    },
    {
      "epoch": 0.16,
      "grad_norm": 6.064030094421469e-05,
      "learning_rate": 0.00019328499465707331,
      "loss": 0.0,
      "step": 1020
    },
    {
      "epoch": 0.16,
      "grad_norm": 5.975174281047657e-05,
      "learning_rate": 0.00019309511426028104,
      "loss": 0.0,
      "step": 1030
    },
    {
      "epoch": 0.16,
      "grad_norm": 5.8623936638468876e-05,
      "learning_rate": 0.0001929026824644399,
      "loss": 0.0,
      "step": 1040
    },
    {
      "epoch": 0.17,
      "grad_norm": 5.8019992138724774e-05,
      "learning_rate": 0.00019270770454340674,
      "loss": 0.0,
      "step": 1050
    },
    {
      "epoch": 0.17,
      "grad_norm": 5.6902637879829854e-05,
      "learning_rate": 0.00019251018584081839,
      "loss": 0.0,
      "step": 1060
    },
    {
      "epoch": 0.17,
      "grad_norm": 5.611980304820463e-05,
      "learning_rate": 0.00019231013176994528,
      "loss": 0.0,
      "step": 1070
    },
    {
      "epoch": 0.17,
      "grad_norm": 5.536168464459479e-05,
      "learning_rate": 0.0001921075478135431,
      "loss": 0.0,
      "step": 1080
    },
    {
      "epoch": 0.17,
      "grad_norm": 5.442432302515954e-05,
      "learning_rate": 0.00019190243952370246,
      "loss": 0.0,
      "step": 1090
    },
    {
      "epoch": 0.17,
      "grad_norm": 5.364285607356578e-05,
      "learning_rate": 0.00019169481252169688,
      "loss": 0.0,
      "step": 1100
    },
    {
      "epoch": 0.18,
      "grad_norm": 5.2675430197268724e-05,
      "learning_rate": 0.00019148467249782857,
      "loss": 0.0,
      "step": 1110
    },
    {
      "epoch": 0.18,
      "grad_norm": 5.180203152121976e-05,
      "learning_rate": 0.00019127202521127253,
      "loss": 0.0,
      "step": 1120
    },
    {
      "epoch": 0.18,
      "grad_norm": 5.1117694965796545e-05,
      "learning_rate": 0.00019105687648991875,
      "loss": 0.0,
      "step": 1130
    },
    {
      "epoch": 0.18,
      "grad_norm": 5.032157787354663e-05,
      "learning_rate": 0.00019083923223021248,
      "loss": 0.0,
      "step": 1140
    },
    {
      "epoch": 0.18,
      "grad_norm": 4.9563124775886536e-05,
      "learning_rate": 0.00019061909839699262,
      "loss": 0.0,
      "step": 1150
    },
    {
      "epoch": 0.18,
      "grad_norm": 4.8855301429284737e-05,
      "learning_rate": 0.00019039648102332822,
      "loss": 0.0,
      "step": 1160
    },
    {
      "epoch": 0.19,
      "grad_norm": 4.8094869271153584e-05,
      "learning_rate": 0.0001901713862103531,
      "loss": 0.0,
      "step": 1170
    },
    {
      "epoch": 0.19,
      "grad_norm": 4.7465840907534584e-05,
      "learning_rate": 0.00018994382012709884,
      "loss": 0.0,
      "step": 1180
    },
    {
      "epoch": 0.19,
      "grad_norm": 4.675550735555589e-05,
      "learning_rate": 0.00018971378901032538,
      "loss": 0.0,
      "step": 1190
    },
    {
      "epoch": 0.19,
      "grad_norm": 4.6109249524306506e-05,
      "learning_rate": 0.00018948129916435046,
      "loss": 0.0,
      "step": 1200
    },
    {
      "epoch": 0.19,
      "eval_loss": 5.517257704923395e-06,
      "eval_runtime": 678.5963,
      "eval_samples_per_second": 3.921,
      "eval_steps_per_second": 3.921,
      "step": 1200
    },
    {
      "epoch": 0.19,
      "grad_norm": 4.556746716843918e-05,
      "learning_rate": 0.00018924635696087656,
      "loss": 0.0,
      "step": 1210
    },
    {
      "epoch": 0.19,
      "grad_norm": 4.498475755099207e-05,
      "learning_rate": 0.00018900896883881636,
      "loss": 0.0,
      "step": 1220
    },
    {
      "epoch": 0.19,
      "grad_norm": 4.422712663654238e-05,
      "learning_rate": 0.00018876914130411634,
      "loss": 0.0,
      "step": 1230
    },
    {
      "epoch": 0.2,
      "grad_norm": 4.358746809884906e-05,
      "learning_rate": 0.00018852688092957837,
      "loss": 0.0,
      "step": 1240
    },
    {
      "epoch": 0.2,
      "grad_norm": 4.317778439144604e-05,
      "learning_rate": 0.0001882821943546797,
      "loss": 0.0,
      "step": 1250
    },
    {
      "epoch": 0.2,
      "grad_norm": 4.255426028976217e-05,
      "learning_rate": 0.00018803508828539088,
      "loss": 0.0,
      "step": 1260
    },
    {
      "epoch": 0.2,
      "grad_norm": 4.208676182315685e-05,
      "learning_rate": 0.00018778556949399194,
      "loss": 0.0,
      "step": 1270
    },
    {
      "epoch": 0.2,
      "grad_norm": 4.1539660742273554e-05,
      "learning_rate": 0.000187533644818887,
      "loss": 0.0,
      "step": 1280
    },
    {
      "epoch": 0.2,
      "grad_norm": 4.1055274778045714e-05,
      "learning_rate": 0.00018727932116441667,
      "loss": 0.0,
      "step": 1290
    },
    {
      "epoch": 0.21,
      "grad_norm": 4.03260528401006e-05,
      "learning_rate": 0.00018702260550066882,
      "loss": 0.0,
      "step": 1300
    },
    {
      "epoch": 0.21,
      "grad_norm": 3.97946969314944e-05,
      "learning_rate": 0.00018676350486328765,
      "loss": 0.0,
      "step": 1310
    },
    {
      "epoch": 0.21,
      "grad_norm": 3.927663419744931e-05,
      "learning_rate": 0.00018650202635328084,
      "loss": 0.0,
      "step": 1320
    },
    {
      "epoch": 0.21,
      "grad_norm": 3.872224624501541e-05,
      "learning_rate": 0.0001862381771368249,
      "loss": 0.0,
      "step": 1330
    },
    {
      "epoch": 0.21,
      "grad_norm": 3.8312784454319626e-05,
      "learning_rate": 0.00018597196444506878,
      "loss": 0.0,
      "step": 1340
    },
    {
      "epoch": 0.21,
      "grad_norm": 3.7926682125544176e-05,
      "learning_rate": 0.00018570339557393576,
      "loss": 0.0,
      "step": 1350
    },
    {
      "epoch": 0.22,
      "grad_norm": 3.739940075320192e-05,
      "learning_rate": 0.0001854324778839234,
      "loss": 0.0,
      "step": 1360
    },
    {
      "epoch": 0.22,
      "grad_norm": 3.68890177924186e-05,
      "learning_rate": 0.00018515921879990187,
      "loss": 0.0,
      "step": 1370
    },
    {
      "epoch": 0.22,
      "grad_norm": 3.658093555714004e-05,
      "learning_rate": 0.00018488362581091044,
      "loss": 0.0,
      "step": 1380
    },
    {
      "epoch": 0.22,
      "grad_norm": 3.6252939025871456e-05,
      "learning_rate": 0.00018460570646995224,
      "loss": 0.0,
      "step": 1390
    },
    {
      "epoch": 0.22,
      "grad_norm": 3.566632221918553e-05,
      "learning_rate": 0.0001843254683937873,
      "loss": 0.0,
      "step": 1400
    },
    {
      "epoch": 0.22,
      "eval_loss": 4.310215445002541e-06,
      "eval_runtime": 677.7713,
      "eval_samples_per_second": 3.926,
      "eval_steps_per_second": 3.926,
      "step": 1400
    },
    {
      "epoch": 0.22,
      "grad_norm": 3.536800795700401e-05,
      "learning_rate": 0.00018404291926272368,
      "loss": 0.0,
      "step": 1410
    },
    {
      "epoch": 0.22,
      "grad_norm": 3.4788034099619836e-05,
      "learning_rate": 0.00018375806682040716,
      "loss": 0.0,
      "step": 1420
    },
    {
      "epoch": 0.23,
      "grad_norm": 3.466546695563011e-05,
      "learning_rate": 0.00018347091887360882,
      "loss": 0.0,
      "step": 1430
    },
    {
      "epoch": 0.23,
      "grad_norm": 3.415128230699338e-05,
      "learning_rate": 0.0001831814832920113,
      "loss": 0.0,
      "step": 1440
    },
    {
      "epoch": 0.23,
      "grad_norm": 3.3692907891236246e-05,
      "learning_rate": 0.00018288976800799286,
      "loss": 0.0,
      "step": 1450
    },
    {
      "epoch": 0.23,
      "grad_norm": 3.305283826193772e-05,
      "learning_rate": 0.00018259578101641023,
      "loss": 0.0,
      "step": 1460
    },
    {
      "epoch": 0.23,
      "grad_norm": 3.287609069957398e-05,
      "learning_rate": 0.00018229953037437942,
      "loss": 0.0,
      "step": 1470
    },
    {
      "epoch": 0.23,
      "grad_norm": 3.256453055655584e-05,
      "learning_rate": 0.00018200102420105476,
      "loss": 0.0,
      "step": 1480
    },
    {
      "epoch": 0.24,
      "grad_norm": 3.209506030543707e-05,
      "learning_rate": 0.0001817002706774066,
      "loss": 0.0,
      "step": 1490
    },
    {
      "epoch": 0.24,
      "grad_norm": 3.183619264746085e-05,
      "learning_rate": 0.00018139727804599695,
      "loss": 0.0,
      "step": 1500
    },
    {
      "epoch": 0.24,
      "grad_norm": 3.140209082630463e-05,
      "learning_rate": 0.00018109205461075368,
      "loss": 0.0,
      "step": 1510
    },
    {
      "epoch": 0.24,
      "grad_norm": 3.0993491236586124e-05,
      "learning_rate": 0.0001807846087367429,
      "loss": 0.0,
      "step": 1520
    },
    {
      "epoch": 0.24,
      "grad_norm": 3.067683428525925e-05,
      "learning_rate": 0.00018047494884993959,
      "loss": 0.0,
      "step": 1530
    },
    {
      "epoch": 0.24,
      "grad_norm": 3.0400608011404984e-05,
      "learning_rate": 0.00018016308343699687,
      "loss": 0.0,
      "step": 1540
    },
    {
      "epoch": 0.25,
      "grad_norm": 3.0034292649361305e-05,
      "learning_rate": 0.00017984902104501335,
      "loss": 0.0,
      "step": 1550
    },
    {
      "epoch": 0.25,
      "grad_norm": 2.979646342282649e-05,
      "learning_rate": 0.00017953277028129883,
      "loss": 0.0,
      "step": 1560
    },
    {
      "epoch": 0.25,
      "grad_norm": 2.9413335141725838e-05,
      "learning_rate": 0.00017921433981313834,
      "loss": 0.0,
      "step": 1570
    },
    {
      "epoch": 0.25,
      "grad_norm": 2.9106640795362182e-05,
      "learning_rate": 0.00017889373836755485,
      "loss": 0.0,
      "step": 1580
    },
    {
      "epoch": 0.25,
      "grad_norm": 2.8897004085592926e-05,
      "learning_rate": 0.00017857097473106986,
      "loss": 0.0,
      "step": 1590
    },
    {
      "epoch": 0.25,
      "grad_norm": 2.8534617740660906e-05,
      "learning_rate": 0.00017824605774946262,
      "loss": 0.0,
      "step": 1600
    },
    {
      "epoch": 0.25,
      "eval_loss": 3.4715505989879603e-06,
      "eval_runtime": 678.1094,
      "eval_samples_per_second": 3.924,
      "eval_steps_per_second": 3.924,
      "step": 1600
    },
    {
      "epoch": 0.25,
      "grad_norm": 2.825512274284847e-05,
      "learning_rate": 0.00017791899632752783,
      "loss": 0.0,
      "step": 1610
    },
    {
      "epoch": 0.26,
      "grad_norm": 2.7753631002269685e-05,
      "learning_rate": 0.00017758979942883153,
      "loss": 0.0,
      "step": 1620
    },
    {
      "epoch": 0.26,
      "grad_norm": 2.7518633942236193e-05,
      "learning_rate": 0.00017725847607546533,
      "loss": 0.0,
      "step": 1630
    },
    {
      "epoch": 0.26,
      "grad_norm": 2.7194701033295132e-05,
