{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.0,
  "eval_steps": 30,
  "global_step": 3128,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0012801024081926554,
      "grad_norm": 4.318057537078857,
      "learning_rate": 0.0,
      "loss": 1.0778,
      "step": 1
    },
    {
      "epoch": 0.002560204816385311,
      "grad_norm": 4.247881889343262,
      "learning_rate": 3.3898305084745766e-07,
      "loss": 1.1177,
      "step": 2
    },
    {
      "epoch": 0.0038403072245779663,
      "grad_norm": 4.42292594909668,
      "learning_rate": 6.779661016949153e-07,
      "loss": 1.0451,
      "step": 3
    },
    {
      "epoch": 0.005120409632770622,
      "grad_norm": 4.248280048370361,
      "learning_rate": 1.016949152542373e-06,
      "loss": 1.0984,
      "step": 4
    },
    {
      "epoch": 0.006400512040963277,
      "grad_norm": 3.9760775566101074,
      "learning_rate": 1.3559322033898307e-06,
      "loss": 1.1407,
      "step": 5
    },
    {
      "epoch": 0.007680614449155933,
      "grad_norm": 4.0757951736450195,
      "learning_rate": 1.6949152542372882e-06,
      "loss": 1.1223,
      "step": 6
    },
    {
      "epoch": 0.008960716857348588,
      "grad_norm": 3.6940793991088867,
      "learning_rate": 2.033898305084746e-06,
      "loss": 1.104,
      "step": 7
    },
    {
      "epoch": 0.010240819265541243,
      "grad_norm": 3.3989737033843994,
      "learning_rate": 2.372881355932204e-06,
      "loss": 1.0761,
      "step": 8
    },
    {
      "epoch": 0.011520921673733898,
      "grad_norm": 2.556053400039673,
      "learning_rate": 2.7118644067796613e-06,
      "loss": 1.0753,
      "step": 9
    },
    {
      "epoch": 0.012801024081926553,
      "grad_norm": 2.3599398136138916,
      "learning_rate": 3.0508474576271192e-06,
      "loss": 1.0595,
      "step": 10
    },
    {
      "epoch": 0.01408112649011921,
      "grad_norm": 2.174004316329956,
      "learning_rate": 3.3898305084745763e-06,
      "loss": 1.0986,
      "step": 11
    },
    {
      "epoch": 0.015361228898311865,
      "grad_norm": 4.076415061950684,
      "learning_rate": 3.7288135593220342e-06,
      "loss": 1.0546,
      "step": 12
    },
    {
      "epoch": 0.016641331306504522,
      "grad_norm": 3.8682055473327637,
      "learning_rate": 4.067796610169492e-06,
      "loss": 1.0498,
      "step": 13
    },
    {
      "epoch": 0.017921433714697177,
      "grad_norm": 3.5528361797332764,
      "learning_rate": 4.40677966101695e-06,
      "loss": 1.0399,
      "step": 14
    },
    {
      "epoch": 0.019201536122889832,
      "grad_norm": 2.4921932220458984,
      "learning_rate": 4.745762711864408e-06,
      "loss": 1.0343,
      "step": 15
    },
    {
      "epoch": 0.020481638531082487,
      "grad_norm": 2.5666632652282715,
      "learning_rate": 5.084745762711865e-06,
      "loss": 1.0366,
      "step": 16
    },
    {
      "epoch": 0.021761740939275142,
      "grad_norm": 2.5974960327148438,
      "learning_rate": 5.423728813559323e-06,
      "loss": 0.9794,
      "step": 17
    },
    {
      "epoch": 0.023041843347467797,
      "grad_norm": 2.314732789993286,
      "learning_rate": 5.7627118644067805e-06,
      "loss": 1.0135,
      "step": 18
    },
    {
      "epoch": 0.024321945755660452,
      "grad_norm": 1.8715646266937256,
      "learning_rate": 6.1016949152542385e-06,
      "loss": 1.0432,
      "step": 19
    },
    {
      "epoch": 0.025602048163853107,
      "grad_norm": 1.6317925453186035,
      "learning_rate": 6.440677966101695e-06,
      "loss": 1.0005,
      "step": 20
    },
    {
      "epoch": 0.026882150572045762,
      "grad_norm": 1.4418320655822754,
      "learning_rate": 6.779661016949153e-06,
      "loss": 1.0038,
      "step": 21
    },
    {
      "epoch": 0.02816225298023842,
      "grad_norm": 1.424547553062439,
      "learning_rate": 7.1186440677966106e-06,
      "loss": 1.0127,
      "step": 22
    },
    {
      "epoch": 0.029442355388431075,
      "grad_norm": 1.3697491884231567,
      "learning_rate": 7.4576271186440685e-06,
      "loss": 0.9887,
      "step": 23
    },
    {
      "epoch": 0.03072245779662373,
      "grad_norm": 1.2942132949829102,
      "learning_rate": 7.796610169491526e-06,
      "loss": 0.9546,
      "step": 24
    },
    {
      "epoch": 0.032002560204816385,
      "grad_norm": 1.2515391111373901,
      "learning_rate": 8.135593220338983e-06,
      "loss": 0.9461,
      "step": 25
    },
    {
      "epoch": 0.033282662613009044,
      "grad_norm": 1.2862663269042969,
      "learning_rate": 8.47457627118644e-06,
      "loss": 1.0312,
      "step": 26
    },
    {
      "epoch": 0.034562765021201695,
      "grad_norm": 1.198714256286621,
      "learning_rate": 8.8135593220339e-06,
      "loss": 1.0194,
      "step": 27
    },
    {
      "epoch": 0.035842867429394354,
      "grad_norm": 1.059179663658142,
      "learning_rate": 9.152542372881356e-06,
      "loss": 0.9944,
      "step": 28
    },
    {
      "epoch": 0.037122969837587005,
      "grad_norm": 0.9127014875411987,
      "learning_rate": 9.491525423728815e-06,
      "loss": 0.9496,
      "step": 29
    },
    {
      "epoch": 0.038403072245779664,
      "grad_norm": 0.9680947065353394,
      "learning_rate": 9.830508474576272e-06,
      "loss": 0.9746,
      "step": 30
    },
    {
      "epoch": 0.038403072245779664,
      "eval_loss": 0.015015463344752789,
      "eval_runtime": 10.8244,
      "eval_samples_per_second": 45.822,
      "eval_steps_per_second": 5.728,
      "step": 30
    },
    {
      "epoch": 0.039683174653972315,
      "grad_norm": 0.9460412263870239,
      "learning_rate": 1.016949152542373e-05,
      "loss": 0.9383,
      "step": 31
    },
    {
      "epoch": 0.040963277062164974,
      "grad_norm": 0.9197554588317871,
      "learning_rate": 1.0508474576271188e-05,
      "loss": 0.9388,
      "step": 32
    },
    {
      "epoch": 0.042243379470357625,
      "grad_norm": 0.8614938259124756,
      "learning_rate": 1.0847457627118645e-05,
      "loss": 0.9249,
      "step": 33
    },
    {
      "epoch": 0.043523481878550284,
      "grad_norm": 0.7609007954597473,
      "learning_rate": 1.1186440677966102e-05,
      "loss": 0.94,
      "step": 34
    },
    {
      "epoch": 0.04480358428674294,
      "grad_norm": 0.7822310924530029,
      "learning_rate": 1.1525423728813561e-05,
      "loss": 0.9407,
      "step": 35
    },
    {
      "epoch": 0.046083686694935594,
      "grad_norm": 0.7825785875320435,
      "learning_rate": 1.1864406779661018e-05,
      "loss": 0.9003,
      "step": 36
    },
    {
      "epoch": 0.04736378910312825,
      "grad_norm": 0.7560637593269348,
      "learning_rate": 1.2203389830508477e-05,
      "loss": 0.9513,
      "step": 37
    },
    {
      "epoch": 0.048643891511320904,
      "grad_norm": 0.698431134223938,
      "learning_rate": 1.2542372881355932e-05,
      "loss": 0.901,
      "step": 38
    },
    {
      "epoch": 0.04992399391951356,
      "grad_norm": 0.7895458936691284,
      "learning_rate": 1.288135593220339e-05,
      "loss": 0.8922,
      "step": 39
    },
    {
      "epoch": 0.051204096327706214,
      "grad_norm": 0.7163522243499756,
      "learning_rate": 1.3220338983050848e-05,
      "loss": 0.9065,
      "step": 40
    },
    {
      "epoch": 0.05248419873589887,
      "grad_norm": 0.670813798904419,
      "learning_rate": 1.3559322033898305e-05,
      "loss": 0.8912,
      "step": 41
    },
    {
      "epoch": 0.053764301144091524,
      "grad_norm": 0.7033433318138123,
      "learning_rate": 1.3898305084745764e-05,
      "loss": 0.9136,
      "step": 42
    },
    {
      "epoch": 0.05504440355228418,
      "grad_norm": 0.7405217289924622,
      "learning_rate": 1.4237288135593221e-05,
      "loss": 0.9028,
      "step": 43
    },
    {
      "epoch": 0.05632450596047684,
      "grad_norm": 0.6802017092704773,
      "learning_rate": 1.4576271186440678e-05,
      "loss": 0.8928,
      "step": 44
    },
    {
      "epoch": 0.05760460836866949,
      "grad_norm": 0.6827060580253601,
      "learning_rate": 1.4915254237288137e-05,
      "loss": 0.9427,
      "step": 45
    },
    {
      "epoch": 0.05888471077686215,
      "grad_norm": 0.7236905694007874,
      "learning_rate": 1.5254237288135594e-05,
      "loss": 0.9082,
      "step": 46
    },
    {
      "epoch": 0.0601648131850548,
      "grad_norm": 0.7171168923377991,
      "learning_rate": 1.5593220338983053e-05,
      "loss": 0.9184,
      "step": 47
    },
    {
      "epoch": 0.06144491559324746,
      "grad_norm": 0.6564812660217285,
      "learning_rate": 1.593220338983051e-05,
      "loss": 0.856,
      "step": 48
    },
    {
      "epoch": 0.06272501800144012,
      "grad_norm": 0.7057896852493286,
      "learning_rate": 1.6271186440677967e-05,
      "loss": 0.8904,
      "step": 49
    },
    {
      "epoch": 0.06400512040963277,
      "grad_norm": 0.7270225882530212,
      "learning_rate": 1.6610169491525424e-05,
      "loss": 0.8885,
      "step": 50
    },
    {
      "epoch": 0.06528522281782542,
      "grad_norm": 0.7114495038986206,
      "learning_rate": 1.694915254237288e-05,
      "loss": 0.8972,
      "step": 51
    },
    {
      "epoch": 0.06656532522601809,
      "grad_norm": 0.7182011008262634,
      "learning_rate": 1.728813559322034e-05,
      "loss": 0.9454,
      "step": 52
    },
    {
      "epoch": 0.06784542763421074,
      "grad_norm": 0.7269001007080078,
      "learning_rate": 1.76271186440678e-05,
      "loss": 0.9026,
      "step": 53
    },
    {
      "epoch": 0.06912553004240339,
      "grad_norm": 0.695046603679657,
      "learning_rate": 1.7966101694915256e-05,
      "loss": 0.8912,
      "step": 54
    },
    {
      "epoch": 0.07040563245059604,
      "grad_norm": 0.7239252924919128,
      "learning_rate": 1.8305084745762713e-05,
      "loss": 0.9019,
      "step": 55
    },
    {
      "epoch": 0.07168573485878871,
      "grad_norm": 0.6834789514541626,
      "learning_rate": 1.864406779661017e-05,
      "loss": 0.8904,
      "step": 56
    },
    {
      "epoch": 0.07296583726698136,
      "grad_norm": 0.7371823191642761,
      "learning_rate": 1.898305084745763e-05,
      "loss": 0.9121,
      "step": 57
    },
    {
      "epoch": 0.07424593967517401,
      "grad_norm": 0.6945562362670898,
      "learning_rate": 1.9322033898305087e-05,
      "loss": 0.8754,
      "step": 58
    },
    {
      "epoch": 0.07552604208336668,
      "grad_norm": 0.717341423034668,
      "learning_rate": 1.9661016949152545e-05,
      "loss": 0.8548,
      "step": 59
    },
    {
      "epoch": 0.07680614449155933,
      "grad_norm": 0.7238847017288208,
      "learning_rate": 2e-05,
      "loss": 0.8478,
      "step": 60
    },
    {
      "epoch": 0.07680614449155933,
      "eval_loss": 0.013981991447508335,
      "eval_runtime": 10.8389,
      "eval_samples_per_second": 45.761,
      "eval_steps_per_second": 5.72,
      "step": 60
    },
    {
      "epoch": 0.07808624689975198,
      "grad_norm": 0.6857765913009644,
      "learning_rate": 2.033898305084746e-05,
      "loss": 0.9078,
      "step": 61
    },
    {
      "epoch": 0.07936634930794463,
      "grad_norm": 0.7827364802360535,
      "learning_rate": 2.0677966101694916e-05,
      "loss": 0.8982,
      "step": 62
    },
    {
      "epoch": 0.0806464517161373,
      "grad_norm": 0.7178738713264465,
      "learning_rate": 2.1016949152542376e-05,
      "loss": 0.882,
      "step": 63
    },
    {
      "epoch": 0.08192655412432995,
      "grad_norm": 0.7412232160568237,
      "learning_rate": 2.1355932203389833e-05,
      "loss": 0.9023,
      "step": 64
    },
    {
      "epoch": 0.0832066565325226,
      "grad_norm": 0.7248542904853821,
      "learning_rate": 2.169491525423729e-05,
      "loss": 0.8355,
      "step": 65
    },
    {
      "epoch": 0.08448675894071525,
      "grad_norm": 0.6934853196144104,
      "learning_rate": 2.2033898305084748e-05,
      "loss": 0.8826,
      "step": 66
    },
    {
      "epoch": 0.08576686134890792,
      "grad_norm": 0.7710501551628113,
      "learning_rate": 2.2372881355932205e-05,
      "loss": 0.8677,
      "step": 67
    },
    {
      "epoch": 0.08704696375710057,
      "grad_norm": 0.720954418182373,
      "learning_rate": 2.2711864406779665e-05,
      "loss": 0.8817,
      "step": 68
    },
    {
      "epoch": 0.08832706616529322,
      "grad_norm": 0.792184591293335,
      "learning_rate": 2.3050847457627122e-05,
      "loss": 0.8755,
      "step": 69
    },
    {
      "epoch": 0.08960716857348588,
      "grad_norm": 0.8244057297706604,
      "learning_rate": 2.338983050847458e-05,
      "loss": 0.8696,
      "step": 70
    },
    {
      "epoch": 0.09088727098167854,
      "grad_norm": 0.7293037176132202,
      "learning_rate": 2.3728813559322036e-05,
      "loss": 0.8693,
      "step": 71
    },
    {
      "epoch": 0.09216737338987119,
      "grad_norm": 0.8354009985923767,
      "learning_rate": 2.406779661016949e-05,
      "loss": 0.8964,
      "step": 72
    },
    {
      "epoch": 0.09344747579806384,
      "grad_norm": 0.7932097911834717,
      "learning_rate": 2.4406779661016954e-05,
      "loss": 0.8834,
      "step": 73
    },
    {
      "epoch": 0.0947275782062565,
      "grad_norm": 0.7310692071914673,
      "learning_rate": 2.474576271186441e-05,
      "loss": 0.8576,
      "step": 74
    },
    {
      "epoch": 0.09600768061444916,
      "grad_norm": 0.7643473148345947,
      "learning_rate": 2.5084745762711865e-05,
      "loss": 0.8414,
      "step": 75
    },
    {
      "epoch": 0.09728778302264181,
      "grad_norm": 0.7004395127296448,
      "learning_rate": 2.5423728813559322e-05,
      "loss": 0.8518,
      "step": 76
    },
    {
      "epoch": 0.09856788543083447,
      "grad_norm": 0.7725978493690491,
      "learning_rate": 2.576271186440678e-05,
      "loss": 0.8464,
      "step": 77
    },
    {
      "epoch": 0.09984798783902712,
      "grad_norm": 0.7045527696609497,
      "learning_rate": 2.610169491525424e-05,
      "loss": 0.8779,
      "step": 78
    },
    {
      "epoch": 0.10112809024721978,
      "grad_norm": 0.7486863136291504,
      "learning_rate": 2.6440677966101696e-05,
      "loss": 0.8712,
      "step": 79
    },
    {
      "epoch": 0.10240819265541243,
      "grad_norm": 0.7594077587127686,
      "learning_rate": 2.6779661016949153e-05,
      "loss": 0.8862,
      "step": 80
    },
    {
      "epoch": 0.1036882950636051,
      "grad_norm": 0.7708922028541565,
      "learning_rate": 2.711864406779661e-05,
      "loss": 0.8716,
      "step": 81
    },
    {
      "epoch": 0.10496839747179774,
      "grad_norm": 0.8713701367378235,
      "learning_rate": 2.7457627118644068e-05,
      "loss": 0.8824,
      "step": 82
    },
    {
      "epoch": 0.1062484998799904,
      "grad_norm": 0.7621445059776306,
      "learning_rate": 2.7796610169491528e-05,
      "loss": 0.8523,
      "step": 83
    },
    {
      "epoch": 0.10752860228818305,
      "grad_norm": 0.8088158965110779,
      "learning_rate": 2.8135593220338985e-05,
      "loss": 0.8374,
      "step": 84
    },
    {
      "epoch": 0.10880870469637571,
      "grad_norm": 0.7840273380279541,
      "learning_rate": 2.8474576271186442e-05,
      "loss": 0.9006,
      "step": 85
    },
    {
      "epoch": 0.11008880710456836,
      "grad_norm": 0.8660479187965393,
      "learning_rate": 2.88135593220339e-05,
      "loss": 0.8457,
      "step": 86
    },
    {
      "epoch": 0.11136890951276102,
      "grad_norm": 0.8039504289627075,
      "learning_rate": 2.9152542372881356e-05,
      "loss": 0.8649,
      "step": 87
    },
    {
      "epoch": 0.11264901192095368,
      "grad_norm": 0.97013258934021,
      "learning_rate": 2.9491525423728817e-05,
      "loss": 0.9029,
      "step": 88
    },
    {
      "epoch": 0.11392911432914633,
      "grad_norm": 0.8036187291145325,
      "learning_rate": 2.9830508474576274e-05,
      "loss": 0.8297,
      "step": 89
    },
    {
      "epoch": 0.11520921673733898,
      "grad_norm": 0.8849989175796509,
      "learning_rate": 3.016949152542373e-05,
      "loss": 0.8567,
      "step": 90
    },
    {
      "epoch": 0.11520921673733898,
      "eval_loss": 0.013601833023130894,
      "eval_runtime": 10.8134,
      "eval_samples_per_second": 45.869,
      "eval_steps_per_second": 5.734,
      "step": 90
    },
    {
      "epoch": 0.11648931914553164,
      "grad_norm": 0.8388200402259827,
      "learning_rate": 3.0508474576271188e-05,
      "loss": 0.8421,
      "step": 91
    },
    {
      "epoch": 0.1177694215537243,
      "grad_norm": 1.01188063621521,
      "learning_rate": 3.084745762711865e-05,
      "loss": 0.8578,
      "step": 92
    },
    {
      "epoch": 0.11904952396191695,
      "grad_norm": 0.7491028308868408,
      "learning_rate": 3.1186440677966106e-05,
      "loss": 0.8573,
      "step": 93
    },
    {
      "epoch": 0.1203296263701096,
      "grad_norm": 1.0739816427230835,
      "learning_rate": 3.152542372881356e-05,
      "loss": 0.9224,
      "step": 94
    },
    {
      "epoch": 0.12160972877830227,
      "grad_norm": 0.9446407556533813,
      "learning_rate": 3.186440677966102e-05,
      "loss": 0.8538,
      "step": 95
    },
    {
      "epoch": 0.12288983118649492,
      "grad_norm": 0.9192800521850586,
      "learning_rate": 3.2203389830508473e-05,
      "loss": 0.853,
      "step": 96
    },
    {
      "epoch": 0.12416993359468757,
      "grad_norm": 1.107749342918396,
      "learning_rate": 3.2542372881355934e-05,