      "learning_rate": 0.00017692503534779937,
      "loss": 0.0,
      "step": 1640
    },
    {
      "epoch": 0.26,
      "grad_norm": 2.694679278647527e-05,
      "learning_rate": 0.00017658948638423313,
      "loss": 0.0,
      "step": 1650
    },
    {
      "epoch": 0.26,
      "grad_norm": 2.6594472728902474e-05,
      "learning_rate": 0.0001762518383809454,
      "loss": 0.0,
      "step": 1660
    },
    {
      "epoch": 0.26,
      "grad_norm": 2.6295565476175398e-05,
      "learning_rate": 0.00017591210059164194,
      "loss": 0.0,
      "step": 1670
    },
    {
      "epoch": 0.27,
      "grad_norm": 2.6117490051547065e-05,
      "learning_rate": 0.0001755702823273019,
      "loss": 0.0,
      "step": 1680
    },
    {
      "epoch": 0.27,
      "grad_norm": 2.586103255453054e-05,
      "learning_rate": 0.00017522639295592277,
      "loss": 0.0,
      "step": 1690
    },
    {
      "epoch": 0.27,
      "grad_norm": 2.5608789655962028e-05,
      "learning_rate": 0.0001748804419022635,
      "loss": 0.0,
      "step": 1700
    },
    {
      "epoch": 0.27,
      "grad_norm": 2.5343721063109115e-05,
      "learning_rate": 0.00017453243864758638,
      "loss": 0.0,
      "step": 1710
    },
    {
      "epoch": 0.27,
      "grad_norm": 2.508814759494271e-05,
      "learning_rate": 0.00017418239272939693,
      "loss": 0.0,
      "step": 1720
    },
    {
      "epoch": 0.27,
      "grad_norm": 2.4780534658930264e-05,
      "learning_rate": 0.0001738303137411828,
      "loss": 0.0,
      "step": 1730
    },
    {
      "epoch": 0.28,
      "grad_norm": 2.452327862556558e-05,
      "learning_rate": 0.0001734762113321506,
      "loss": 0.0,
      "step": 1740
    },
    {
      "epoch": 0.28,
      "grad_norm": 2.4306553314090706e-05,
      "learning_rate": 0.0001731200952069616,
      "loss": 0.0,
      "step": 1750
    },
    {
      "epoch": 0.28,
      "grad_norm": 2.406495696050115e-05,
      "learning_rate": 0.00017276197512546573,
      "loss": 0.0,
      "step": 1760
    },
    {
      "epoch": 0.28,
      "grad_norm": 2.372384915361181e-05,
      "learning_rate": 0.0001724018609024341,
      "loss": 0.0,
      "step": 1770
    },
    {
      "epoch": 0.28,
      "grad_norm": 2.3568389224237762e-05,
      "learning_rate": 0.00017203976240728989,
      "loss": 0.0,
      "step": 1780
    },
    {
      "epoch": 0.28,
      "grad_norm": 2.3447633793693967e-05,
      "learning_rate": 0.00017167568956383812,
      "loss": 0.0,
      "step": 1790
    },
    {
      "epoch": 0.28,
      "grad_norm": 2.3247848730534315e-05,
      "learning_rate": 0.00017130965234999338,
      "loss": 0.0,
      "step": 1800
    },
    {
      "epoch": 0.28,
      "eval_loss": 2.874105121009052e-06,
      "eval_runtime": 677.576,
      "eval_samples_per_second": 3.927,
      "eval_steps_per_second": 3.927,
      "step": 1800
    },
    {
      "epoch": 0.29,
      "grad_norm": 2.304599547642283e-05,
      "learning_rate": 0.00017094166079750672,
      "loss": 0.0,
      "step": 1810
    },
    {
      "epoch": 0.29,
      "grad_norm": 2.2788086425862275e-05,
      "learning_rate": 0.00017057172499169033,
      "loss": 0.0,
      "step": 1820
    },
    {
      "epoch": 0.29,
      "grad_norm": 2.2592799723497592e-05,
      "learning_rate": 0.0001701998550711414,
      "loss": 0.0,
      "step": 1830
    },
    {
      "epoch": 0.29,
      "grad_norm": 2.249391036457382e-05,
      "learning_rate": 0.0001698260612274642,
      "loss": 0.0,
      "step": 1840
    },
    {
      "epoch": 0.29,
      "grad_norm": 2.2216292563825846e-05,
      "learning_rate": 0.00016945035370499064,
      "loss": 0.0,
      "step": 1850
    },
    {
      "epoch": 0.29,
      "grad_norm": 2.208605656051077e-05,
      "learning_rate": 0.0001690727428004998,
      "loss": 0.0,
      "step": 1860
    },
    {
      "epoch": 0.3,
      "grad_norm": 2.177110218326561e-05,
      "learning_rate": 0.0001686932388629354,
      "loss": 0.0,
      "step": 1870
    },
    {
      "epoch": 0.3,
      "grad_norm": 2.1624189685098827e-05,
      "learning_rate": 0.00016831185229312237,
      "loss": 0.0,
      "step": 1880
    },
    {
      "epoch": 0.3,
      "grad_norm": 2.1496376575669274e-05,
      "learning_rate": 0.00016792859354348175,
      "loss": 0.0,
      "step": 1890
    },
    {
      "epoch": 0.3,
      "grad_norm": 2.117209805874154e-05,
      "learning_rate": 0.00016754347311774423,
      "loss": 0.0,
      "step": 1900
    },
    {
      "epoch": 0.3,
      "grad_norm": 2.09246118174633e-05,
      "learning_rate": 0.00016715650157066235,
      "loss": 0.0,
      "step": 1910
    },
    {
      "epoch": 0.3,
      "grad_norm": 2.0770705305039883e-05,
      "learning_rate": 0.000166767689507721,
      "loss": 0.0,
      "step": 1920
    },
    {
      "epoch": 0.31,
      "grad_norm": 2.0503721316345036e-05,
      "learning_rate": 0.0001663770475848471,
      "loss": 0.0,
      "step": 1930
    },
    {
      "epoch": 0.31,
      "grad_norm": 2.0482279069256037e-05,
      "learning_rate": 0.00016598458650811732,
      "loss": 0.0,
      "step": 1940
    },
    {
      "epoch": 0.31,
      "grad_norm": 2.016001963056624e-05,
      "learning_rate": 0.00016559031703346469,
      "loss": 0.0,
      "step": 1950
    },
    {
      "epoch": 0.31,
      "grad_norm": 2.0079261958017014e-05,
      "learning_rate": 0.00016519424996638393,
      "loss": 0.0,
      "step": 1960
    },
    {
      "epoch": 0.31,
      "grad_norm": 1.9872393750119954e-05,
      "learning_rate": 0.0001647963961616352,
      "loss": 0.0,
      "step": 1970
    },
    {
      "epoch": 0.31,
      "grad_norm": 1.9629405869636685e-05,
      "learning_rate": 0.00016439676652294663,
      "loss": 0.0,
      "step": 1980
    },
    {
      "epoch": 0.31,
      "grad_norm": 1.9679631805047393e-05,
      "learning_rate": 0.00016399537200271556,
      "loss": 0.0,
      "step": 1990
    },
    {
      "epoch": 0.32,
      "grad_norm": 1.92725783563219e-05,
      "learning_rate": 0.0001635922236017083,
      "loss": 0.0,
      "step": 2000
    },
    {
      "epoch": 0.32,
      "eval_loss": 2.4346688860532595e-06,
      "eval_runtime": 678.1413,
      "eval_samples_per_second": 3.924,
      "eval_steps_per_second": 3.924,
      "step": 2000
    },
    {
      "epoch": 0.32,
      "grad_norm": 1.919512578751892e-05,
      "learning_rate": 0.0001631873323687587,
      "loss": 0.0,
      "step": 2010
    },
    {
      "epoch": 0.32,
      "grad_norm": 1.9078650439041667e-05,
      "learning_rate": 0.00016278070940046527,
      "loss": 0.0,
      "step": 2020
    },
    {
      "epoch": 0.32,
      "grad_norm": 1.8961709429277107e-05,
      "learning_rate": 0.00016237236584088705,
      "loss": 0.0,
      "step": 2030
    },
    {
      "epoch": 0.32,
      "grad_norm": 1.8703172827372327e-05,
      "learning_rate": 0.00016196231288123837,
      "loss": 0.0,
      "step": 2040
    },
    {
      "epoch": 0.32,
      "grad_norm": 1.852524292189628e-05,
      "learning_rate": 0.0001615505617595819,
      "loss": 0.0,
      "step": 2050
    },
    {
      "epoch": 0.33,
      "grad_norm": 1.839043034124188e-05,
      "learning_rate": 0.00016113712376052077,
      "loss": 0.0,
      "step": 2060
    },
    {
      "epoch": 0.33,
      "grad_norm": 1.8239614291815087e-05,
      "learning_rate": 0.00016072201021488934,
      "loss": 0.0,
      "step": 2070
    },
    {
      "epoch": 0.33,
      "grad_norm": 1.8058430214296095e-05,
      "learning_rate": 0.00016030523249944257,
      "loss": 0.0,
      "step": 2080
    },
    {
      "epoch": 0.33,
      "grad_norm": 1.7962598576559685e-05,
      "learning_rate": 0.00015988680203654434,
      "loss": 0.0,
      "step": 2090
    },
    {
      "epoch": 0.33,
      "grad_norm": 1.7849572031991556e-05,
      "learning_rate": 0.00015946673029385426,
      "loss": 0.0,
      "step": 2100
    },
    {
      "epoch": 0.33,
      "grad_norm": 1.770034577930346e-05,
      "learning_rate": 0.00015904502878401355,
      "loss": 0.0,
      "step": 2110
    },
    {
      "epoch": 0.34,
      "grad_norm": 1.757799691404216e-05,
      "learning_rate": 0.00015862170906432934,
      "loss": 0.0,
      "step": 2120
    },
    {
      "epoch": 0.34,
      "grad_norm": 1.7467971701989882e-05,
      "learning_rate": 0.00015819678273645802,
      "loss": 0.0,
      "step": 2130
    },
    {
      "epoch": 0.34,
      "grad_norm": 1.730651274556294e-05,
      "learning_rate": 0.00015777026144608738,
      "loss": 0.0,
      "step": 2140
    },
    {
      "epoch": 0.34,
      "grad_norm": 1.7148164261016063e-05,
      "learning_rate": 0.00015734215688261718,
      "loss": 0.0,
      "step": 2150
    },
    {
      "epoch": 0.34,
      "grad_norm": 1.7010233932523988e-05,
      "learning_rate": 0.00015691248077883912,
      "loss": 0.0,
      "step": 2160
    },
    {
      "epoch": 0.34,
      "grad_norm": 1.685299321252387e-05,
      "learning_rate": 0.000156481244910615,
      "loss": 0.0,
      "step": 2170
    },
    {
      "epoch": 0.35,
      "grad_norm": 1.664710907789413e-05,
      "learning_rate": 0.00015604846109655408,
      "loss": 0.0,
      "step": 2180
    },
    {
      "epoch": 0.35,
      "grad_norm": 1.640819027670659e-05,
      "learning_rate": 0.00015561414119768933,
      "loss": 0.0,
      "step": 2190
    },
    {
      "epoch": 0.35,
      "grad_norm": 1.634788168303203e-05,
      "learning_rate": 0.00015517829711715207,
      "loss": 0.0,
      "step": 2200
    },
    {
      "epoch": 0.35,
      "eval_loss": 2.087794882754679e-06,
      "eval_runtime": 678.3684,
      "eval_samples_per_second": 3.923,
      "eval_steps_per_second": 3.923,
      "step": 2200
    },
    {
      "epoch": 0.35,
      "grad_norm": 1.6288742699543945e-05,
      "learning_rate": 0.000154740940799846,
      "loss": 0.0,
      "step": 2210
    },
    {
      "epoch": 0.35,
      "grad_norm": 1.611598599993158e-05,
      "learning_rate": 0.00015430208423211975,
      "loss": 0.0,
      "step": 2220
    },
    {
      "epoch": 0.35,
      "grad_norm": 1.593867091287393e-05,
      "learning_rate": 0.0001538617394414383,
      "loss": 0.0,
      "step": 2230
    },
    {
      "epoch": 0.35,
      "grad_norm": 1.5858329788898118e-05,
      "learning_rate": 0.00015341991849605337,
      "loss": 0.0,
      "step": 2240
    },
    {
      "epoch": 0.36,
      "grad_norm": 1.5770854588481598e-05,
      "learning_rate": 0.00015297663350467294,
      "loss": 0.0,
      "step": 2250
    },
    {
      "epoch": 0.36,
      "grad_norm": 1.5689287465647794e-05,
      "learning_rate": 0.00015253189661612897,
      "loss": 0.0,
      "step": 2260
    },