      "loss": 0.8723,
      "step": 97
    },
    {
      "epoch": 0.12545003600288024,
      "grad_norm": 0.9199897050857544,
      "learning_rate": 3.2881355932203394e-05,
      "loss": 0.8362,
      "step": 98
    },
    {
      "epoch": 0.1267301384110729,
      "grad_norm": 1.0593056678771973,
      "learning_rate": 3.322033898305085e-05,
      "loss": 0.8373,
      "step": 99
    },
    {
      "epoch": 0.12801024081926554,
      "grad_norm": 0.975534975528717,
      "learning_rate": 3.355932203389831e-05,
      "loss": 0.8604,
      "step": 100
    },
    {
      "epoch": 0.1292903432274582,
      "grad_norm": 1.0486565828323364,
      "learning_rate": 3.389830508474576e-05,
      "loss": 0.8732,
      "step": 101
    },
    {
      "epoch": 0.13057044563565084,
      "grad_norm": 1.0505051612854004,
      "learning_rate": 3.423728813559322e-05,
      "loss": 0.8588,
      "step": 102
    },
    {
      "epoch": 0.1318505480438435,
      "grad_norm": 0.8159765005111694,
      "learning_rate": 3.457627118644068e-05,
      "loss": 0.8383,
      "step": 103
    },
    {
      "epoch": 0.13313065045203618,
      "grad_norm": 0.9945966601371765,
      "learning_rate": 3.491525423728814e-05,
      "loss": 0.8391,
      "step": 104
    },
    {
      "epoch": 0.13441075286022883,
      "grad_norm": 0.9792317152023315,
      "learning_rate": 3.52542372881356e-05,
      "loss": 0.9045,
      "step": 105
    },
    {
      "epoch": 0.13569085526842148,
      "grad_norm": 0.9425814747810364,
      "learning_rate": 3.559322033898305e-05,
      "loss": 0.8713,
      "step": 106
    },
    {
      "epoch": 0.13697095767661413,
      "grad_norm": 1.0463520288467407,
      "learning_rate": 3.593220338983051e-05,
      "loss": 0.8833,
      "step": 107
    },
    {
      "epoch": 0.13825106008480678,
      "grad_norm": 0.9036206007003784,
      "learning_rate": 3.627118644067797e-05,
      "loss": 0.8142,
      "step": 108
    },
    {
      "epoch": 0.13953116249299943,
      "grad_norm": 0.8945938944816589,
      "learning_rate": 3.6610169491525426e-05,
      "loss": 0.8189,
      "step": 109
    },
    {
      "epoch": 0.14081126490119208,
      "grad_norm": 1.0331628322601318,
      "learning_rate": 3.6949152542372886e-05,
      "loss": 0.8251,
      "step": 110
    },
    {
      "epoch": 0.14209136730938476,
      "grad_norm": 1.019895076751709,
      "learning_rate": 3.728813559322034e-05,
      "loss": 0.8176,
      "step": 111
    },
    {
      "epoch": 0.14337146971757742,
      "grad_norm": 0.9001871347427368,
      "learning_rate": 3.76271186440678e-05,
      "loss": 0.842,
      "step": 112
    },
    {
      "epoch": 0.14465157212577007,
      "grad_norm": 0.842754065990448,
      "learning_rate": 3.796610169491526e-05,
      "loss": 0.8192,
      "step": 113
    },
    {
      "epoch": 0.14593167453396272,
      "grad_norm": 0.998187243938446,
      "learning_rate": 3.8305084745762714e-05,
      "loss": 0.9035,
      "step": 114
    },
    {
      "epoch": 0.14721177694215537,
      "grad_norm": 0.8333083987236023,
      "learning_rate": 3.8644067796610175e-05,
      "loss": 0.7836,
      "step": 115
    },
    {
      "epoch": 0.14849187935034802,
      "grad_norm": 1.0127038955688477,
      "learning_rate": 3.898305084745763e-05,
      "loss": 0.811,
      "step": 116
    },
    {
      "epoch": 0.14977198175854067,
      "grad_norm": 0.7711405754089355,
      "learning_rate": 3.932203389830509e-05,
      "loss": 0.814,
      "step": 117
    },
    {
      "epoch": 0.15105208416673335,
      "grad_norm": 0.9136852622032166,
      "learning_rate": 3.966101694915255e-05,
      "loss": 0.8363,
      "step": 118
    },
    {
      "epoch": 0.152332186574926,
      "grad_norm": 0.8900148868560791,
      "learning_rate": 4e-05,
      "loss": 0.8484,
      "step": 119
    },
    {
      "epoch": 0.15361228898311866,
      "grad_norm": 0.8631476163864136,
      "learning_rate": 3.999999382260215e-05,
      "loss": 0.8252,
      "step": 120
    },
    {
      "epoch": 0.15361228898311866,
      "eval_loss": 0.013392439112067223,
      "eval_runtime": 10.8197,
      "eval_samples_per_second": 45.842,
      "eval_steps_per_second": 5.73,
      "step": 120
    },
    {
      "epoch": 0.1548923913913113,
      "grad_norm": 0.8755859136581421,
      "learning_rate": 3.9999975290412844e-05,
      "loss": 0.8276,
      "step": 121
    },
    {
      "epoch": 0.15617249379950396,
      "grad_norm": 0.8674336671829224,
      "learning_rate": 3.9999944403444785e-05,
      "loss": 0.8313,
      "step": 122
    },
    {
      "epoch": 0.1574525962076966,
      "grad_norm": 0.7986403107643127,
      "learning_rate": 3.999990116171919e-05,
      "loss": 0.8785,
      "step": 123
    },
    {
      "epoch": 0.15873269861588926,
      "grad_norm": 0.8159152865409851,
      "learning_rate": 3.999984556526574e-05,
      "loss": 0.8822,
      "step": 124
    },
    {
      "epoch": 0.1600128010240819,
      "grad_norm": 0.8714810013771057,
      "learning_rate": 3.9999777614122576e-05,
      "loss": 0.8251,
      "step": 125
    },
    {
      "epoch": 0.1612929034322746,
      "grad_norm": 0.7786198854446411,
      "learning_rate": 3.9999697308336365e-05,
      "loss": 0.8785,
      "step": 126
    },
    {
      "epoch": 0.16257300584046724,
      "grad_norm": 1.0131577253341675,
      "learning_rate": 3.99996046479622e-05,
      "loss": 0.8619,
      "step": 127
    },
    {
      "epoch": 0.1638531082486599,
      "grad_norm": 0.8323366641998291,
      "learning_rate": 3.99994996330637e-05,
      "loss": 0.8224,
      "step": 128
    },
    {
      "epoch": 0.16513321065685255,
      "grad_norm": 0.9837428331375122,
      "learning_rate": 3.999938226371294e-05,
      "loss": 0.831,
      "step": 129
    },
    {
      "epoch": 0.1664133130650452,
      "grad_norm": 0.9995911717414856,
      "learning_rate": 3.999925253999048e-05,
      "loss": 0.865,
      "step": 130
    },
    {
      "epoch": 0.16769341547323785,
      "grad_norm": 0.8621081113815308,
      "learning_rate": 3.999911046198536e-05,
      "loss": 0.8698,
      "step": 131
    },
    {
      "epoch": 0.1689735178814305,
      "grad_norm": 1.2333866357803345,
      "learning_rate": 3.9998956029795096e-05,
      "loss": 0.8409,
      "step": 132
    },
    {
      "epoch": 0.17025362028962318,
      "grad_norm": 1.014843225479126,
      "learning_rate": 3.999878924352568e-05,
      "loss": 0.891,
      "step": 133
    },
    {
      "epoch": 0.17153372269781583,
      "grad_norm": 1.112094521522522,
      "learning_rate": 3.99986101032916e-05,
      "loss": 0.8653,
      "step": 134
    },
    {
      "epoch": 0.17281382510600848,
      "grad_norm": 0.9054744839668274,
      "learning_rate": 3.9998418609215826e-05,
      "loss": 0.7658,
      "step": 135
    },
    {
      "epoch": 0.17409392751420114,
      "grad_norm": 0.9942414164543152,
      "learning_rate": 3.999821476142977e-05,
      "loss": 0.8341,
      "step": 136
    },
    {
      "epoch": 0.1753740299223938,
      "grad_norm": 0.9464170336723328,
      "learning_rate": 3.9997998560073364e-05,
      "loss": 0.798,
      "step": 137
    },
    {
      "epoch": 0.17665413233058644,
      "grad_norm": 1.042913556098938,
      "learning_rate": 3.9997770005294996e-05,
      "loss": 0.8067,
      "step": 138
    },
    {
      "epoch": 0.1779342347387791,
      "grad_norm": 0.8861857652664185,
      "learning_rate": 3.999752909725154e-05,
      "loss": 0.8492,
      "step": 139
    },
    {
      "epoch": 0.17921433714697177,
      "grad_norm": 0.9364302158355713,
      "learning_rate": 3.9997275836108366e-05,
      "loss": 0.8356,
      "step": 140
    },
    {
      "epoch": 0.18049443955516442,
      "grad_norm": 0.901847779750824,
      "learning_rate": 3.9997010222039297e-05,
      "loss": 0.8278,
      "step": 141
    },
    {
      "epoch": 0.18177454196335707,
      "grad_norm": 1.061315655708313,
      "learning_rate": 3.999673225522664e-05,
      "loss": 0.8284,
      "step": 142
    },
    {
      "epoch": 0.18305464437154972,
      "grad_norm": 0.8948325514793396,
      "learning_rate": 3.999644193586118e-05,
      "loss": 0.8294,
      "step": 143
    },
    {
      "epoch": 0.18433474677974238,
      "grad_norm": 1.0115327835083008,
      "learning_rate": 3.9996139264142205e-05,
      "loss": 0.8784,
      "step": 144
    },
    {
      "epoch": 0.18561484918793503,
      "grad_norm": 0.9907302856445312,
      "learning_rate": 3.999582424027744e-05,
      "loss": 0.8236,
      "step": 145
    },
    {
      "epoch": 0.18689495159612768,
      "grad_norm": 0.9442008137702942,
      "learning_rate": 3.999549686448313e-05,
      "loss": 0.8438,