    {
      "epoch": 0.36,
      "grad_norm": 1.5531790268141776e-05,
      "learning_rate": 0.00015208572001904468,
      "loss": 0.0,
      "step": 2270
    },
    {
      "epoch": 0.36,
      "grad_norm": 1.5504576367675327e-05,
      "learning_rate": 0.00015163811594150057,
      "loss": 0.0,
      "step": 2280
    },
    {
      "epoch": 0.36,
      "grad_norm": 1.5426132449647412e-05,
      "learning_rate": 0.00015118909665069906,
      "loss": 0.0,
      "step": 2290
    },
    {
      "epoch": 0.36,
      "grad_norm": 1.528296706965193e-05,
      "learning_rate": 0.0001507386744526286,
      "loss": 0.0,
      "step": 2300
    },
    {
      "epoch": 0.37,
      "grad_norm": 1.5145509678404778e-05,
      "learning_rate": 0.00015028686169172602,
      "loss": 0.0,
      "step": 2310
    },
    {
      "epoch": 0.37,
      "grad_norm": 1.509943376731826e-05,
      "learning_rate": 0.0001498336707505386,
      "loss": 0.0,
      "step": 2320
    },
    {
      "epoch": 0.37,
      "grad_norm": 1.4934204045857769e-05,
      "learning_rate": 0.00014937911404938451,
      "loss": 0.0,
      "step": 2330
    },
    {
      "epoch": 0.37,
      "grad_norm": 1.477322894061217e-05,
      "learning_rate": 0.00014892320404601232,
      "loss": 0.0,
      "step": 2340
    },
    {
      "epoch": 0.37,
      "grad_norm": 1.4727381312695798e-05,
      "learning_rate": 0.0001484659532352599,
      "loss": 0.0,
      "step": 2350
    },
    {
      "epoch": 0.37,
      "grad_norm": 1.4560227100446355e-05,
      "learning_rate": 0.0001480073741487116,
      "loss": 0.0,
      "step": 2360
    },
    {
      "epoch": 0.38,
      "grad_norm": 1.449589035473764e-05,
      "learning_rate": 0.00014754747935435508,
      "loss": 0.0,
      "step": 2370
    },
    {
      "epoch": 0.38,
      "grad_norm": 1.443142809876008e-05,
      "learning_rate": 0.0001470862814562367,
      "loss": 0.0,
      "step": 2380
    },
    {
      "epoch": 0.38,
      "grad_norm": 1.4300727343652397e-05,
      "learning_rate": 0.0001466237930941163,
      "loss": 0.0,
      "step": 2390
    },
    {
      "epoch": 0.38,
      "grad_norm": 1.429617350368062e-05,
      "learning_rate": 0.00014616002694312056,
      "loss": 0.0,
      "step": 2400
    },
    {
      "epoch": 0.38,
      "eval_loss": 1.813380436033185e-06,
      "eval_runtime": 677.1885,
      "eval_samples_per_second": 3.929,
      "eval_steps_per_second": 3.929,
      "step": 2400
    },
    {
      "epoch": 0.38,
      "grad_norm": 1.425579557690071e-05,
      "learning_rate": 0.0001456949957133956,
      "loss": 0.0,
      "step": 2410
    },
    {
      "epoch": 0.38,
      "grad_norm": 1.4058313354325946e-05,
      "learning_rate": 0.00014522871214975907,
      "loss": 0.0,
      "step": 2420
    },
    {
      "epoch": 0.38,
      "grad_norm": 1.4009660844749305e-05,
      "learning_rate": 0.00014476118903135018,
      "loss": 0.0,
      "step": 2430
    },
    {
      "epoch": 0.39,
      "grad_norm": 1.3869190297555178e-05,
      "learning_rate": 0.0001442924391712801,
      "loss": 0.0,
      "step": 2440
    },
    {
      "epoch": 0.39,
      "grad_norm": 1.3833836419507861e-05,
      "learning_rate": 0.00014382247541628036,
      "loss": 0.0,
      "step": 2450
    },
    {
      "epoch": 0.39,
      "grad_norm": 1.3745286196353845e-05,
      "learning_rate": 0.00014335131064635112,
      "loss": 0.0,
      "step": 2460
    },
    {
      "epoch": 0.39,
      "grad_norm": 1.3666711311088875e-05,
      "learning_rate": 0.00014287895777440786,
      "loss": 0.0,
      "step": 2470
    },
    {
      "epoch": 0.39,
      "grad_norm": 1.3573147043643985e-05,
      "learning_rate": 0.00014240542974592764,
      "loss": 0.0,
      "step": 2480
    },
    {
      "epoch": 0.39,
      "grad_norm": 1.3365749509830493e-05,
      "learning_rate": 0.00014193073953859434,
      "loss": 0.0,
      "step": 2490
    },
    {
      "epoch": 0.4,
      "grad_norm": 1.331144812866114e-05,
      "learning_rate": 0.0001414549001619429,
      "loss": 0.0,
      "step": 2500
    },
    {
      "epoch": 0.4,
      "grad_norm": 1.3190996469347738e-05,
      "learning_rate": 0.00014097792465700282,
      "loss": 0.0,
      "step": 2510
    },
    {
      "epoch": 0.4,
      "grad_norm": 1.3139730071998201e-05,
      "learning_rate": 0.00014049982609594085,
      "loss": 0.0,
      "step": 2520
    },
    {
      "epoch": 0.4,
      "grad_norm": 1.306840840697987e-05,
      "learning_rate": 0.0001400206175817025,
      "loss": 0.0,
      "step": 2530
    },
    {
      "epoch": 0.4,
      "grad_norm": 1.3012773706577718e-05,
      "learning_rate": 0.00013954031224765306,
      "loss": 0.0,
      "step": 2540
    },
    {
      "epoch": 0.4,
      "grad_norm": 1.2865250937466044e-05,
      "learning_rate": 0.0001390589232572178,
      "loss": 0.0,
      "step": 2550
    },
    {
      "epoch": 0.41,
      "grad_norm": 1.2822238204535097e-05,
      "learning_rate": 0.00013857646380352102,
      "loss": 0.0,
      "step": 2560
    },
    {
      "epoch": 0.41,
      "grad_norm": 1.2771665751643013e-05,
      "learning_rate": 0.0001380929471090244,
      "loss": 0.0,
      "step": 2570
    },
    {
      "epoch": 0.41,
      "grad_norm": 1.2760150639223866e-05,
      "learning_rate": 0.00013760838642516495,
      "loss": 0.0,
      "step": 2580
    },
    {
      "epoch": 0.41,
      "grad_norm": 1.2568410056701396e-05,
      "learning_rate": 0.00013712279503199154,
      "loss": 0.0,
      "step": 2590
    },
    {
      "epoch": 0.41,
      "grad_norm": 1.2498898286139593e-05,
      "learning_rate": 0.00013663618623780104,
      "loss": 0.0,
      "step": 2600
    },
    {
      "epoch": 0.41,
      "eval_loss": 1.5974021607689792e-06,
      "eval_runtime": 677.165,
      "eval_samples_per_second": 3.93,
      "eval_steps_per_second": 3.93,
      "step": 2600
    },
    {
      "epoch": 0.41,
      "grad_norm": 1.245233761437703e-05,
      "learning_rate": 0.0001361485733787736,
      "loss": 0.0,
      "step": 2610
    },
    {
      "epoch": 0.41,
      "grad_norm": 1.2358401363599114e-05,
      "learning_rate": 0.00013565996981860712,
      "loss": 0.0,
      "step": 2620
    },
    {
      "epoch": 0.42,
      "grad_norm": 1.2247163795109373e-05,
      "learning_rate": 0.00013517038894815108,
      "loss": 0.0,
      "step": 2630
    },
    {
      "epoch": 0.42,
      "grad_norm": 1.2209315173095092e-05,
      "learning_rate": 0.0001346798441850394,
      "loss": 0.0,
      "step": 2640
    },
    {
      "epoch": 0.42,
      "grad_norm": 1.2194812370580621e-05,
      "learning_rate": 0.00013418834897332288,
      "loss": 0.0,
      "step": 2650
    },
    {
      "epoch": 0.42,
      "grad_norm": 1.2058239917678293e-05,
      "learning_rate": 0.00013369591678310063,
      "loss": 0.0,
      "step": 2660
    },
    {
      "epoch": 0.42,
      "grad_norm": 1.2058473657816648e-05,
      "learning_rate": 0.00013320256111015087,
      "loss": 0.0,
      "step": 2670
    },
    {
      "epoch": 0.42,
      "grad_norm": 1.194212836708175e-05,
      "learning_rate": 0.00013270829547556133,
      "loss": 0.0,
      "step": 2680
    },
    {
      "epoch": 0.43,
      "grad_norm": 1.1821592124761082e-05,
      "learning_rate": 0.0001322131334253582,
      "loss": 0.0,
      "step": 2690
    },
    {
      "epoch": 0.43,
      "grad_norm": 1.1817806807812303e-05,
      "learning_rate": 0.00013171708853013545,
      "loss": 0.0,
      "step": 2700
    },
    {
      "epoch": 0.43,
      "grad_norm": 1.1675842870317865e-05,
      "learning_rate": 0.0001312201743846824,
      "loss": 0.0,
      "step": 2710
    },
    {
      "epoch": 0.43,
      "grad_norm": 1.170699397334829e-05,
      "learning_rate": 0.00013072240460761147,
      "loss": 0.0,
      "step": 2720
    },
    {
      "epoch": 0.43,
      "grad_norm": 1.1621774319792166e-05,
      "learning_rate": 0.00013022379284098487,
      "loss": 0.0,
      "step": 2730
    },
    {
      "epoch": 0.43,
      "grad_norm": 1.1526318303367589e-05,
      "learning_rate": 0.00012972435274994062,
      "loss": 0.0,
      "step": 2740
    },
    {
      "epoch": 0.44,
      "grad_norm": 1.1440261914685834e-05,
      "learning_rate": 0.0001292240980223181,
      "loss": 0.0,
      "step": 2750
    },
    {
      "epoch": 0.44,
      "grad_norm": 1.1386619917175267e-05,
      "learning_rate": 0.00012872304236828295,
      "loss": 0.0,
      "step": 2760
    },
    {
      "epoch": 0.44,
      "grad_norm": 1.1362284567439929e-05,
      "learning_rate": 0.00012822119951995132,
      "loss": 0.0,
      "step": 2770
    },
    {
      "epoch": 0.44,
      "grad_norm": 1.1351637112966273e-05,
      "learning_rate": 0.0001277185832310134,
      "loss": 0.0,
      "step": 2780
    },
    {
      "epoch": 0.44,
      "grad_norm": 1.1198649190191645e-05,
      "learning_rate": 0.00012721520727635674,
      "loss": 0.0,
      "step": 2790
    },
    {
      "epoch": 0.44,
      "grad_norm": 1.113287635234883e-05,
      "learning_rate": 0.00012671108545168842,
      "loss": 0.0,
      "step": 2800
    },
    {
      "epoch": 0.44,
      "eval_loss": 1.416017994415597e-06,
      "eval_runtime": 676.7498,
      "eval_samples_per_second": 3.932,
      "eval_steps_per_second": 3.932,
      "step": 2800
    },
    {
      "epoch": 0.44,
      "grad_norm": 1.1048445230699144e-05,
      "learning_rate": 0.00012620623157315715,
      "loss": 0.0,
      "step": 2810
    },
    {
      "epoch": 0.45,
      "grad_norm": 1.1035486750188284e-05,
      "learning_rate": 0.00012570065947697457,
      "loss": 0.0,
      "step": 2820
    },
    {
      "epoch": 0.45,
      "grad_norm": 1.0959458450088277e-05,
      "learning_rate": 0.00012519438301903616,
      "loss": 0.0,
      "step": 2830
    },
    {
      "epoch": 0.45,
      "grad_norm": 1.092481215891894e-05,
      "learning_rate": 0.00012468741607454123,
      "loss": 0.0,
      "step": 2840
    },
    {
      "epoch": 0.45,
      "grad_norm": 1.082818562281318e-05,
      "learning_rate": 0.000124179772537613,
      "loss": 0.0,
      "step": 2850
    },
    {
      "epoch": 0.45,
      "grad_norm": 1.0677342288545333e-05,
      "learning_rate": 0.00012367146632091743,
      "loss": 0.0,
      "step": 2860
    },
    {
      "epoch": 0.45,
      "grad_norm": 1.066176173480926e-05,
      "learning_rate": 0.00012316251135528235,
      "loss": 0.0,
      "step": 2870
    },
    {
      "epoch": 0.46,
      "grad_norm": 1.0617589396133553e-05,
      "learning_rate": 0.00012265292158931523,
      "loss": 0.0,
      "step": 2880
    },
    {
      "epoch": 0.46,
      "grad_norm": 1.0559007932897657e-05,
      "learning_rate": 0.00012214271098902124,
      "loss": 0.0,
      "step": 2890
    },
    {
      "epoch": 0.46,
      "grad_norm": 1.0487300642125774e-05,
      "learning_rate": 0.00012163189353742035,
      "loss": 0.0,
      "step": 2900
    },
    {
      "epoch": 0.46,
      "grad_norm": 1.0520334399188869e-05,
      "learning_rate": 0.00012112048323416407,
      "loss": 0.0,
      "step": 2910
    },
    {
      "epoch": 0.46,
      "grad_norm": 1.0428723726363387e-05,
      "learning_rate": 0.00012060849409515191,
      "loss": 0.0,
      "step": 2920
    },
    {
      "epoch": 0.46,
      "grad_norm": 1.0382533218944445e-05,
      "learning_rate": 0.00012009594015214705,
      "loss": 0.0,
      "step": 2930
    },
    {
      "epoch": 0.47,
      "grad_norm": 1.0258096153847873e-05,
      "learning_rate": 0.00011958283545239204,
      "loss": 0.0,
      "step": 2940
    },
    {
      "epoch": 0.47,
      "grad_norm": 1.0247929822071455e-05,
      "learning_rate": 0.00011906919405822348,
      "loss": 0.0,
      "step": 2950
    },
    {
      "epoch": 0.47,
      "grad_norm": 1.0227300663245842e-05,
      "learning_rate": 0.00011855503004668704,
      "loss": 0.0,
      "step": 2960
    },
    {
      "epoch": 0.47,
      "grad_norm": 1.0172416295972653e-05,
      "learning_rate": 0.00011804035750915123,
      "loss": 0.0,
      "step": 2970
    },
    {
      "epoch": 0.47,
      "grad_norm": 1.015808720694622e-05,
      "learning_rate": 0.00011752519055092156,
      "loss": 0.0,
      "step": 2980
    },
    {
      "epoch": 0.47,
      "grad_norm": 1.0122079402208328e-05,
      "learning_rate": 0.00011700954329085374,
      "loss": 0.0,
      "step": 2990
    },
    {
      "epoch": 0.47,
      "grad_norm": 1.0053732694359496e-05,
      "learning_rate": 0.0001164934298609668,
      "loss": 0.0,
      "step": 3000
    },
    {
      "epoch": 0.47,
      "eval_loss": 1.2696945077550481e-06,
      "eval_runtime": 677.3894,
      "eval_samples_per_second": 3.928,
      "eval_steps_per_second": 3.928,
      "step": 3000
    },
    {
      "epoch": 0.48,
      "grad_norm": 9.959580893337261e-06,
      "learning_rate": 0.00011597686440605589,
      "loss": 0.0,
      "step": 3010
    },
    {
      "epoch": 0.48,
      "grad_norm": 9.868744200502988e-06,
      "learning_rate": 0.00011545986108330436,
      "loss": 0.0,
      "step": 3020
    },
    {
      "epoch": 0.48,
      "grad_norm": 9.799139661481604e-06,
      "learning_rate": 0.00011494243406189611,
      "loss": 0.0,
      "step": 3030
    },
    {
      "epoch": 0.48,
      "grad_norm": 9.899444194161333e-06,
      "learning_rate": 0.00011442459752262693,
      "loss": 0.0,
      "step": 3040
    },
    {
      "epoch": 0.48,
      "grad_norm": 9.762599802343175e-06,
      "learning_rate": 0.00011390636565751608,
      "loss": 0.0,
      "step": 3050
    },
    {
      "epoch": 0.48,
      "grad_norm": 9.824057997320779e-06,
      "learning_rate": 0.0001133877526694173,
      "loss": 0.0,
      "step": 3060
    },
    {
      "epoch": 0.49,
      "grad_norm": 9.707560820970684e-06,
      "learning_rate": 0.00011286877277162943,
      "loss": 0.0,
      "step": 3070
    },
    {
      "epoch": 0.49,
      "grad_norm": 9.551651601213962e-06,
      "learning_rate": 0.00011234944018750708,
      "loss": 0.0,
      "step": 3080
    },
    {
      "epoch": 0.49,
      "grad_norm": 9.524844244879205e-06,
      "learning_rate": 0.0001118297691500706,
      "loss": 0.0,
      "step": 3090
    },
    {
      "epoch": 0.49,
      "grad_norm": 9.5066816356848e-06,
      "learning_rate": 0.00011130977390161623,
      "loss": 0.0,
      "step": 3100
    },
    {
      "epoch": 0.49,
      "grad_norm": 9.528801456326619e-06,
      "learning_rate": 0.00011078946869332556,
      "loss": 0.0,
      "step": 3110
    },
    {
      "epoch": 0.49,
      "grad_norm": 9.463856258662418e-06,
      "learning_rate": 0.00011026886778487511,
      "loss": 0.0,
      "step": 3120
    },
    {
      "epoch": 0.5,
      "grad_norm": 9.388229955220595e-06,
      "learning_rate": 0.00010974798544404545,
      "loss": 0.0,
      "step": 3130
    },
    {
      "epoch": 0.5,
      "grad_norm": 9.261138075089548e-06,
      "learning_rate": 0.00010922683594633021,
      "loss": 0.0,
      "step": 3140
    },
    {
      "epoch": 0.5,
      "grad_norm": 9.190160199068487e-06,
      "learning_rate": 0.0001087054335745448,
      "loss": 0.0,
      "step": 3150
    },
    {
      "epoch": 0.5,
      "grad_norm": 9.165185474557802e-06,
      "learning_rate": 0.00010818379261843492,
      "loss": 0.0,
      "step": 3160
    },
    {
      "epoch": 0.5,
      "grad_norm": 9.157519343716558e-06,
      "learning_rate": 0.0001076619273742852,
      "loss": 0.0,
      "step": 3170
    },
    {
      "epoch": 0.5,
      "grad_norm": 9.194201084028464e-06,
      "learning_rate": 0.00010713985214452696,
      "loss": 0.0,
      "step": 3180
    },
    {
      "epoch": 0.5,
      "grad_norm": 9.106612196774222e-06,
      "learning_rate": 0.00010661758123734657,
      "loss": 0.0,
      "step": 3190
    },
    {
      "epoch": 0.51,
      "grad_norm": 9.073876753973309e-06,
      "learning_rate": 0.00010609512896629327,
      "loss": 0.0,
      "step": 3200
    },
    {
      "epoch": 0.51,
      "eval_loss": 1.1537576938280836e-06,
      "eval_runtime": 677.6361,
      "eval_samples_per_second": 3.927,
      "eval_steps_per_second": 3.927,
      "step": 3200
    },
    {
      "epoch": 0.51,
      "grad_norm": 8.928679562814068e-06,
      "learning_rate": 0.00010557250964988672,
      "loss": 0.0,
      "step": 3210
    },
    {
      "epoch": 0.51,
      "grad_norm": 8.87604801391717e-06,
      "learning_rate": 0.00010504973761122476,
      "loss": 0.0,
      "step": 3220
    },
    {
      "epoch": 0.51,
      "grad_norm": 8.847343451634515e-06,
      "learning_rate": 0.00010452682717759066,
      "loss": 0.0,
      "step": 3230
    },
    {
      "epoch": 0.51,
      "grad_norm": 8.800505383987911e-06,
      "learning_rate": 0.00010400379268006082,
      "loss": 0.0,
      "step": 3240
    },
    {
      "epoch": 0.51,
      "grad_norm": 8.778545634413604e-06,
      "learning_rate": 0.00010348064845311157,
      "loss": 0.0,
      "step": 3250
    },
    {
      "epoch": 0.52,
      "grad_norm": 8.74461420607986e-06,
      "learning_rate": 0.0001029574088342266,
      "loss": 0.0,
      "step": 3260
    },
    {
      "epoch": 0.52,
      "grad_norm": 8.718725439393893e-06,
      "learning_rate": 0.00010243408816350403,
      "loss": 0.0,
      "step": 3270
    },
    {
      "epoch": 0.52,
      "grad_norm": 8.643019100418314e-06,
      "learning_rate": 0.0001019107007832632,
      "loss": 0.0,
      "step": 3280
    },
    {
      "epoch": 0.52,
      "grad_norm": 8.676522156747524e-06,
      "learning_rate": 0.00010138726103765171,
      "loss": 0.0,
      "step": 3290
    },
    {
      "epoch": 0.52,
      "grad_norm": 8.580357643950265e-06,
      "learning_rate": 0.00010086378327225241,
      "loss": 0.0,
      "step": 3300
    },
    {
      "epoch": 0.52,
      "grad_norm": 8.526895726390649e-06,
      "learning_rate": 0.00010034028183369008,
      "loss": 0.0,
      "step": 3310
    },
    {
      "epoch": 0.53,
      "grad_norm": 8.501358934154268e-06,
      "learning_rate": 9.981677106923822e-05,
      "loss": 0.0,
      "step": 3320
    },
    {
      "epoch": 0.53,
      "grad_norm": 8.426710337516852e-06,
      "learning_rate": 9.929326532642607e-05,
      "loss": 0.0,
      "step": 3330
    },
    {
      "epoch": 0.53,
      "grad_norm": 8.356075340998359e-06,
      "learning_rate": 9.876977895264516e-05,
      "loss": 0.0,
      "step": 3340
    },
    {
      "epoch": 0.53,
      "grad_norm": 8.328726835316047e-06,
      "learning_rate": 9.824632629475613e-05,
      "loss": 0.0,
      "step": 3350
    },
    {
      "epoch": 0.53,
      "grad_norm": 8.304438779305201e-06,
      "learning_rate": 9.77229216986957e-05,
      "loss": 0.0,
      "step": 3360
    },
    {
      "epoch": 0.53,
      "grad_norm": 8.256480214186013e-06,
      "learning_rate": 9.719957950908331e-05,
      "loss": 0.0,
      "step": 3370
    },
    {
      "epoch": 0.54,
      "grad_norm": 8.222721589845605e-06,
      "learning_rate": 9.66763140688281e-05,
      "loss": 0.0,
      "step": 3380
    },
    {
      "epoch": 0.54,
      "grad_norm": 8.20540572021855e-06,
      "learning_rate": 9.615313971873578e-05,
      "loss": 0.0,
      "step": 3390
    },
    {
      "epoch": 0.54,
      "grad_norm": 8.17977161204908e-06,
      "learning_rate": 9.563007079711565e-05,
      "loss": 0.0,
      "step": 3400
    },
    {
      "epoch": 0.54,
      "eval_loss": 1.0527809308769065e-06,
      "eval_runtime": 678.0937,
      "eval_samples_per_second": 3.924,
      "eval_steps_per_second": 3.924,
      "step": 3400
    },
    {
      "epoch": 0.54,
      "grad_norm": 8.110734597721603e-06,
      "learning_rate": 9.510712163938755e-05,
      "loss": 0.0,
      "step": 3410
    },
    {
      "epoch": 0.54,
      "grad_norm": 8.069450814218726e-06,
      "learning_rate": 9.458430657768902e-05,
      "loss": 0.0,
      "step": 3420
    },
    {
      "epoch": 0.54,
      "grad_norm": 8.081680789473467e-06,
      "learning_rate": 9.406163994048253e-05,
      "loss": 0.0,
      "step": 3430
    },
    {
      "epoch": 0.54,
      "grad_norm": 8.110348971968051e-06,
      "learning_rate": 9.353913605216281e-05,
      "loss": 0.0,
      "step": 3440
    },
    {
      "epoch": 0.55,
      "grad_norm": 8.089009497780353e-06,
      "learning_rate": 9.301680923266413e-05,
      "loss": 0.0,
      "step": 3450
    },
    {
      "epoch": 0.55,
      "grad_norm": 8.064496796578169e-06,
      "learning_rate": 9.24946737970681e-05,
      "loss": 0.0,
      "step": 3460
    },
    {
      "epoch": 0.55,
      "grad_norm": 7.925036697997712e-06,
      "learning_rate": 9.197274405521107e-05,
      "loss": 0.0,
      "step": 3470
    },
    {
      "epoch": 0.55,
      "grad_norm": 7.884138540248387e-06,
      "learning_rate": 9.145103431129213e-05,
      "loss": 0.0,
      "step": 3480
    },
    {
      "epoch": 0.55,
      "grad_norm": 8.020449058676604e-06,
      "learning_rate": 9.092955886348097e-05,
      "loss": 0.0,
      "step": 3490
    },
    {
      "epoch": 0.55,
      "grad_norm": 7.85996780905407e-06,
      "learning_rate": 9.040833200352614e-05,
      "loss": 0.0,
      "step": 3500
    },
    {
      "epoch": 0.56,
      "grad_norm": 7.84209078119602e-06,
      "learning_rate": 8.988736801636329e-05,
      "loss": 0.0,
      "step": 3510
    },
    {
      "epoch": 0.56,