      "step": 146
    },
    {
      "epoch": 0.18817505400432036,
      "grad_norm": 0.9412844777107239,
      "learning_rate": 3.9995157136983966e-05,
      "loss": 0.8731,
      "step": 147
    },
    {
      "epoch": 0.189455156412513,
      "grad_norm": 0.8231805562973022,
      "learning_rate": 3.999480505801313e-05,
      "loss": 0.8479,
      "step": 148
    },
    {
      "epoch": 0.19073525882070566,
      "grad_norm": 1.113372802734375,
      "learning_rate": 3.999444062781229e-05,
      "loss": 0.8655,
      "step": 149
    },
    {
      "epoch": 0.1920153612288983,
      "grad_norm": 0.7836041450500488,
      "learning_rate": 3.9994063846631565e-05,
      "loss": 0.8351,
      "step": 150
    },
    {
      "epoch": 0.1920153612288983,
      "eval_loss": 0.013233460485935211,
      "eval_runtime": 10.8102,
      "eval_samples_per_second": 45.883,
      "eval_steps_per_second": 5.735,
      "step": 150
    },
    {
      "epoch": 0.19329546363709096,
      "grad_norm": 0.9617286920547485,
      "learning_rate": 3.99936747147296e-05,
      "loss": 0.79,
      "step": 151
    },
    {
      "epoch": 0.19457556604528362,
      "grad_norm": 0.8357570767402649,
      "learning_rate": 3.9993273232373455e-05,
      "loss": 0.8284,
      "step": 152
    },
    {
      "epoch": 0.19585566845347627,
      "grad_norm": 1.0674220323562622,
      "learning_rate": 3.999285939983871e-05,
      "loss": 0.8479,
      "step": 153
    },
    {
      "epoch": 0.19713577086166895,
      "grad_norm": 0.9409031271934509,
      "learning_rate": 3.9992433217409414e-05,
      "loss": 0.8149,
      "step": 154
    },
    {
      "epoch": 0.1984158732698616,
      "grad_norm": 1.0493512153625488,
      "learning_rate": 3.999199468537809e-05,
      "loss": 0.809,
      "step": 155
    },
    {
      "epoch": 0.19969597567805425,
      "grad_norm": 1.1253619194030762,
      "learning_rate": 3.999154380404572e-05,
      "loss": 0.8131,
      "step": 156
    },
    {
      "epoch": 0.2009760780862469,
      "grad_norm": 0.8770150542259216,
      "learning_rate": 3.9991080573721805e-05,
      "loss": 0.7923,
      "step": 157
    },
    {
      "epoch": 0.20225618049443955,
      "grad_norm": 1.2235959768295288,
      "learning_rate": 3.9990604994724276e-05,
      "loss": 0.8317,
      "step": 158
    },
    {
      "epoch": 0.2035362829026322,
      "grad_norm": 0.9696475267410278,
      "learning_rate": 3.999011706737957e-05,
      "loss": 0.8265,
      "step": 159
    },
    {
      "epoch": 0.20481638531082486,
      "grad_norm": 1.2137672901153564,
      "learning_rate": 3.998961679202258e-05,
      "loss": 0.8498,
      "step": 160
    },
    {
      "epoch": 0.20609648771901753,
      "grad_norm": 0.9181241393089294,
      "learning_rate": 3.998910416899669e-05,
      "loss": 0.8263,
      "step": 161
    },
    {
      "epoch": 0.2073765901272102,
      "grad_norm": 1.0590282678604126,
      "learning_rate": 3.998857919865376e-05,
      "loss": 0.8282,
      "step": 162
    },
    {
      "epoch": 0.20865669253540284,
      "grad_norm": 0.951016366481781,
      "learning_rate": 3.9988041881354095e-05,
      "loss": 0.8476,
      "step": 163
    },
    {
      "epoch": 0.2099367949435955,
      "grad_norm": 0.8958009481430054,
      "learning_rate": 3.998749221746652e-05,
      "loss": 0.8129,
      "step": 164
    },
    {
      "epoch": 0.21121689735178814,
      "grad_norm": 1.042737603187561,
      "learning_rate": 3.9986930207368306e-05,
      "loss": 0.8522,
      "step": 165
    },
    {
      "epoch": 0.2124969997599808,
      "grad_norm": 0.9709591865539551,
      "learning_rate": 3.998635585144521e-05,
      "loss": 0.8247,
      "step": 166
    },
    {
      "epoch": 0.21377710216817344,
      "grad_norm": 0.9807443618774414,
      "learning_rate": 3.9985769150091434e-05,
      "loss": 0.8134,
      "step": 167
    },
    {
      "epoch": 0.2150572045763661,
      "grad_norm": 0.9329289197921753,
      "learning_rate": 3.99851701037097e-05,
      "loss": 0.8209,
      "step": 168
    },
    {
      "epoch": 0.21633730698455877,
      "grad_norm": 0.9565260410308838,
      "learning_rate": 3.998455871271117e-05,
      "loss": 0.8406,
      "step": 169
    },
    {
      "epoch": 0.21761740939275143,
      "grad_norm": 0.9333802461624146,
      "learning_rate": 3.99839349775155e-05,
      "loss": 0.794,
      "step": 170
    },
    {
      "epoch": 0.21889751180094408,
      "grad_norm": 0.8861300945281982,
      "learning_rate": 3.9983298898550794e-05,
      "loss": 0.8694,
      "step": 171
    },
    {
      "epoch": 0.22017761420913673,
      "grad_norm": 0.8293015360832214,
      "learning_rate": 3.998265047625364e-05,
      "loss": 0.8338,
      "step": 172
    },
    {
      "epoch": 0.22145771661732938,
      "grad_norm": 0.8500891327857971,
      "learning_rate": 3.998198971106912e-05,
      "loss": 0.8035,
      "step": 173
    },
    {
      "epoch": 0.22273781902552203,
      "grad_norm": 0.800555408000946,
      "learning_rate": 3.998131660345075e-05,
      "loss": 0.8699,
      "step": 174
    },
    {
      "epoch": 0.22401792143371468,
      "grad_norm": 0.8893824815750122,
      "learning_rate": 3.998063115386054e-05,
      "loss": 0.7829,
      "step": 175
    },
    {
      "epoch": 0.22529802384190736,
      "grad_norm": 0.7633495330810547,
      "learning_rate": 3.997993336276897e-05,
      "loss": 0.8272,
      "step": 176
    },
    {
      "epoch": 0.22657812625010001,
      "grad_norm": 0.8218979835510254,
      "learning_rate": 3.9979223230654994e-05,
      "loss": 0.7728,
      "step": 177
    },
    {
      "epoch": 0.22785822865829267,
      "grad_norm": 0.7941882014274597,
      "learning_rate": 3.997850075800602e-05,
      "loss": 0.7987,
      "step": 178
    },
    {
      "epoch": 0.22913833106648532,
      "grad_norm": 0.8235889077186584,
      "learning_rate": 3.9977765945317934e-05,
      "loss": 0.8311,
      "step": 179
    },
    {
      "epoch": 0.23041843347467797,
      "grad_norm": 0.8269426226615906,
      "learning_rate": 3.997701879309511e-05,
      "loss": 0.8177,
      "step": 180
    },
    {
      "epoch": 0.23041843347467797,
      "eval_loss": 0.013103107921779156,
      "eval_runtime": 10.8218,
      "eval_samples_per_second": 45.833,
      "eval_steps_per_second": 5.729,
      "step": 180
    },
    {
      "epoch": 0.23169853588287062,
      "grad_norm": 0.9420286417007446,
      "learning_rate": 3.9976259301850375e-05,
      "loss": 0.8359,
      "step": 181
    },
    {
      "epoch": 0.23297863829106327,
      "grad_norm": 0.9078615307807922,
      "learning_rate": 3.997548747210501e-05,
      "loss": 0.7836,
      "step": 182
    },
    {
      "epoch": 0.23425874069925595,
      "grad_norm": 0.9137442111968994,
      "learning_rate": 3.997470330438879e-05,
      "loss": 0.8228,
      "step": 183
    },
    {
      "epoch": 0.2355388431074486,
      "grad_norm": 0.9286066889762878,
      "learning_rate": 3.9973906799239964e-05,
      "loss": 0.8077,
      "step": 184
    },
    {
      "epoch": 0.23681894551564125,
      "grad_norm": 0.922153890132904,
      "learning_rate": 3.997309795720521e-05,
      "loss": 0.8655,
      "step": 185
    },
    {
      "epoch": 0.2380990479238339,
      "grad_norm": 0.8637731075286865,
      "learning_rate": 3.9972276778839725e-05,
      "loss": 0.8416,
      "step": 186
    },
    {
      "epoch": 0.23937915033202656,
      "grad_norm": 0.8832123875617981,
      "learning_rate": 3.997144326470713e-05,
      "loss": 0.8376,
      "step": 187
    },
    {
      "epoch": 0.2406592527402192,
      "grad_norm": 0.9608950614929199,
      "learning_rate": 3.9970597415379535e-05,
      "loss": 0.8298,
      "step": 188
    },
    {
      "epoch": 0.24193935514841186,
      "grad_norm": 0.8668861389160156,
      "learning_rate": 3.9969739231437514e-05,
      "loss": 0.8076,
      "step": 189
    },
    {
      "epoch": 0.24321945755660454,
      "grad_norm": 0.7897141575813293,
      "learning_rate": 3.99688687134701e-05,
      "loss": 0.8099,
      "step": 190
    },
    {
      "epoch": 0.2444995599647972,
      "grad_norm": 0.8441980481147766,
      "learning_rate": 3.9967985862074795e-05,
      "loss": 0.8078,
      "step": 191
    },
    {
      "epoch": 0.24577966237298984,
      "grad_norm": 0.8205133676528931,
      "learning_rate": 3.996709067785758e-05,
      "loss": 0.8296,
      "step": 192
    },
    {
      "epoch": 0.2470597647811825,
      "grad_norm": 0.9071422815322876,
      "learning_rate": 3.996618316143288e-05,
      "loss": 0.8355,
      "step": 193
    },
    {
      "epoch": 0.24833986718937515,
      "grad_norm": 0.7740186452865601,
      "learning_rate": 3.996526331342361e-05,
      "loss": 0.8331,
      "step": 194
    },
    {
      "epoch": 0.2496199695975678,
      "grad_norm": 0.97866290807724,
      "learning_rate": 3.9964331134461104e-05,
      "loss": 0.8096,
      "step": 195
    },
    {
      "epoch": 0.2509000720057605,
      "grad_norm": 0.6750162839889526,
      "learning_rate": 3.996338662518521e-05,
      "loss": 0.797,
      "step": 196
    },
    {
      "epoch": 0.25218017441395313,
      "grad_norm": 0.8180609345436096,
      "learning_rate": 3.996242978624421e-05,
      "loss": 0.8085,
      "step": 197
    },
    {
      "epoch": 0.2534602768221458,
      "grad_norm": 0.8507323861122131,
      "learning_rate": 3.996146061829487e-05,
      "loss": 0.8164,
      "step": 198
    },
    {
      "epoch": 0.25474037923033843,
      "grad_norm": 0.7099197506904602,
      "learning_rate": 3.9960479122002384e-05,
      "loss": 0.8216,
      "step": 199
    },
    {
      "epoch": 0.2560204816385311,
      "grad_norm": 0.9042320847511292,
      "learning_rate": 3.9959485298040436e-05,
      "loss": 0.807,
      "step": 200
    },
    {
      "epoch": 0.25730058404672373,
      "grad_norm": 0.886713445186615,
      "learning_rate": 3.995847914709118e-05,
      "loss": 0.8262,
      "step": 201
    },
    {
      "epoch": 0.2585806864549164,
      "grad_norm": 0.7223711609840393,
      "learning_rate": 3.9957460669845204e-05,
      "loss": 0.7771,
      "step": 202
    },
    {
      "epoch": 0.25986078886310904,
      "grad_norm": 0.7963854074478149,
      "learning_rate": 3.9956429867001554e-05,
      "loss": 0.8313,
      "step": 203
    },
    {
      "epoch": 0.2611408912713017,
      "grad_norm": 0.7575312852859497,
      "learning_rate": 3.9955386739267766e-05,
      "loss": 0.7841,
      "step": 204
    },
    {
      "epoch": 0.26242099367949434,
      "grad_norm": 0.8132676482200623,
      "learning_rate": 3.9954331287359827e-05,
      "loss": 0.8317,
      "step": 205
    },
    {
      "epoch": 0.263701096087687,
      "grad_norm": 0.7840173840522766,
      "learning_rate": 3.995326351200215e-05,
      "loss": 0.8143,
      "step": 206
    },
    {
      "epoch": 0.26498119849587964,
      "grad_norm": 0.740490198135376,
      "learning_rate": 3.995218341392766e-05,
      "loss": 0.8268,
      "step": 207
    },
    {
      "epoch": 0.26626130090407235,
      "grad_norm": 0.8167890906333923,
      "learning_rate": 3.995109099387769e-05,
      "loss": 0.8043,
      "step": 208
    },
    {
      "epoch": 0.267541403312265,
      "grad_norm": 0.7455295324325562,
      "learning_rate": 3.9949986252602074e-05,
      "loss": 0.8113,
      "step": 209
    },
    {
      "epoch": 0.26882150572045765,
      "grad_norm": 0.8637762069702148,
      "learning_rate": 3.994886919085906e-05,
      "loss": 0.8118,
      "step": 210
    },
    {
      "epoch": 0.26882150572045765,
      "eval_loss": 0.012998094782233238,
      "eval_runtime": 10.8341,
      "eval_samples_per_second": 45.781,
      "eval_steps_per_second": 5.723,
      "step": 210
    },
    {
      "epoch": 0.2701016081286503,
      "grad_norm": 0.7905036211013794,
      "learning_rate": 3.9947739809415384e-05,
      "loss": 0.8571,
      "step": 211
    },
    {
      "epoch": 0.27138171053684296,
      "grad_norm": 0.7857114672660828,
      "learning_rate": 3.994659810904623e-05,
      "loss": 0.8029,
      "step": 212
    },
    {
      "epoch": 0.2726618129450356,
      "grad_norm": 0.9537340998649597,
      "learning_rate": 3.994544409053523e-05,
      "loss": 0.8214,
      "step": 213
    },
    {
      "epoch": 0.27394191535322826,
      "grad_norm": 0.8235071301460266,
      "learning_rate": 3.994427775467448e-05,
      "loss": 0.8492,
      "step": 214
    },
    {
      "epoch": 0.2752220177614209,
      "grad_norm": 0.8366528153419495,
      "learning_rate": 3.994309910226453e-05,
      "loss": 0.833,
      "step": 215
    },
    {
      "epoch": 0.27650212016961356,
      "grad_norm": 0.8549700975418091,
      "learning_rate": 3.994190813411437e-05,
      "loss": 0.8229,
      "step": 216
    },
    {
      "epoch": 0.2777822225778062,
      "grad_norm": 0.8313760757446289,
      "learning_rate": 3.9940704851041466e-05,
      "loss": 0.8408,
      "step": 217
    },
    {
      "epoch": 0.27906232498599887,
      "grad_norm": 0.7870802283287048,
      "learning_rate": 3.9939489253871715e-05,
      "loss": 0.8467,
      "step": 218
    },
    {
      "epoch": 0.2803424273941915,
      "grad_norm": 0.8972495794296265,
      "learning_rate": 3.993826134343949e-05,
      "loss": 0.7698,
      "step": 219
    },
    {
      "epoch": 0.28162252980238417,
      "grad_norm": 0.8142496347427368,
      "learning_rate": 3.993702112058757e-05,
      "loss": 0.859,
      "step": 220
    },
    {
      "epoch": 0.2829026322105768,
      "grad_norm": 1.0279487371444702,
      "learning_rate": 3.9935768586167254e-05,
      "loss": 0.8509,
      "step": 221
    },
    {
      "epoch": 0.28418273461876953,
      "grad_norm": 0.7696442604064941,
      "learning_rate": 3.993450374103823e-05,
      "loss": 0.8516,
      "step": 222
    },
    {
      "epoch": 0.2854628370269622,
      "grad_norm": 0.8864192366600037,
      "learning_rate": 3.9933226586068666e-05,
      "loss": 0.8495,
      "step": 223
    },
    {
      "epoch": 0.28674293943515483,
      "grad_norm": 0.8386762738227844,
      "learning_rate": 3.9931937122135176e-05,
      "loss": 0.8755,
      "step": 224
    },
    {
      "epoch": 0.2880230418433475,
      "grad_norm": 0.7864452600479126,
      "learning_rate": 3.9930635350122806e-05,
      "loss": 0.8074,
      "step": 225
    },
    {
      "epoch": 0.28930314425154013,
      "grad_norm": 0.8790215849876404,
      "learning_rate": 3.992932127092508e-05,
      "loss": 0.8277,
      "step": 226
    },
    {
      "epoch": 0.2905832466597328,
      "grad_norm": 0.7088143825531006,
      "learning_rate": 3.9927994885443945e-05,
      "loss": 0.7891,
      "step": 227
    },
    {
      "epoch": 0.29186334906792544,
      "grad_norm": 0.7788523435592651,
      "learning_rate": 3.99266561945898e-05,
      "loss": 0.7683,
      "step": 228
    },
    {
      "epoch": 0.2931434514761181,
      "grad_norm": 0.8953355550765991,
      "learning_rate": 3.99253051992815e-05,
      "loss": 0.816,
      "step": 229
    },
    {
      "epoch": 0.29442355388431074,
      "grad_norm": 0.7870184779167175,
      "learning_rate": 3.992394190044632e-05,
      "loss": 0.8159,
      "step": 230
    },
    {
      "epoch": 0.2957036562925034,
      "grad_norm": 0.8162986636161804,
      "learning_rate": 3.9922566299020015e-05,
      "loss": 0.7715,
      "step": 231
    },
    {
      "epoch": 0.29698375870069604,
      "grad_norm": 0.8836362361907959,
      "learning_rate": 3.992117839594676e-05,
      "loss": 0.8133,
      "step": 232
    },
    {
      "epoch": 0.2982638611088887,
      "grad_norm": 0.742364227771759,
      "learning_rate": 3.991977819217919e-05,
      "loss": 0.8183,
      "step": 233
    },
    {
      "epoch": 0.29954396351708135,
      "grad_norm": 0.9332095980644226,
      "learning_rate": 3.9918365688678354e-05,
      "loss": 0.8485,
      "step": 234
    },
    {
      "epoch": 0.300824065925274,
      "grad_norm": 0.781606912612915,
      "learning_rate": 3.991694088641377e-05,
      "loss": 0.7872,
      "step": 235
    },
    {
      "epoch": 0.3021041683334667,
      "grad_norm": 0.9482907056808472,
      "learning_rate": 3.991550378636341e-05,
      "loss": 0.8303,
      "step": 236
    },
    {
      "epoch": 0.30338427074165936,
      "grad_norm": 0.7216508388519287,
      "learning_rate": 3.991405438951365e-05,
      "loss": 0.7753,
      "step": 237
    },
    {
      "epoch": 0.304664373149852,
      "grad_norm": 0.7598963379859924,
      "learning_rate": 3.991259269685932e-05,
      "loss": 0.7773,
      "step": 238
    },
    {
      "epoch": 0.30594447555804466,
      "grad_norm": 0.7879756689071655,
      "learning_rate": 3.9911118709403694e-05,
      "loss": 0.8307,
      "step": 239
    },
    {
      "epoch": 0.3072245779662373,
      "grad_norm": 0.8073533773422241,
      "learning_rate": 3.99096324281585e-05,
      "loss": 0.7971,
      "step": 240
    },
    {
      "epoch": 0.3072245779662373,
      "eval_loss": 0.012903096154332161,
      "eval_runtime": 10.8174,
      "eval_samples_per_second": 45.852,
      "eval_steps_per_second": 5.732,
      "step": 240
    },
    {
      "epoch": 0.30850468037442996,
      "grad_norm": 0.8798401951789856,
      "learning_rate": 3.9908133854143865e-05,
      "loss": 0.8206,
      "step": 241
    },
    {
      "epoch": 0.3097847827826226,
      "grad_norm": 0.9081045389175415,
      "learning_rate": 3.9906622988388396e-05,