      "grad_norm": 7.831266884750221e-06,
      "learning_rate": 8.936668117972359e-05,
      "loss": 0.0,
      "step": 3520
    },
    {
      "epoch": 0.56,
      "grad_norm": 7.775220183248166e-06,
      "learning_rate": 8.884628576374266e-05,
      "loss": 0.0,
      "step": 3530
    },
    {
      "epoch": 0.56,
      "grad_norm": 7.822431143722497e-06,
      "learning_rate": 8.832619603056925e-05,
      "loss": 0.0,
      "step": 3540
    },
    {
      "epoch": 0.56,
      "grad_norm": 7.721299880358856e-06,
      "learning_rate": 8.780642623397449e-05,
      "loss": 0.0,
      "step": 3550
    },
    {
      "epoch": 0.56,
      "grad_norm": 7.75898115534801e-06,
      "learning_rate": 8.728699061896115e-05,
      "loss": 0.0,
      "step": 3560
    },
    {
      "epoch": 0.57,
      "grad_norm": 7.73134070186643e-06,
      "learning_rate": 8.676790342137337e-05,
      "loss": 0.0,
      "step": 3570
    },
    {
      "epoch": 0.57,
      "grad_norm": 7.776122401992325e-06,
      "learning_rate": 8.624917886750638e-05,
      "loss": 0.0,
      "step": 3580
    },
    {
      "epoch": 0.57,
      "grad_norm": 7.709079000051133e-06,
      "learning_rate": 8.573083117371676e-05,
      "loss": 0.0,
      "step": 3590
    },
    {
      "epoch": 0.57,
      "grad_norm": 7.647149686818011e-06,
      "learning_rate": 8.521287454603258e-05,
      "loss": 0.0,
      "step": 3600
    },
    {
      "epoch": 0.57,
      "eval_loss": 9.793853905648575e-07,
      "eval_runtime": 677.7322,
      "eval_samples_per_second": 3.926,
      "eval_steps_per_second": 3.926,
      "step": 3600
    },
    {
      "epoch": 0.57,
      "grad_norm": 7.622395969519857e-06,
      "learning_rate": 8.46953231797643e-05,
      "loss": 0.0,
      "step": 3610
    },
    {
      "epoch": 0.57,
      "grad_norm": 7.592416750412667e-06,
      "learning_rate": 8.417819125911561e-05,
      "loss": 0.0,
      "step": 3620
    },
    {
      "epoch": 0.57,
      "grad_norm": 7.567623015347635e-06,
      "learning_rate": 8.366149295679471e-05,
      "loss": 0.0,
      "step": 3630
    },
    {
      "epoch": 0.58,
      "grad_norm": 7.557383469247725e-06,
      "learning_rate": 8.314524243362594e-05,
      "loss": 0.0,
      "step": 3640
    },
    {
      "epoch": 0.58,
      "grad_norm": 7.501740128645906e-06,
      "learning_rate": 8.262945383816155e-05,
      "loss": 0.0,
      "step": 3650
    },
    {
      "epoch": 0.58,
      "grad_norm": 7.484216439479496e-06,
      "learning_rate": 8.211414130629417e-05,
      "loss": 0.0,
      "step": 3660
    },
    {
      "epoch": 0.58,
      "grad_norm": 7.407651992252795e-06,
      "learning_rate": 8.159931896086913e-05,
      "loss": 0.0,
      "step": 3670
    },
    {
      "epoch": 0.58,
      "grad_norm": 7.390664904960431e-06,
      "learning_rate": 8.108500091129755e-05,
      "loss": 0.0,
      "step": 3680
    },
    {
      "epoch": 0.58,
      "grad_norm": 7.430302503053099e-06,
      "learning_rate": 8.057120125316973e-05,
      "loss": 0.0,
      "step": 3690
    },
    {
      "epoch": 0.59,
      "grad_norm": 7.307650321308756e-06,
      "learning_rate": 8.005793406786857e-05,
      "loss": 0.0,
      "step": 3700
    },
    {
      "epoch": 0.59,
      "grad_norm": 7.302259746211348e-06,
      "learning_rate": 7.954521342218398e-05,
      "loss": 0.0,
      "step": 3710
    },
    {
      "epoch": 0.59,
      "grad_norm": 7.249198915815214e-06,
      "learning_rate": 7.903305336792708e-05,
      "loss": 0.0,
      "step": 3720
    },
    {
      "epoch": 0.59,
      "grad_norm": 7.208569059002912e-06,
      "learning_rate": 7.852146794154539e-05,
      "loss": 0.0,
      "step": 3730
    },
    {
      "epoch": 0.59,
      "grad_norm": 7.189140433183638e-06,
      "learning_rate": 7.801047116373777e-05,
      "loss": 0.0,
      "step": 3740
    },
    {
      "epoch": 0.59,
      "grad_norm": 7.176822691690177e-06,
      "learning_rate": 7.750007703907046e-05,
      "loss": 0.0,
      "step": 3750
    },
    {
      "epoch": 0.6,
      "grad_norm": 7.194201771199005e-06,
      "learning_rate": 7.699029955559319e-05,
      "loss": 0.0,
      "step": 3760
    },
    {
      "epoch": 0.6,
      "grad_norm": 7.143597031245008e-06,
      "learning_rate": 7.648115268445568e-05,
      "loss": 0.0,
      "step": 3770
    },
    {
      "epoch": 0.6,
      "grad_norm": 7.13780127625796e-06,
      "learning_rate": 7.597265037952499e-05,
      "loss": 0.0,
      "step": 3780
    },
    {
      "epoch": 0.6,
      "grad_norm": 7.1177173595060594e-06,
      "learning_rate": 7.546480657700285e-05,
      "loss": 0.0,
      "step": 3790
    },
    {
      "epoch": 0.6,
      "grad_norm": 7.0835039878147654e-06,
      "learning_rate": 7.495763519504394e-05,
      "loss": 0.0,
      "step": 3800
    },
    {
      "epoch": 0.6,
      "eval_loss": 9.064575579031953e-07,
      "eval_runtime": 678.6169,
      "eval_samples_per_second": 3.921,
      "eval_steps_per_second": 3.921,
      "step": 3800
    },
    {
      "epoch": 0.6,
      "grad_norm": 7.067355909384787e-06,
      "learning_rate": 7.445115013337427e-05,
      "loss": 0.0,
      "step": 3810
    },
    {
      "epoch": 0.6,
      "grad_norm": 7.007452950347215e-06,
      "learning_rate": 7.394536527291025e-05,
      "loss": 0.0,
      "step": 3820
    },
    {
      "epoch": 0.61,
      "grad_norm": 6.989830126258312e-06,
      "learning_rate": 7.344029447537845e-05,
      "loss": 0.0,
      "step": 3830
    },
    {
      "epoch": 0.61,
      "grad_norm": 7.00696318745031e-06,
      "learning_rate": 7.293595158293542e-05,
      "loss": 0.0,
      "step": 3840
    },
    {
      "epoch": 0.61,
      "grad_norm": 6.9599195740011055e-06,
      "learning_rate": 7.243235041778858e-05,
      "loss": 0.0,
      "step": 3850
    },
    {
      "epoch": 0.61,
      "grad_norm": 6.901184860907961e-06,
      "learning_rate": 7.192950478181724e-05,
      "loss": 0.0,
      "step": 3860
    },
    {
      "epoch": 0.61,
      "grad_norm": 6.831253358541289e-06,
      "learning_rate": 7.142742845619448e-05,
      "loss": 0.0,
      "step": 3870
    },
    {
      "epoch": 0.61,
      "grad_norm": 6.842563379905187e-06,
      "learning_rate": 7.09261352010093e-05,
      "loss": 0.0,
      "step": 3880
    },
    {
      "epoch": 0.62,
      "grad_norm": 6.8026229200768284e-06,
      "learning_rate": 7.042563875488959e-05,
      "loss": 0.0,
      "step": 3890
    },
    {
      "epoch": 0.62,
      "grad_norm": 6.846538781246636e-06,
      "learning_rate": 6.992595283462563e-05,
      "loss": 0.0,
      "step": 3900
    },
    {
      "epoch": 0.62,
      "grad_norm": 6.835349267930724e-06,
      "learning_rate": 6.942709113479406e-05,
      "loss": 0.0,
      "step": 3910
    },
    {
      "epoch": 0.62,
      "grad_norm": 6.8236986408010125e-06,
      "learning_rate": 6.892906732738271e-05,
      "loss": 0.0,
      "step": 3920
    },
    {
      "epoch": 0.62,
      "grad_norm": 6.792744443373522e-06,
      "learning_rate": 6.843189506141582e-05,
      "loss": 0.0,
      "step": 3930
    },
    {
      "epoch": 0.62,
      "grad_norm": 6.790230600017821e-06,
      "learning_rate": 6.79355879625799e-05,
      "loss": 0.0,
      "step": 3940
    },
    {
      "epoch": 0.63,
      "grad_norm": 6.77120306136203e-06,
      "learning_rate": 6.744015963285048e-05,
      "loss": 0.0,
      "step": 3950
    },
    {
      "epoch": 0.63,
      "grad_norm": 6.727182153554168e-06,
      "learning_rate": 6.694562365011907e-05,
      "loss": 0.0,
      "step": 3960
    },
    {
      "epoch": 0.63,
      "grad_norm": 6.759685220458778e-06,
      "learning_rate": 6.64519935678213e-05,
      "loss": 0.0,
      "step": 3970
    },
    {
      "epoch": 0.63,
      "grad_norm": 6.710036814183695e-06,
      "learning_rate": 6.595928291456537e-05,
      "loss": 0.0,
      "step": 3980
    },
    {
      "epoch": 0.63,
      "grad_norm": 6.672820290987147e-06,
      "learning_rate": 6.546750519376115e-05,
      "loss": 0.0,
      "step": 3990
    },
    {
      "epoch": 0.63,
      "grad_norm": 6.715794370393269e-06,
      "learning_rate": 6.497667388325035e-05,
      "loss": 0.0,
      "step": 4000
    },
    {
      "epoch": 0.63,
      "eval_loss": 8.466194003631244e-07,
      "eval_runtime": 677.4668,
      "eval_samples_per_second": 3.928,
      "eval_steps_per_second": 3.928,
      "step": 4000
    },
    {
      "epoch": 0.63,
      "grad_norm": 6.650427621934796e-06,
      "learning_rate": 6.448680243493697e-05,
      "loss": 0.0,
      "step": 4010
    },
    {
      "epoch": 0.64,
      "grad_norm": 6.633624707319541e-06,
      "learning_rate": 6.399790427441864e-05,
      "loss": 0.0,
      "step": 4020
    },
    {
      "epoch": 0.64,
      "grad_norm": 6.6454022089601494e-06,
      "learning_rate": 6.350999280061872e-05,
      "loss": 0.0,
      "step": 4030
    },
    {
      "epoch": 0.64,
      "grad_norm": 6.565022431459511e-06,
      "learning_rate": 6.302308138541908e-05,
      "loss": 0.0,
      "step": 4040
    },
    {
      "epoch": 0.64,
      "grad_norm": 6.468826995842392e-06,
      "learning_rate": 6.253718337329364e-05,
      "loss": 0.0,
      "step": 4050
    },
    {
      "epoch": 0.64,
      "grad_norm": 6.430604116758332e-06,
      "learning_rate": 6.205231208094252e-05,
      "loss": 0.0,
      "step": 4060
    },
    {
      "epoch": 0.64,
      "grad_norm": 6.451509761973284e-06,
      "learning_rate": 6.156848079692731e-05,
      "loss": 0.0,
      "step": 4070
    },
    {
      "epoch": 0.65,
      "grad_norm": 6.3997758843470365e-06,
      "learning_rate": 6.10857027813067e-05,
      "loss": 0.0,
      "step": 4080
    },
    {
      "epoch": 0.65,
      "grad_norm": 6.348835086100735e-06,
      "learning_rate": 6.0603991265273074e-05,
      "loss": 0.0,
      "step": 4090
    },
    {
      "epoch": 0.65,
      "grad_norm": 6.375953944370849e-06,
      "learning_rate": 6.0123359450789975e-05,
      "loss": 0.0,
      "step": 4100
    },
    {
      "epoch": 0.65,
      "grad_norm": 6.320291959127644e-06,
      "learning_rate": 5.964382051023024e-05,
      "loss": 0.0,
      "step": 4110
    },
    {
      "epoch": 0.65,
      "grad_norm": 6.349788691295544e-06,
      "learning_rate": 5.916538758601502e-05,
      "loss": 0.0,
      "step": 4120
    },
    {
      "epoch": 0.65,
      "grad_norm": 6.3071929616853595e-06,
      "learning_rate": 5.8688073790253494e-05,
      "loss": 0.0,
      "step": 4130
    },
    {
      "epoch": 0.66,
      "grad_norm": 6.314509391813772e-06,
      "learning_rate": 5.821189220438371e-05,
      "loss": 0.0,
      "step": 4140
    },
    {
      "epoch": 0.66,
      "grad_norm": 6.26232713329955e-06,
      "learning_rate": 5.773685587881389e-05,
      "loss": 0.0,
      "step": 4150
    },
    {
      "epoch": 0.66,
      "grad_norm": 6.290351393545279e-06,
      "learning_rate": 5.7262977832564826e-05,
      "loss": 0.0,
      "step": 4160
    },
    {
      "epoch": 0.66,
      "grad_norm": 6.278557066252688e-06,
      "learning_rate": 5.679027105291312e-05,
      "loss": 0.0,
      "step": 4170
    },
    {
      "epoch": 0.66,
      "grad_norm": 6.206881607795367e-06,
      "learning_rate": 5.6318748495035136e-05,
      "loss": 0.0,
      "step": 4180
    },
    {
      "epoch": 0.66,
      "grad_norm": 6.211484105733689e-06,
      "learning_rate": 5.584842308165207e-05,
      "loss": 0.0,
      "step": 4190
    },
    {
      "epoch": 0.66,
      "grad_norm": 6.166925231809728e-06,
      "learning_rate": 5.5379307702675696e-05,
      "loss": 0.0,
      "step": 4200
    },
    {
      "epoch": 0.66,
      "eval_loss": 7.956632543937303e-07,
      "eval_runtime": 677.4881,
      "eval_samples_per_second": 3.928,
      "eval_steps_per_second": 3.928,
      "step": 4200
    },
    {
      "epoch": 0.67,
      "grad_norm": 6.2032509049458895e-06,
      "learning_rate": 5.491141521485525e-05,
      "loss": 0.0,
      "step": 4210
    },
    {
      "epoch": 0.67,
      "grad_norm": 6.100659902585903e-06,
      "learning_rate": 5.4444758441424894e-05,
      "loss": 0.0,
      "step": 4220
    },
    {
      "epoch": 0.67,
      "grad_norm": 6.096037850511493e-06,
      "learning_rate": 5.3979350171752244e-05,
      "loss": 0.0,
      "step": 4230
    },
    {
      "epoch": 0.67,
      "grad_norm": 6.062749434931902e-06,
      "learning_rate": 5.351520316098808e-05,
      "loss": 0.0,
      "step": 4240
    },
    {
      "epoch": 0.67,
      "grad_norm": 6.080766524974024e-06,
      "learning_rate": 5.3052330129716654e-05,
      "loss": 0.0,
      "step": 4250
    },
    {
      "epoch": 0.67,
      "grad_norm": 6.079821559978882e-06,
      "learning_rate": 5.259074376360701e-05,
      "loss": 0.0,
      "step": 4260
    },
    {
      "epoch": 0.68,
      "grad_norm": 6.042651421012124e-06,
      "learning_rate": 5.2130456713065415e-05,
      "loss": 0.0,
      "step": 4270
    },
    {
      "epoch": 0.68,
      "grad_norm": 6.043023404345149e-06,
      "learning_rate": 5.1671481592888595e-05,
      "loss": 0.0,
      "step": 4280
    },
    {
      "epoch": 0.68,
      "grad_norm": 5.995801984681748e-06,
      "learning_rate": 5.1213830981918034e-05,
      "loss": 0.0,
      "step": 4290
    },
    {
      "epoch": 0.68,
      "grad_norm": 5.988891189190326e-06,
      "learning_rate": 5.0757517422695214e-05,
      "loss": 0.0,
      "step": 4300
    },
    {
      "epoch": 0.68,
      "grad_norm": 5.986105861666147e-06,
      "learning_rate": 5.0302553421117913e-05,
      "loss": 0.0,
      "step": 4310
    },
    {
      "epoch": 0.68,
      "grad_norm": 5.938798494753428e-06,
      "learning_rate": 4.984895144609743e-05,
      "loss": 0.0,
      "step": 4320
    },
    {
      "epoch": 0.69,
      "grad_norm": 5.95130632063956e-06,
      "learning_rate": 4.9396723929216784e-05,
      "loss": 0.0,
      "step": 4330
    },
    {
      "epoch": 0.69,
      "grad_norm": 5.944618806097424e-06,
      "learning_rate": 4.8945883264390217e-05,
      "loss": 0.0,
      "step": 4340
    },
    {
      "epoch": 0.69,
      "grad_norm": 5.940526989434147e-06,
      "learning_rate": 4.849644180752331e-05,
      "loss": 0.0,
      "step": 4350
    },
    {
      "epoch": 0.69,
      "grad_norm": 5.880401658941992e-06,
      "learning_rate": 4.804841187617448e-05,
      "loss": 0.0,
      "step": 4360
    },
    {
      "epoch": 0.69,
      "grad_norm": 5.834131570736645e-06,
      "learning_rate": 4.7601805749217267e-05,
      "loss": 0.0,
      "step": 4370
    },
    {
      "epoch": 0.69,
      "grad_norm": 5.80765390623128e-06,
      "learning_rate": 4.715663566650398e-05,
      "loss": 0.0,
      "step": 4380
    },
    {
      "epoch": 0.69,
      "grad_norm": 5.785386747447774e-06,
      "learning_rate": 4.6712913828530156e-05,
      "loss": 0.0,
      "step": 4390
    },
    {
      "epoch": 0.7,
      "grad_norm": 5.7291763368994e-06,
      "learning_rate": 4.62706523961002e-05,
      "loss": 0.0,
      "step": 4400
    },
    {
      "epoch": 0.7,
      "eval_loss": 7.554591547886957e-07,
      "eval_runtime": 676.6257,
      "eval_samples_per_second": 3.933,
      "eval_steps_per_second": 3.933,
      "step": 4400
    },
    {
      "epoch": 0.7,
      "grad_norm": 5.772773874923587e-06,
      "learning_rate": 4.5829863489994105e-05,
      "loss": 0.0,
      "step": 4410
    },
    {
      "epoch": 0.7,
      "grad_norm": 5.740222150052432e-06,
      "learning_rate": 4.539055919063526e-05,
      "loss": 0.0,
      "step": 4420
    },
    {
      "epoch": 0.7,
      "grad_norm": 5.725092705688439e-06,
      "learning_rate": 4.495275153775937e-05,
      "loss": 0.0,
      "step": 4430
    },
    {
      "epoch": 0.7,
      "grad_norm": 5.695530944649363e-06,
      "learning_rate": 4.451645253008451e-05,
      "loss": 0.0,
      "step": 4440
    },
    {
      "epoch": 0.7,
      "grad_norm": 5.710885488952044e-06,
      "learning_rate": 4.408167412498224e-05,
      "loss": 0.0,
      "step": 4450
    },
    {
      "epoch": 0.71,
      "grad_norm": 5.695631443813909e-06,
      "learning_rate": 4.3648428238149985e-05,
      "loss": 0.0,
      "step": 4460
    },
    {
      "epoch": 0.71,
      "grad_norm": 5.678127763530938e-06,
      "learning_rate": 4.3216726743284266e-05,
      "loss": 0.0,
      "step": 4470
    },
    {
      "epoch": 0.71,
      "grad_norm": 5.683846666215686e-06,
      "learning_rate": 4.278658147175561e-05,
      "loss": 0.0,
      "step": 4480
    },
    {
      "epoch": 0.71,
      "grad_norm": 5.663964202540228e-06,
      "learning_rate": 4.235800421228403e-05,
      "loss": 0.0,
      "step": 4490
    },
    {
      "epoch": 0.71,
      "grad_norm": 5.672467068507103e-06,
      "learning_rate": 4.193100671061604e-05,
      "loss": 0.0,
      "step": 4500
    },
    {
      "epoch": 0.71,
      "grad_norm": 5.680505182681372e-06,
      "learning_rate": 4.1505600669202626e-05,
      "loss": 0.0,
      "step": 4510
    },
    {
      "epoch": 0.72,
      "grad_norm": 5.664197487931233e-06,
      "learning_rate": 4.108179774687877e-05,
      "loss": 0.0,
      "step": 4520
    },
    {
      "epoch": 0.72,
      "grad_norm": 5.6741005209914874e-06,
      "learning_rate": 4.065960955854371e-05,
      "loss": 0.0,
      "step": 4530
    },
    {
      "epoch": 0.72,
      "grad_norm": 5.665102889906848e-06,
      "learning_rate": 4.0239047674842676e-05,
      "loss": 0.0,
      "step": 4540
    },
    {
      "epoch": 0.72,
      "grad_norm": 5.6059820963128e-06,
      "learning_rate": 3.9820123621849914e-05,
      "loss": 0.0,
      "step": 4550
    },
    {
      "epoch": 0.72,
      "grad_norm": 5.637057711282978e-06,
      "learning_rate": 3.9402848880752495e-05,
      "loss": 0.0,
      "step": 4560
    },
    {
      "epoch": 0.72,
      "grad_norm": 5.659976523020305e-06,
      "learning_rate": 3.898723488753594e-05,
      "loss": 0.0,
      "step": 4570
    },
    {
      "epoch": 0.72,
      "grad_norm": 5.5967493608477525e-06,
      "learning_rate": 3.85732930326707e-05,
      "loss": 0.0,
      "step": 4580
    },
    {
      "epoch": 0.73,
      "grad_norm": 5.650239927490475e-06,
      "learning_rate": 3.816103466079996e-05,
      "loss": 0.0,
      "step": 4590
    },
    {
      "epoch": 0.73,
      "grad_norm": 5.624555797112407e-06,
      "learning_rate": 3.775047107042883e-05,
      "loss": 0.0,
      "step": 4600
    },
    {
      "epoch": 0.73,
      "eval_loss": 7.241376920319453e-07,
      "eval_runtime": 679.1475,
      "eval_samples_per_second": 3.918,
      "eval_steps_per_second": 3.918,
      "step": 4600
    },
    {
      "epoch": 0.73,
      "grad_norm": 5.63035382583621e-06,
      "learning_rate": 3.73416135136144e-05,
      "loss": 0.0,
      "step": 4610
    },
    {
      "epoch": 0.73,
      "grad_norm": 5.586789939115988e-06,
      "learning_rate": 3.693447319565785e-05,
      "loss": 0.0,
      "step": 4620
    },
    {
      "epoch": 0.73,
      "grad_norm": 5.556290034292033e-06,
      "learning_rate": 3.6529061274796904e-05,
      "loss": 0.0,
      "step": 4630
    },
    {
      "epoch": 0.73,
      "grad_norm": 5.571641395363258e-06,
      "learning_rate": 3.612538886190028e-05,
      "loss": 0.0,
      "step": 4640
    },
    {
      "epoch": 0.74,
      "grad_norm": 5.598154984909343e-06,
      "learning_rate": 3.572346702016299e-05,
      "loss": 0.0,
      "step": 4650
    },
    {
      "epoch": 0.74,
      "grad_norm": 5.605736532743322e-06,
      "learning_rate": 3.532330676480335e-05,
      "loss": 0.0,
      "step": 4660
    },
    {
      "epoch": 0.74,
      "grad_norm": 5.584050995821599e-06,
      "learning_rate": 3.492491906276093e-05,
      "loss": 0.0,
      "step": 4670
    },
    {
      "epoch": 0.74,
      "grad_norm": 5.5939735830179416e-06,
      "learning_rate": 3.4528314832396134e-05,
      "loss": 0.0,
      "step": 4680
    },
    {
      "epoch": 0.74,
      "grad_norm": 5.486252575792605e-06,
      "learning_rate": 3.413350494319083e-05,
      "loss": 0.0,
      "step": 4690
    },
    {
      "epoch": 0.74,
      "grad_norm": 5.496313406183617e-06,
      "learning_rate": 3.374050021545045e-05,
      "loss": 0.0,
      "step": 4700
    },
    {
      "epoch": 0.75,
      "grad_norm": 5.51309494767338e-06,
      "learning_rate": 3.334931142000759e-05,
      "loss": 0.0,
      "step": 4710
    },
    {
      "epoch": 0.75,
      "grad_norm": 5.535066975426162e-06,
      "learning_rate": 3.295994927792669e-05,
      "loss": 0.0,
      "step": 4720
    },
    {
      "epoch": 0.75,
      "grad_norm": 5.510259597940603e-06,
      "learning_rate": 3.257242446021029e-05,
      "loss": 0.0,
      "step": 4730
    },
    {
      "epoch": 0.75,
      "grad_norm": 5.482118922373047e-06,
      "learning_rate": 3.2186747587506514e-05,
      "loss": 0.0,
      "step": 4740
    },
    {
      "epoch": 0.75,
      "grad_norm": 5.511979452421656e-06,
      "learning_rate": 3.180292922981805e-05,
      "loss": 0.0,
      "step": 4750
    },
    {
      "epoch": 0.75,
      "grad_norm": 5.525758297153516e-06,
      "learning_rate": 3.142097990621238e-05,
      "loss": 0.0,
      "step": 4760
    },
    {
      "epoch": 0.76,
      "grad_norm": 5.4763736443419475e-06,
      "learning_rate": 3.1040910084533614e-05,
      "loss": 0.0,
      "step": 4770
    },