      "loss": 0.8528,
      "step": 242
    },
    {
      "epoch": 0.31106488519081527,
      "grad_norm": 0.8628330230712891,
      "learning_rate": 3.99050998319291e-05,
      "loss": 0.8297,
      "step": 243
    },
    {
      "epoch": 0.3123449875990079,
      "grad_norm": 0.7450794577598572,
      "learning_rate": 3.9903564385811456e-05,
      "loss": 0.7667,
      "step": 244
    },
    {
      "epoch": 0.31362509000720057,
      "grad_norm": 0.7316635251045227,
      "learning_rate": 3.990201665108934e-05,
      "loss": 0.867,
      "step": 245
    },
    {
      "epoch": 0.3149051924153932,
      "grad_norm": 0.903037965297699,
      "learning_rate": 3.99004566288251e-05,
      "loss": 0.8491,
      "step": 246
    },
    {
      "epoch": 0.31618529482358587,
      "grad_norm": 0.7979913949966431,
      "learning_rate": 3.989888432008948e-05,
      "loss": 0.7929,
      "step": 247
    },
    {
      "epoch": 0.3174653972317785,
      "grad_norm": 0.859156608581543,
      "learning_rate": 3.989729972596171e-05,
      "loss": 0.7699,
      "step": 248
    },
    {
      "epoch": 0.3187454996399712,
      "grad_norm": 0.7992383241653442,
      "learning_rate": 3.989570284752939e-05,
      "loss": 0.8275,
      "step": 249
    },
    {
      "epoch": 0.3200256020481638,
      "grad_norm": 0.9154579043388367,
      "learning_rate": 3.9894093685888584e-05,
      "loss": 0.8396,
      "step": 250
    },
    {
      "epoch": 0.32130570445635653,
      "grad_norm": 0.7978524565696716,
      "learning_rate": 3.98924722421438e-05,
      "loss": 0.8499,
      "step": 251
    },
    {
      "epoch": 0.3225858068645492,
      "grad_norm": 0.6886290907859802,
      "learning_rate": 3.989083851740795e-05,
      "loss": 0.7635,
      "step": 252
    },
    {
      "epoch": 0.32386590927274184,
      "grad_norm": 0.8453454971313477,
      "learning_rate": 3.9889192512802395e-05,
      "loss": 0.8194,
      "step": 253
    },
    {
      "epoch": 0.3251460116809345,
      "grad_norm": 0.7061597108840942,
      "learning_rate": 3.9887534229456896e-05,
      "loss": 0.7581,
      "step": 254
    },
    {
      "epoch": 0.32642611408912714,
      "grad_norm": 0.8182460069656372,
      "learning_rate": 3.988586366850968e-05,
      "loss": 0.79,
      "step": 255
    },
    {
      "epoch": 0.3277062164973198,
      "grad_norm": 0.8259915709495544,
      "learning_rate": 3.988418083110738e-05,
      "loss": 0.8186,
      "step": 256
    },
    {
      "epoch": 0.32898631890551244,
      "grad_norm": 0.7348259687423706,
      "learning_rate": 3.9882485718405054e-05,
      "loss": 0.8269,
      "step": 257
    },
    {
      "epoch": 0.3302664213137051,
      "grad_norm": 0.7757918834686279,
      "learning_rate": 3.988077833156619e-05,
      "loss": 0.7898,
      "step": 258
    },
    {
      "epoch": 0.33154652372189775,
      "grad_norm": 0.883405864238739,
      "learning_rate": 3.9879058671762695e-05,
      "loss": 0.7759,
      "step": 259
    },
    {
      "epoch": 0.3328266261300904,
      "grad_norm": 0.7318927645683289,
      "learning_rate": 3.987732674017491e-05,
      "loss": 0.8345,
      "step": 260
    },
    {
      "epoch": 0.33410672853828305,
      "grad_norm": 0.9309760332107544,
      "learning_rate": 3.98755825379916e-05,
      "loss": 0.7986,
      "step": 261
    },
    {
      "epoch": 0.3353868309464757,
      "grad_norm": 0.895551860332489,
      "learning_rate": 3.987382606640993e-05,
      "loss": 0.8246,
      "step": 262
    },
    {
      "epoch": 0.33666693335466835,
      "grad_norm": 0.9246336817741394,
      "learning_rate": 3.9872057326635525e-05,
      "loss": 0.7803,
      "step": 263
    },
    {
      "epoch": 0.337947035762861,
      "grad_norm": 0.7929348349571228,
      "learning_rate": 3.987027631988239e-05,
      "loss": 0.7914,
      "step": 264
    },
    {
      "epoch": 0.3392271381710537,
      "grad_norm": 0.8753546476364136,
      "learning_rate": 3.9868483047372974e-05,
      "loss": 0.7532,
      "step": 265
    },
    {
      "epoch": 0.34050724057924636,
      "grad_norm": 0.7170605063438416,
      "learning_rate": 3.9866677510338136e-05,
      "loss": 0.8136,
      "step": 266
    },
    {
      "epoch": 0.341787342987439,
      "grad_norm": 0.8197619915008545,
      "learning_rate": 3.986485971001716e-05,
      "loss": 0.8157,
      "step": 267
    },
    {
      "epoch": 0.34306744539563166,
      "grad_norm": 0.8301973938941956,
      "learning_rate": 3.986302964765774e-05,
      "loss": 0.8593,
      "step": 268
    },
    {
      "epoch": 0.3443475478038243,
      "grad_norm": 0.7686181664466858,
      "learning_rate": 3.9861187324516e-05,
      "loss": 0.7886,
      "step": 269
    },
    {
      "epoch": 0.34562765021201697,
      "grad_norm": 0.8302679061889648,
      "learning_rate": 3.985933274185646e-05,
      "loss": 0.7867,
      "step": 270
    },
    {
      "epoch": 0.34562765021201697,
      "eval_loss": 0.012820749543607235,
      "eval_runtime": 10.8139,
      "eval_samples_per_second": 45.867,
      "eval_steps_per_second": 5.733,
      "step": 270
    },
    {
      "epoch": 0.3469077526202096,
      "grad_norm": 0.8299188017845154,
      "learning_rate": 3.985746590095206e-05,
      "loss": 0.792,
      "step": 271
    },
    {
      "epoch": 0.34818785502840227,
      "grad_norm": 0.7753996849060059,
      "learning_rate": 3.985558680308416e-05,
      "loss": 0.7904,
      "step": 272
    },
    {
      "epoch": 0.3494679574365949,
      "grad_norm": 0.9009094834327698,
      "learning_rate": 3.985369544954254e-05,
      "loss": 0.82,
      "step": 273
    },
    {
      "epoch": 0.3507480598447876,
      "grad_norm": 0.7270292043685913,
      "learning_rate": 3.985179184162537e-05,
      "loss": 0.7553,
      "step": 274
    },
    {
      "epoch": 0.3520281622529802,
      "grad_norm": 0.9866331219673157,
      "learning_rate": 3.984987598063925e-05,
      "loss": 0.8549,
      "step": 275
    },
    {
      "epoch": 0.3533082646611729,
      "grad_norm": 0.8446618914604187,
      "learning_rate": 3.984794786789919e-05,
      "loss": 0.7954,
      "step": 276
    },
    {
      "epoch": 0.35458836706936553,
      "grad_norm": 0.7152864933013916,
      "learning_rate": 3.9846007504728593e-05,
      "loss": 0.806,
      "step": 277
    },
    {
      "epoch": 0.3558684694775582,
      "grad_norm": 0.756089985370636,
      "learning_rate": 3.984405489245928e-05,
      "loss": 0.7996,
      "step": 278
    },
    {
      "epoch": 0.3571485718857509,
      "grad_norm": 0.7793037295341492,
      "learning_rate": 3.984209003243149e-05,
      "loss": 0.8117,
      "step": 279
    },
    {
      "epoch": 0.35842867429394354,
      "grad_norm": 0.708320140838623,
      "learning_rate": 3.984011292599385e-05,
      "loss": 0.7597,
      "step": 280
    },
    {
      "epoch": 0.3597087767021362,
      "grad_norm": 0.7415875196456909,
      "learning_rate": 3.983812357450341e-05,
      "loss": 0.771,
      "step": 281
    },
    {
      "epoch": 0.36098887911032884,
      "grad_norm": 0.7343917489051819,
      "learning_rate": 3.983612197932561e-05,
      "loss": 0.8324,
      "step": 282
    },
    {
      "epoch": 0.3622689815185215,
      "grad_norm": 0.722980260848999,
      "learning_rate": 3.98341081418343e-05,
      "loss": 0.7972,
      "step": 283
    },
    {
      "epoch": 0.36354908392671414,
      "grad_norm": 0.7382889986038208,
      "learning_rate": 3.983208206341173e-05,
      "loss": 0.7891,
      "step": 284
    },
    {
      "epoch": 0.3648291863349068,
      "grad_norm": 0.7934734225273132,
      "learning_rate": 3.983004374544856e-05,
      "loss": 0.8272,
      "step": 285
    },
    {
      "epoch": 0.36610928874309945,
      "grad_norm": 0.7105775475502014,
      "learning_rate": 3.982799318934385e-05,
      "loss": 0.8011,
      "step": 286
    },
    {
      "epoch": 0.3673893911512921,
      "grad_norm": 0.7467436194419861,
      "learning_rate": 3.9825930396505036e-05,
      "loss": 0.8085,
      "step": 287
    },
    {
      "epoch": 0.36866949355948475,
      "grad_norm": 0.7933696508407593,
      "learning_rate": 3.982385536834799e-05,
      "loss": 0.8297,
      "step": 288
    },
    {
      "epoch": 0.3699495959676774,
      "grad_norm": 0.6954778432846069,
      "learning_rate": 3.9821768106296975e-05,
      "loss": 0.8231,
      "step": 289
    },
    {
      "epoch": 0.37122969837587005,
      "grad_norm": 0.8585456609725952,
      "learning_rate": 3.981966861178462e-05,
      "loss": 0.7696,
      "step": 290
    },
    {
      "epoch": 0.3725098007840627,
      "grad_norm": 0.7246525883674622,
      "learning_rate": 3.981755688625197e-05,
      "loss": 0.8118,