    {
      "epoch": 0.76,
      "grad_norm": 5.465680260385852e-06,
      "learning_rate": 3.066273018111551e-05,
      "loss": 0.0,
      "step": 4780
    },
    {
      "epoch": 0.76,
      "grad_norm": 5.468723884405335e-06,
      "learning_rate": 3.028645056049606e-05,
      "loss": 0.0,
      "step": 4790
    },
    {
      "epoch": 0.76,
      "grad_norm": 5.448248884931672e-06,
      "learning_rate": 2.9912081535133297e-05,
      "loss": 0.0,
      "step": 4800
    },
    {
      "epoch": 0.76,
      "eval_loss": 6.918809845046781e-07,
      "eval_runtime": 676.7787,
      "eval_samples_per_second": 3.932,
      "eval_steps_per_second": 3.932,
      "step": 4800
    },
    {
      "epoch": 0.76,
      "grad_norm": 5.457012321130605e-06,
      "learning_rate": 2.9539633365122877e-05,
      "loss": 0.0,
      "step": 4810
    },
    {
      "epoch": 0.76,
      "grad_norm": 5.470580163091654e-06,
      "learning_rate": 2.9169116257916808e-05,
      "loss": 0.0,
      "step": 4820
    },
    {
      "epoch": 0.76,
      "grad_norm": 5.4234992603596766e-06,
      "learning_rate": 2.8800540368043638e-05,
      "loss": 0.0,
      "step": 4830
    },
    {
      "epoch": 0.77,
      "grad_norm": 5.431846147985198e-06,
      "learning_rate": 2.8433915796830135e-05,
      "loss": 0.0,
      "step": 4840
    },
    {
      "epoch": 0.77,
      "grad_norm": 5.3348458095570095e-06,
      "learning_rate": 2.8069252592124583e-05,
      "loss": 0.0,
      "step": 4850
    },
    {
      "epoch": 0.77,
      "grad_norm": 5.307606897986261e-06,
      "learning_rate": 2.7706560748021315e-05,
      "loss": 0.0,
      "step": 4860
    },
    {
      "epoch": 0.77,
      "grad_norm": 5.340937150322134e-06,
      "learning_rate": 2.734585020458683e-05,
      "loss": 0.0,
      "step": 4870
    },
    {
      "epoch": 0.77,
      "grad_norm": 5.310743290465325e-06,
      "learning_rate": 2.698713084758737e-05,
      "loss": 0.0,
      "step": 4880
    },
    {
      "epoch": 0.77,
      "grad_norm": 5.335717105481308e-06,
      "learning_rate": 2.6630412508217974e-05,
      "loss": 0.0,
      "step": 4890
    },
    {
      "epoch": 0.78,
      "grad_norm": 5.354706445359625e-06,
      "learning_rate": 2.627570496283306e-05,
      "loss": 0.0,
      "step": 4900
    },
    {
      "epoch": 0.78,
      "grad_norm": 5.330066414899193e-06,
      "learning_rate": 2.5923017932678494e-05,
      "loss": 0.0,
      "step": 4910
    },
    {
      "epoch": 0.78,
      "grad_norm": 5.34015543962596e-06,
      "learning_rate": 2.557236108362514e-05,
      "loss": 0.0,
      "step": 4920
    },
    {
      "epoch": 0.78,
      "grad_norm": 5.212670657783747e-06,
      "learning_rate": 2.5223744025904006e-05,
      "loss": 0.0,
      "step": 4930
    },
    {
      "epoch": 0.78,
      "grad_norm": 5.223550488153705e-06,
      "learning_rate": 2.4877176313842753e-05,
      "loss": 0.0,
      "step": 4940
    },
    {
      "epoch": 0.78,
      "grad_norm": 5.22407617609133e-06,
      "learning_rate": 2.453266744560395e-05,
      "loss": 0.0,
      "step": 4950
    },
    {
      "epoch": 0.79,
      "grad_norm": 5.1980200623802375e-06,
      "learning_rate": 2.4190226862924826e-05,
      "loss": 0.0,
      "step": 4960
    },
    {
      "epoch": 0.79,
      "grad_norm": 5.193377546675038e-06,
      "learning_rate": 2.3849863950858298e-05,
      "loss": 0.0,
      "step": 4970
    },
    {
      "epoch": 0.79,
      "grad_norm": 5.195761787035735e-06,
      "learning_rate": 2.3511588037515886e-05,
      "loss": 0.0,
      "step": 4980
    },
    {
      "epoch": 0.79,
      "grad_norm": 5.1700949370570015e-06,
      "learning_rate": 2.3175408393812082e-05,
      "loss": 0.0,
      "step": 4990
    },
    {
      "epoch": 0.79,
      "grad_norm": 5.151221557753161e-06,
      "learning_rate": 2.2841334233210255e-05,
      "loss": 0.0,
      "step": 5000
    },
    {
      "epoch": 0.79,
      "eval_loss": 6.666368221885932e-07,
      "eval_runtime": 678.3655,
      "eval_samples_per_second": 3.923,
      "eval_steps_per_second": 3.923,
      "step": 5000
    },
    {
      "epoch": 0.79,
      "grad_norm": 5.246104592515621e-06,
      "learning_rate": 2.250937471147009e-05,
      "loss": 0.0,
      "step": 5010
    },
    {
      "epoch": 0.79,
      "grad_norm": 5.123742084833793e-06,
      "learning_rate": 2.2179538926396726e-05,
      "loss": 0.0,
      "step": 5020
    },
    {
      "epoch": 0.8,
      "grad_norm": 5.122204584040446e-06,
      "learning_rate": 2.185183591759138e-05,
      "loss": 0.0,
      "step": 5030
    },
    {
      "epoch": 0.8,
      "grad_norm": 5.1354249990254175e-06,
      "learning_rate": 2.1526274666203617e-05,
      "loss": 0.0,
      "step": 5040
    },
    {
      "epoch": 0.8,
      "grad_norm": 5.090547801955836e-06,
      "learning_rate": 2.1202864094685217e-05,
      "loss": 0.0,
      "step": 5050
    },
    {
      "epoch": 0.8,
      "grad_norm": 5.138384494784987e-06,
      "learning_rate": 2.08816130665456e-05,
      "loss": 0.0,
      "step": 5060
    },
    {
      "epoch": 0.8,
      "grad_norm": 5.1392339628364425e-06,
      "learning_rate": 2.0562530386109026e-05,
      "loss": 0.0,
      "step": 5070
    },
    {
      "epoch": 0.8,
      "grad_norm": 5.080198661744362e-06,
      "learning_rate": 2.0245624798273066e-05,
      "loss": 0.0,
      "step": 5080
    },
    {
      "epoch": 0.81,
      "grad_norm": 5.097656412544893e-06,
      "learning_rate": 1.993090498826925e-05,
      "loss": 0.0,
      "step": 5090
    },
    {
      "epoch": 0.81,
      "grad_norm": 5.086645614937879e-06,
      "learning_rate": 1.9618379581424805e-05,
      "loss": 0.0,
      "step": 5100
    },
    {
      "epoch": 0.81,
      "grad_norm": 5.065458481112728e-06,
      "learning_rate": 1.930805714292634e-05,
      "loss": 0.0,
      "step": 5110
    },
    {
      "epoch": 0.81,
      "grad_norm": 5.060294824943412e-06,
      "learning_rate": 1.8999946177585072e-05,
      "loss": 0.0,
      "step": 5120
    },
    {
      "epoch": 0.81,
      "grad_norm": 5.091289949632483e-06,
      "learning_rate": 1.8694055129603794e-05,
      "loss": 0.0,
      "step": 5130
    },
    {
      "epoch": 0.81,
      "grad_norm": 5.040332325734198e-06,
      "learning_rate": 1.8390392382345435e-05,
      "loss": 0.0,
      "step": 5140
    },
    {
      "epoch": 0.82,
      "grad_norm": 5.0679514060902875e-06,
      "learning_rate": 1.8088966258103268e-05,
      "loss": 0.0,
      "step": 5150
    },
    {
      "epoch": 0.82,
      "grad_norm": 5.058467195340199e-06,
      "learning_rate": 1.7789785017872916e-05,
      "loss": 0.0,
      "step": 5160
    },
    {
      "epoch": 0.82,
      "grad_norm": 5.010219410905847e-06,
      "learning_rate": 1.7492856861125783e-05,
      "loss": 0.0,
      "step": 5170
    },
    {
      "epoch": 0.82,
      "grad_norm": 5.0275575631530955e-06,
      "learning_rate": 1.7198189925584507e-05,
      "loss": 0.0,
      "step": 5180
    },
    {
      "epoch": 0.82,
      "grad_norm": 5.025058726459974e-06,
      "learning_rate": 1.6905792286999823e-05,
      "loss": 0.0,
      "step": 5190
    },
    {
      "epoch": 0.82,
      "grad_norm": 5.064585366199026e-06,
      "learning_rate": 1.66156719589293e-05,
      "loss": 0.0,
      "step": 5200
    },
    {
      "epoch": 0.82,
      "eval_loss": 6.479372700596286e-07,
      "eval_runtime": 677.7823,
      "eval_samples_per_second": 3.926,
      "eval_steps_per_second": 3.926,
      "step": 5200
    },
    {
      "epoch": 0.82,
      "grad_norm": 5.059666364104487e-06,
      "learning_rate": 1.632783689251771e-05,
      "loss": 0.0,
      "step": 5210
    },
    {
      "epoch": 0.83,
      "grad_norm": 5.040225005359389e-06,
      "learning_rate": 1.6042294976279025e-05,
      "loss": 0.0,
      "step": 5220
    },
    {
      "epoch": 0.83,
      "grad_norm": 5.047268587077269e-06,
      "learning_rate": 1.5759054035880418e-05,
      "loss": 0.0,
      "step": 5230
    },
    {
      "epoch": 0.83,
      "grad_norm": 5.036600214225473e-06,
      "learning_rate": 1.5478121833927563e-05,
      "loss": 0.0,
      "step": 5240
    },
    {
      "epoch": 0.83,
      "grad_norm": 5.034959940530825e-06,
      "learning_rate": 1.5199506069752078e-05,
      "loss": 0.0,
      "step": 5250
    },
    {
      "epoch": 0.83,
      "grad_norm": 4.9835912250273395e-06,
      "learning_rate": 1.4923214379200335e-05,
      "loss": 0.0,
      "step": 5260
    },
    {
      "epoch": 0.83,
      "grad_norm": 5.027083261666121e-06,
      "learning_rate": 1.4649254334424355e-05,
      "loss": 0.0,
      "step": 5270
    },
    {
      "epoch": 0.84,
      "grad_norm": 4.9774353101383895e-06,
      "learning_rate": 1.4377633443674233e-05,
      "loss": 0.0,
      "step": 5280
    },
    {
      "epoch": 0.84,
      "grad_norm": 4.998076747142477e-06,
      "learning_rate": 1.4108359151092277e-05,
      "loss": 0.0,
      "step": 5290
    },
    {
      "epoch": 0.84,
      "grad_norm": 5.016227532905759e-06,
      "learning_rate": 1.3841438836509157e-05,
      "loss": 0.0,
      "step": 5300
    },
    {
      "epoch": 0.84,
      "grad_norm": 5.005045295547461e-06,
      "learning_rate": 1.3576879815241439e-05,
      "loss": 0.0,
      "step": 5310
    },
    {
      "epoch": 0.84,
      "grad_norm": 5.001534645998618e-06,
      "learning_rate": 1.3314689337891285e-05,
      "loss": 0.0,
      "step": 5320
    },
    {
      "epoch": 0.84,
      "grad_norm": 5.0081639528798405e-06,
      "learning_rate": 1.3054874590147648e-05,
      "loss": 0.0,
      "step": 5330
    },
    {
      "epoch": 0.85,
      "grad_norm": 4.970066584064625e-06,
      "learning_rate": 1.2797442692589378e-05,
      "loss": 0.0,
      "step": 5340
    },
    {
      "epoch": 0.85,
      "grad_norm": 4.98696681461297e-06,
      "learning_rate": 1.254240070049003e-05,
      "loss": 0.0,
      "step": 5350
    },
    {
      "epoch": 0.85,
      "grad_norm": 4.9790855882747564e-06,
      "learning_rate": 1.2289755603624542e-05,
      "loss": 0.0,
      "step": 5360
    },
    {
      "epoch": 0.85,
      "grad_norm": 4.94319328936399e-06,
      "learning_rate": 1.203951432607764e-05,
      "loss": 0.0,
      "step": 5370
    },
    {
      "epoch": 0.85,
      "grad_norm": 4.973221166437725e-06,
      "learning_rate": 1.1791683726054114e-05,
      "loss": 0.0,
      "step": 5380
    },
    {
      "epoch": 0.85,
      "grad_norm": 4.990554771211464e-06,
      "learning_rate": 1.1546270595690822e-05,
      "loss": 0.0,
      "step": 5390
    },
    {
      "epoch": 0.85,
      "grad_norm": 4.974987405148568e-06,
      "learning_rate": 1.1303281660870591e-05,
      "loss": 0.0,
      "step": 5400
    },
    {
      "epoch": 0.85,
      "eval_loss": 6.367176297317201e-07,
      "eval_runtime": 677.1253,
      "eval_samples_per_second": 3.93,
      "eval_steps_per_second": 3.93,
      "step": 5400
    },
    {
      "epoch": 0.86,
      "grad_norm": 4.9774153012549505e-06,
      "learning_rate": 1.1062723581037759e-05,
      "loss": 0.0,
      "step": 5410
    },
    {
      "epoch": 0.86,
      "grad_norm": 4.9803270485426765e-06,
      "learning_rate": 1.0824602949015816e-05,
      "loss": 0.0,
      "step": 5420
    },
    {
      "epoch": 0.86,
      "grad_norm": 4.961484592058696e-06,
      "learning_rate": 1.0588926290826717e-05,
      "loss": 0.0,
      "step": 5430
    },
    {
      "epoch": 0.86,
      "grad_norm": 4.986853582522599e-06,
      "learning_rate": 1.0355700065511898e-05,
      "loss": 0.0,
      "step": 5440
    },
    {
      "epoch": 0.86,
      "grad_norm": 4.96260508953128e-06,
      "learning_rate": 1.0124930664955301e-05,
      "loss": 0.0,
      "step": 5450
    },
    {
      "epoch": 0.86,
      "grad_norm": 4.938865004078252e-06,
      "learning_rate": 9.896624413708278e-06,
      "loss": 0.0,
      "step": 5460
    },
    {
      "epoch": 0.87,
      "grad_norm": 4.9825462156150024e-06,
      "learning_rate": 9.670787568816175e-06,
      "loss": 0.0,
      "step": 5470
    },
    {
      "epoch": 0.87,
      "grad_norm": 4.957390956406016e-06,
      "learning_rate": 9.447426319646902e-06,
      "loss": 0.0,
      "step": 5480
    },
    {
      "epoch": 0.87,
      "grad_norm": 4.943620297126472e-06,
      "learning_rate": 9.226546787721257e-06,
      "loss": 0.0,
      "step": 5490
    },
    {
      "epoch": 0.87,
      "grad_norm": 4.928577709506499e-06,
      "learning_rate": 9.008155026545184e-06,
      "loss": 0.0,
      "step": 5500
    },
    {
      "epoch": 0.87,
      "grad_norm": 4.950798484060215e-06,
      "learning_rate": 8.792257021443872e-06,
      "loss": 0.0,
      "step": 5510
    },
    {
      "epoch": 0.87,
      "grad_norm": 4.966905635228613e-06,
      "learning_rate": 8.578858689397695e-06,
      "loss": 0.0,
      "step": 5520
    },
    {
      "epoch": 0.88,
      "grad_norm": 4.940853159496328e-06,
      "learning_rate": 8.36796587888007e-06,
      "loss": 0.0,
      "step": 5530
    },
    {
      "epoch": 0.88,
      "grad_norm": 4.948326022713445e-06,
      "learning_rate": 8.159584369697205e-06,
      "loss": 0.0,
      "step": 5540
    },
    {
      "epoch": 0.88,
      "grad_norm": 4.948916739522247e-06,
      "learning_rate": 7.953719872829568e-06,
      "loss": 0.0,
      "step": 5550
    },
    {
      "epoch": 0.88,
      "grad_norm": 4.938838628731901e-06,
      "learning_rate": 7.75037803027554e-06,
      "loss": 0.0,
      "step": 5560
    },
    {
      "epoch": 0.88,
      "grad_norm": 4.90622232973692e-06,
      "learning_rate": 7.549564414896692e-06,
      "loss": 0.0,
      "step": 5570
    },
    {
      "epoch": 0.88,
      "grad_norm": 4.9328236855217256e-06,
      "learning_rate": 7.351284530265057e-06,
      "loss": 0.0,
      "step": 5580
    },
    {
      "epoch": 0.88,
      "grad_norm": 4.9284103624813724e-06,
      "learning_rate": 7.1555438105123e-06,
      "loss": 0.0,
      "step": 5590
    },
    {
      "epoch": 0.89,
      "grad_norm": 4.904401521343971e-06,
      "learning_rate": 6.962347620180787e-06,
      "loss": 0.0,
      "step": 5600
    },
    {
      "epoch": 0.89,
      "eval_loss": 6.259655265239417e-07,
      "eval_runtime": 677.9255,
      "eval_samples_per_second": 3.925,
      "eval_steps_per_second": 3.925,
      "step": 5600
    },
    {
      "epoch": 0.89,
      "grad_norm": 4.910792085865978e-06,
      "learning_rate": 6.771701254076623e-06,
      "loss": 0.0,
      "step": 5610
    },
    {
      "epoch": 0.89,
      "grad_norm": 4.910984443995403e-06,
      "learning_rate": 6.583609937124435e-06,
      "loss": 0.0,
      "step": 5620
    },
    {
      "epoch": 0.89,
      "grad_norm": 4.911421456199605e-06,
      "learning_rate": 6.398078824224252e-06,
      "loss": 0.0,
      "step": 5630
    },
    {
      "epoch": 0.89,
      "grad_norm": 4.912932581646601e-06,
      "learning_rate": 6.215113000110217e-06,
      "loss": 0.0,
      "step": 5640
    },
    {
      "epoch": 0.89,
      "grad_norm": 4.893081495538354e-06,
      "learning_rate": 6.034717479211205e-06,
      "loss": 0.0,
      "step": 5650
    },
    {
      "epoch": 0.9,
      "grad_norm": 4.8864626478462014e-06,
      "learning_rate": 5.856897205513423e-06,
      "loss": 0.0,
      "step": 5660
    },
    {
      "epoch": 0.9,
      "grad_norm": 4.90334423375316e-06,
      "learning_rate": 5.6816570524248936e-06,
      "loss": 0.0,
      "step": 5670
    },
    {
      "epoch": 0.9,
      "grad_norm": 4.878108484263066e-06,
      "learning_rate": 5.509001822641913e-06,
      "loss": 0.0,
      "step": 5680
    },
    {
      "epoch": 0.9,
      "grad_norm": 4.879219432041282e-06,
      "learning_rate": 5.338936248017401e-06,
      "loss": 0.0,
      "step": 5690
    },
    {
      "epoch": 0.9,
      "grad_norm": 4.860779426962836e-06,
      "learning_rate": 5.17146498943123e-06,
      "loss": 0.0,
      "step": 5700
    },
    {
      "epoch": 0.9,
      "grad_norm": 4.886381248070393e-06,
      "learning_rate": 5.006592636662522e-06,
      "loss": 0.0,
      "step": 5710
    },
    {
      "epoch": 0.91,
      "grad_norm": 4.866678409598535e-06,
      "learning_rate": 4.844323708263821e-06,
      "loss": 0.0,
      "step": 5720
    },
    {
      "epoch": 0.91,
      "grad_norm": 4.864375114266295e-06,
      "learning_rate": 4.684662651437211e-06,
      "loss": 0.0,
      "step": 5730
    },
    {
      "epoch": 0.91,
      "grad_norm": 4.8656829676474445e-06,
      "learning_rate": 4.527613841912548e-06,
      "loss": 0.0,
      "step": 5740
    },
    {
      "epoch": 0.91,
      "grad_norm": 4.792886556970188e-06,
      "learning_rate": 4.373181583827435e-06,
      "loss": 0.0,
      "step": 5750
    },
    {
      "epoch": 0.91,
      "grad_norm": 4.8668603085388895e-06,
      "learning_rate": 4.221370109609302e-06,
      "loss": 0.0,
      "step": 5760
    },
    {
      "epoch": 0.91,
      "grad_norm": 4.859807631873991e-06,
      "learning_rate": 4.072183579859468e-06,
      "loss": 0.0,
      "step": 5770
    },
    {
      "epoch": 0.91,
      "grad_norm": 4.887266186415218e-06,
      "learning_rate": 3.925626083238976e-06,
      "loss": 0.0,
      "step": 5780
    },
    {
      "epoch": 0.92,
      "grad_norm": 4.8928186515695415e-06,
      "learning_rate": 3.7817016363566493e-06,
      "loss": 0.0,
      "step": 5790
    },
    {
      "epoch": 0.92,
      "grad_norm": 4.834650553675601e-06,
      "learning_rate": 3.640414183658991e-06,
      "loss": 0.0,
      "step": 5800
    },
    {
      "epoch": 0.92,
      "eval_loss": 6.198881123964384e-07,
      "eval_runtime": 677.5476,
      "eval_samples_per_second": 3.927,
      "eval_steps_per_second": 3.927,
      "step": 5800
    },
    {
      "epoch": 0.92,
      "grad_norm": 4.888122475676937e-06,
      "learning_rate": 3.5017675973220586e-06,
      "loss": 0.0,
      "step": 5810
    },
    {
      "epoch": 0.92,
      "grad_norm": 4.765689027408371e-06,
      "learning_rate": 3.3657656771453495e-06,
      "loss": 0.0,
      "step": 5820
    },
    {
      "epoch": 0.92,
      "grad_norm": 4.868131327384617e-06,
      "learning_rate": 3.232412150447639e-06,
      "loss": 0.0,
      "step": 5830
    },
    {
      "epoch": 0.92,
      "grad_norm": 4.876651019003475e-06,
      "learning_rate": 3.1017106719649193e-06,
      "loss": 0.0,
      "step": 5840
    },
    {
      "epoch": 0.93,
      "grad_norm": 4.8540955503995065e-06,
      "learning_rate": 2.9736648237501218e-06,
      "loss": 0.0,
      "step": 5850
    },
    {
      "epoch": 0.93,
      "grad_norm": 4.787048965226859e-06,
      "learning_rate": 2.8482781150749872e-06,
      "loss": 0.0,
      "step": 5860
    },
    {
      "epoch": 0.93,
      "grad_norm": 4.8615252126182895e-06,
      "learning_rate": 2.7255539823339503e-06,
      "loss": 0.0,
      "step": 5870
    },
    {
      "epoch": 0.93,
      "grad_norm": 4.885783255303977e-06,
      "learning_rate": 2.605495788949841e-06,
      "loss": 0.0,
      "step": 5880
    },
    {
      "epoch": 0.93,
      "grad_norm": 4.888171588390833e-06,
      "learning_rate": 2.4881068252817994e-06,
      "loss": 0.0,
      "step": 5890
    },
    {
      "epoch": 0.93,
      "grad_norm": 4.836798325413838e-06,
      "learning_rate": 2.373390308535084e-06,
      "loss": 0.0,
      "step": 5900
    },
    {
      "epoch": 0.94,
      "grad_norm": 4.849552624364151e-06,
      "learning_rate": 2.261349382672895e-06,
      "loss": 0.0,
      "step": 5910
    },
    {
      "epoch": 0.94,
      "grad_norm": 4.768218786921352e-06,
      "learning_rate": 2.1519871183301675e-06,
      "loss": 0.0,
      "step": 5920
    },
    {
      "epoch": 0.94,
      "grad_norm": 4.835112576984102e-06,
      "learning_rate": 2.0453065127294614e-06,
      "loss": 0.0,
      "step": 5930
    },
    {
      "epoch": 0.94,
      "grad_norm": 4.815165539184818e-06,
      "learning_rate": 1.941310489598813e-06,
      "loss": 0.0,
      "step": 5940
    },
    {
      "epoch": 0.94,
      "grad_norm": 4.780455128639005e-06,
      "learning_rate": 1.8400018990916256e-06,
      "loss": 0.0,
      "step": 5950
    },
    {
      "epoch": 0.94,
      "grad_norm": 4.8534307097725105e-06,
      "learning_rate": 1.7413835177084835e-06,
      "loss": 0.0,
      "step": 5960
    },
    {
      "epoch": 0.94,
      "grad_norm": 4.769017323269509e-06,
      "learning_rate": 1.6454580482211379e-06,
      "loss": 0.0,
      "step": 5970
    },
    {
      "epoch": 0.95,
      "grad_norm": 4.866690233029658e-06,
      "learning_rate": 1.5522281195983978e-06,
      "loss": 0.0,
      "step": 5980
    },
    {
      "epoch": 0.95,
      "grad_norm": 4.864501534029841e-06,
      "learning_rate": 1.4616962869340889e-06,
      "loss": 0.0,
      "step": 5990
    },
    {
      "epoch": 0.95,
      "grad_norm": 4.847107902605785e-06,
      "learning_rate": 1.3738650313770196e-06,
      "loss": 0.0,
      "step": 6000
    },
    {
      "epoch": 0.95,
      "eval_loss": 6.170831738927518e-07,
      "eval_runtime": 676.4032,
      "eval_samples_per_second": 3.934,
      "eval_steps_per_second": 3.934,
      "step": 6000
    }
  ],
  "logging_steps": 10,
  "max_steps": 6317,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 200,
  "total_flos": 7.7202657902592e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}