      "step": 291
    },
    {
      "epoch": 0.37378990319225536,
      "grad_norm": 0.7333858609199524,
      "learning_rate": 3.981543293114849e-05,
      "loss": 0.8244,
      "step": 292
    },
    {
      "epoch": 0.375070005600448,
      "grad_norm": 0.742274284362793,
      "learning_rate": 3.981329674793198e-05,
      "loss": 0.793,
      "step": 293
    },
    {
      "epoch": 0.3763501080086407,
      "grad_norm": 0.6696738600730896,
      "learning_rate": 3.98111483380687e-05,
      "loss": 0.7953,
      "step": 294
    },
    {
      "epoch": 0.37763021041683337,
      "grad_norm": 0.8201417326927185,
      "learning_rate": 3.9808987703033256e-05,
      "loss": 0.7991,
      "step": 295
    },
    {
      "epoch": 0.378910312825026,
      "grad_norm": 0.7530762553215027,
      "learning_rate": 3.980681484430866e-05,
      "loss": 0.7908,
      "step": 296
    },
    {
      "epoch": 0.38019041523321867,
      "grad_norm": 0.7389295101165771,
      "learning_rate": 3.980462976338631e-05,
      "loss": 0.7866,
      "step": 297
    },
    {
      "epoch": 0.3814705176414113,
      "grad_norm": 0.743794322013855,
      "learning_rate": 3.9802432461766006e-05,
      "loss": 0.7843,
      "step": 298
    },
    {
      "epoch": 0.382750620049604,
      "grad_norm": 0.8816423416137695,
      "learning_rate": 3.980022294095592e-05,
      "loss": 0.814,
      "step": 299
    },
    {
      "epoch": 0.3840307224577966,
      "grad_norm": 0.7187123894691467,
      "learning_rate": 3.979800120247262e-05,
      "loss": 0.7907,
      "step": 300
    },
    {
      "epoch": 0.3840307224577966,
      "eval_loss": 0.01275508850812912,
      "eval_runtime": 10.8305,
      "eval_samples_per_second": 45.797,
      "eval_steps_per_second": 5.725,
      "step": 300
    },
    {
      "epoch": 0.3853108248659893,
      "grad_norm": 0.7645212411880493,
      "learning_rate": 3.979576724784105e-05,
      "loss": 0.7965,
      "step": 301
    },
    {
      "epoch": 0.38659092727418193,
      "grad_norm": 0.8202989101409912,
      "learning_rate": 3.9793521078594556e-05,
      "loss": 0.7962,
      "step": 302
    },
    {
      "epoch": 0.3878710296823746,
      "grad_norm": 0.7811014652252197,
      "learning_rate": 3.9791262696274854e-05,
      "loss": 0.7973,
      "step": 303
    },
    {
      "epoch": 0.38915113209056723,
      "grad_norm": 0.8531347513198853,
      "learning_rate": 3.9788992102432046e-05,
      "loss": 0.7826,
      "step": 304
    },
    {
      "epoch": 0.3904312344987599,
      "grad_norm": 0.7572221159934998,
      "learning_rate": 3.978670929862462e-05,
      "loss": 0.7612,
      "step": 305
    },
    {
      "epoch": 0.39171133690695253,
      "grad_norm": 0.8583319783210754,
      "learning_rate": 3.978441428641944e-05,
      "loss": 0.7906,
      "step": 306
    },
    {
      "epoch": 0.3929914393151452,
      "grad_norm": 0.8188625574111938,
      "learning_rate": 3.978210706739175e-05,
      "loss": 0.8348,
      "step": 307
    },
    {
      "epoch": 0.3942715417233379,
      "grad_norm": 1.0000718832015991,
      "learning_rate": 3.9779787643125174e-05,
      "loss": 0.8365,
      "step": 308
    },
    {
      "epoch": 0.39555164413153054,
      "grad_norm": 0.9057092070579529,
      "learning_rate": 3.977745601521171e-05,
      "loss": 0.8261,
      "step": 309
    },
    {
      "epoch": 0.3968317465397232,
      "grad_norm": 0.9990969300270081,
      "learning_rate": 3.9775112185251734e-05,
      "loss": 0.7879,
      "step": 310
    },
    {
      "epoch": 0.39811184894791585,
      "grad_norm": 0.8068785667419434,
      "learning_rate": 3.977275615485401e-05,
      "loss": 0.8057,
      "step": 311
    },
    {
      "epoch": 0.3993919513561085,
      "grad_norm": 0.9002063870429993,
      "learning_rate": 3.9770387925635645e-05,
      "loss": 0.8241,
      "step": 312
    },
    {
      "epoch": 0.40067205376430115,
      "grad_norm": 0.8019387125968933,
      "learning_rate": 3.976800749922215e-05,
      "loss": 0.7514,
      "step": 313
    },
    {
      "epoch": 0.4019521561724938,
      "grad_norm": 0.9053210616111755,
      "learning_rate": 3.97656148772474e-05,
      "loss": 0.8185,
      "step": 314
    },
    {
      "epoch": 0.40323225858068645,
      "grad_norm": 0.775916337966919,
      "learning_rate": 3.976321006135363e-05,
      "loss": 0.794,
      "step": 315
    },
    {
      "epoch": 0.4045123609888791,
      "grad_norm": 0.8259150385856628,
      "learning_rate": 3.9760793053191455e-05,
      "loss": 0.7842,
      "step": 316
    },
    {
      "epoch": 0.40579246339707176,
      "grad_norm": 0.7507197260856628,
      "learning_rate": 3.975836385441986e-05,
      "loss": 0.8159,
      "step": 317
    },
    {
      "epoch": 0.4070725658052644,
      "grad_norm": 0.8853293061256409,
      "learning_rate": 3.975592246670618e-05,
      "loss": 0.7754,
      "step": 318
    },
    {
      "epoch": 0.40835266821345706,
      "grad_norm": 0.8459185361862183,
      "learning_rate": 3.975346889172613e-05,
      "loss": 0.8205,
      "step": 319
    },
    {
      "epoch": 0.4096327706216497,
      "grad_norm": 0.8047323226928711,
      "learning_rate": 3.9751003131163805e-05,
      "loss": 0.7521,
      "step": 320
    },
    {
      "epoch": 0.41091287302984236,
      "grad_norm": 0.8247420191764832,
      "learning_rate": 3.974852518671163e-05,
      "loss": 0.7756,
      "step": 321
    },
    {
      "epoch": 0.41219297543803507,
      "grad_norm": 0.8434611558914185,
      "learning_rate": 3.974603506007042e-05,
      "loss": 0.7922,
      "step": 322
    },
    {
      "epoch": 0.4134730778462277,
      "grad_norm": 0.8224117755889893,
      "learning_rate": 3.9743532752949336e-05,
      "loss": 0.8011,
      "step": 323
    },
    {
      "epoch": 0.4147531802544204,
      "grad_norm": 0.7743213772773743,
      "learning_rate": 3.9741018267065904e-05,
      "loss": 0.7856,
      "step": 324
    },
    {
      "epoch": 0.416033282662613,
      "grad_norm": 0.7555202841758728,
      "learning_rate": 3.973849160414603e-05,
      "loss": 0.8128,
      "step": 325
    },
    {
      "epoch": 0.4173133850708057,
      "grad_norm": 0.7445181608200073,
      "learning_rate": 3.9735952765923934e-05,
      "loss": 0.7887,
      "step": 326
    },
    {
      "epoch": 0.4185934874789983,
      "grad_norm": 0.7434578537940979,
      "learning_rate": 3.973340175414222e-05,
      "loss": 0.7589,
      "step": 327
    },
    {
      "epoch": 0.419873589887191,
      "grad_norm": 0.7456271052360535,
      "learning_rate": 3.973083857055186e-05,
      "loss": 0.7783,
      "step": 328
    },
    {
      "epoch": 0.42115369229538363,
      "grad_norm": 0.8270319700241089,
      "learning_rate": 3.9728263216912164e-05,
      "loss": 0.8276,
      "step": 329
    },
    {
      "epoch": 0.4224337947035763,
      "grad_norm": 0.7310093641281128,
      "learning_rate": 3.972567569499078e-05,
      "loss": 0.8132,
      "step": 330
    },
    {
      "epoch": 0.4224337947035763,
      "eval_loss": 0.012658854946494102,
      "eval_runtime": 10.8224,
      "eval_samples_per_second": 45.831,
      "eval_steps_per_second": 5.729,
      "step": 330
    },
    {
      "epoch": 0.42371389711176893,
      "grad_norm": 0.8494604825973511,
      "learning_rate": 3.972307600656374e-05,
      "loss": 0.7795,
      "step": 331
    },
    {
      "epoch": 0.4249939995199616,
      "grad_norm": 0.7604356408119202,
      "learning_rate": 3.9720464153415416e-05,
      "loss": 0.8092,
      "step": 332
    },
    {
      "epoch": 0.42627410192815424,
      "grad_norm": 0.8328589200973511,
      "learning_rate": 3.97178401373385e-05,
      "loss": 0.8181,
      "step": 333
    },
    {
      "epoch": 0.4275542043363469,
      "grad_norm": 0.8251141309738159,
      "learning_rate": 3.971520396013409e-05,
      "loss": 0.8205,
      "step": 334
    },
    {
      "epoch": 0.42883430674453954,
      "grad_norm": 0.7609400153160095,
      "learning_rate": 3.9712555623611575e-05,
      "loss": 0.7965,
      "step": 335
    },
    {
      "epoch": 0.4301144091527322,
      "grad_norm": 0.8191712498664856,
      "learning_rate": 3.9709895129588726e-05,
      "loss": 0.8136,
      "step": 336
    },
    {
      "epoch": 0.4313945115609249,
      "grad_norm": 0.7615634202957153,
      "learning_rate": 3.970722247989164e-05,
      "loss": 0.7411,
      "step": 337
    },
    {
      "epoch": 0.43267461396911755,
      "grad_norm": 0.7785950303077698,
      "learning_rate": 3.9704537676354765e-05,
      "loss": 0.7888,
      "step": 338
    },
    {
      "epoch": 0.4339547163773102,
      "grad_norm": 0.6841676831245422,
      "learning_rate": 3.9701840720820894e-05,
      "loss": 0.7852,
      "step": 339
    },
    {
      "epoch": 0.43523481878550285,
      "grad_norm": 0.8268980979919434,
      "learning_rate": 3.9699131615141156e-05,
      "loss": 0.8169,
      "step": 340
    },
    {
      "epoch": 0.4365149211936955,
      "grad_norm": 0.7552794814109802,
      "learning_rate": 3.9696410361175014e-05,
      "loss": 0.7858,
      "step": 341
    },
    {
      "epoch": 0.43779502360188816,
      "grad_norm": 0.8271946907043457,
      "learning_rate": 3.9693676960790275e-05,
      "loss": 0.7858,
      "step": 342
    },
    {
      "epoch": 0.4390751260100808,
      "grad_norm": 0.7914268374443054,
      "learning_rate": 3.96909314158631e-05,
      "loss": 0.8279,
      "step": 343
    },
    {
      "epoch": 0.44035522841827346,
      "grad_norm": 0.9711649417877197,
      "learning_rate": 3.968817372827794e-05,
      "loss": 0.8156,
      "step": 344
    },
    {
      "epoch": 0.4416353308264661,
      "grad_norm": 0.80335533618927,
      "learning_rate": 3.968540389992765e-05,
      "loss": 0.7973,
      "step": 345
    },
    {
      "epoch": 0.44291543323465876,
      "grad_norm": 0.8563199043273926,
      "learning_rate": 3.968262193271334e-05,
      "loss": 0.8212,
      "step": 346
    },
    {
      "epoch": 0.4441955356428514,
      "grad_norm": 0.8743078112602234,
      "learning_rate": 3.967982782854451e-05,
      "loss": 0.7941,
      "step": 347
    },
    {
      "epoch": 0.44547563805104406,
      "grad_norm": 0.8524693846702576,
      "learning_rate": 3.967702158933896e-05,
      "loss": 0.7768,
      "step": 348
    },
    {
      "epoch": 0.4467557404592367,
      "grad_norm": 0.757647693157196,
      "learning_rate": 3.967420321702285e-05,
      "loss": 0.8124,
      "step": 349
    },
    {
      "epoch": 0.44803584286742937,
      "grad_norm": 0.8333066701889038,
      "learning_rate": 3.967137271353061e-05,
      "loss": 0.8022,
      "step": 350
    },
    {
      "epoch": 0.4493159452756221,
      "grad_norm": 0.7583582997322083,
      "learning_rate": 3.9668530080805076e-05,
      "loss": 0.777,
      "step": 351
    },
    {
      "epoch": 0.4505960476838147,
      "grad_norm": 0.8073552250862122,
      "learning_rate": 3.966567532079734e-05,
      "loss": 0.7815,
      "step": 352
    },
    {
      "epoch": 0.4518761500920074,
      "grad_norm": 0.678827702999115,
      "learning_rate": 3.966280843546685e-05,
      "loss": 0.8078,
      "step": 353
    },
    {
      "epoch": 0.45315625250020003,
      "grad_norm": 0.8286870718002319,
      "learning_rate": 3.965992942678139e-05,
      "loss": 0.7792,
      "step": 354
    },
    {
      "epoch": 0.4544363549083927,
      "grad_norm": 0.778968095779419,
      "learning_rate": 3.965703829671701e-05,
      "loss": 0.7639,
      "step": 355
    },
    {
      "epoch": 0.45571645731658533,
      "grad_norm": 0.7844759225845337,
      "learning_rate": 3.965413504725815e-05,
      "loss": 0.7913,
      "step": 356
    },
    {
      "epoch": 0.456996559724778,
      "grad_norm": 0.9054631590843201,
      "learning_rate": 3.965121968039753e-05,
      "loss": 0.8442,
      "step": 357
    },
    {
      "epoch": 0.45827666213297064,
      "grad_norm": 0.7725024223327637,
      "learning_rate": 3.9648292198136174e-05,
      "loss": 0.7878,
      "step": 358
    },
    {
      "epoch": 0.4595567645411633,
      "grad_norm": 0.8029242753982544,
      "learning_rate": 3.9645352602483456e-05,
      "loss": 0.7865,
      "step": 359
    },
    {
      "epoch": 0.46083686694935594,
      "grad_norm": 0.7134058475494385,
      "learning_rate": 3.964240089545704e-05,
      "loss": 0.8009,
      "step": 360
    },
    {
      "epoch": 0.46083686694935594,
      "eval_loss": 0.01259357575327158,
      "eval_runtime": 10.8203,
      "eval_samples_per_second": 45.84,
      "eval_steps_per_second": 5.73,
      "step": 360
    },
    {
      "epoch": 0.4621169693575486,
      "grad_norm": 0.6435779333114624,
      "learning_rate": 3.9639437079082926e-05,
      "loss": 0.7858,
      "step": 361
    },
    {
      "epoch": 0.46339707176574124,
      "grad_norm": 0.8096379637718201,
      "learning_rate": 3.963646115539539e-05,
      "loss": 0.7728,
      "step": 362
    },
    {
      "epoch": 0.4646771741739339,
      "grad_norm": 0.7162945866584778,
      "learning_rate": 3.9633473126437054e-05,
      "loss": 0.792,
      "step": 363
    },
    {
      "epoch": 0.46595727658212654,
      "grad_norm": 0.8785635232925415,
      "learning_rate": 3.963047299425884e-05,
      "loss": 0.807,
      "step": 364
    },
    {
      "epoch": 0.46723737899031925,
      "grad_norm": 0.6872981786727905,
      "learning_rate": 3.962746076091994e-05,
      "loss": 0.8063,
      "step": 365
    },
    {
      "epoch": 0.4685174813985119,
      "grad_norm": 0.7540907859802246,
      "learning_rate": 3.9624436428487925e-05,
      "loss": 0.7666,
      "step": 366
    },
    {
      "epoch": 0.46979758380670456,
      "grad_norm": 0.7178943157196045,
      "learning_rate": 3.96213999990386e-05,
      "loss": 0.7518,
      "step": 367
    },
    {
      "epoch": 0.4710776862148972,
      "grad_norm": 0.7184255123138428,
      "learning_rate": 3.961835147465611e-05,
      "loss": 0.772,
      "step": 368
    },
    {
      "epoch": 0.47235778862308986,
      "grad_norm": 0.8011344075202942,
      "learning_rate": 3.961529085743289e-05,
      "loss": 0.7815,
      "step": 369
    },
    {
      "epoch": 0.4736378910312825,
      "grad_norm": 0.7954471111297607,
      "learning_rate": 3.961221814946969e-05,
      "loss": 0.8032,
      "step": 370
    },
    {
      "epoch": 0.47491799343947516,
      "grad_norm": 0.7110918164253235,
      "learning_rate": 3.960913335287554e-05,
      "loss": 0.7966,
      "step": 371
    },
    {
      "epoch": 0.4761980958476678,
      "grad_norm": 0.8362737894058228,
      "learning_rate": 3.960603646976777e-05,
      "loss": 0.801,
      "step": 372
    },
    {
      "epoch": 0.47747819825586046,
      "grad_norm": 0.7137751579284668,
      "learning_rate": 3.9602927502272025e-05,
      "loss": 0.8186,
      "step": 373
    },
    {
      "epoch": 0.4787583006640531,
      "grad_norm": 0.7931938171386719,
      "learning_rate": 3.959980645252222e-05,
      "loss": 0.801,
      "step": 374
    },
    {
      "epoch": 0.48003840307224577,
      "grad_norm": 0.8369014263153076,
      "learning_rate": 3.959667332266057e-05,
      "loss": 0.8138,
      "step": 375
    },
    {
      "epoch": 0.4813185054804384,
      "grad_norm": 0.7229605317115784,
      "learning_rate": 3.95935281148376e-05,
      "loss": 0.7823,
      "step": 376
    },
    {
      "epoch": 0.48259860788863107,
      "grad_norm": 0.9785498976707458,
      "learning_rate": 3.959037083121209e-05,
      "loss": 0.7791,
      "step": 377
    },
    {
      "epoch": 0.4838787102968237,
      "grad_norm": 0.8153207302093506,
      "learning_rate": 3.958720147395115e-05,
      "loss": 0.7933,
      "step": 378
    },
    {
      "epoch": 0.48515881270501643,
      "grad_norm": 0.8983792662620544,
      "learning_rate": 3.958402004523014e-05,
      "loss": 0.7732,
      "step": 379
    },
    {
      "epoch": 0.4864389151132091,
      "grad_norm": 0.7396347522735596,
      "learning_rate": 3.958082654723273e-05,
      "loss": 0.8084,
      "step": 380
    },
    {
      "epoch": 0.48771901752140173,
      "grad_norm": 0.9474884867668152,
      "learning_rate": 3.957762098215085e-05,
      "loss": 0.7876,
      "step": 381
    },
    {
      "epoch": 0.4889991199295944,
      "grad_norm": 0.7463030815124512,
      "learning_rate": 3.9574403352184746e-05,
      "loss": 0.8113,
      "step": 382
    },
    {
      "epoch": 0.49027922233778704,
      "grad_norm": 0.9061795473098755,
      "learning_rate": 3.957117365954292e-05,
      "loss": 0.7946,
      "step": 383
    },
    {
      "epoch": 0.4915593247459797,
      "grad_norm": 0.7946358919143677,
      "learning_rate": 3.956793190644216e-05,
      "loss": 0.8142,
      "step": 384
    },
    {
      "epoch": 0.49283942715417234,
      "grad_norm": 0.7612783312797546,
      "learning_rate": 3.956467809510753e-05,
      "loss": 0.7787,
      "step": 385
    },
    {
      "epoch": 0.494119529562365,
      "grad_norm": 0.8278447389602661,
      "learning_rate": 3.956141222777239e-05,
      "loss": 0.8473,
      "step": 386
    },
    {
      "epoch": 0.49539963197055764,
      "grad_norm": 0.7168108224868774,
      "learning_rate": 3.955813430667833e-05,
      "loss": 0.7594,
      "step": 387
    },
    {
      "epoch": 0.4966797343787503,
      "grad_norm": 0.8312520980834961,
      "learning_rate": 3.955484433407526e-05,
      "loss": 0.8063,
      "step": 388
    },
    {
      "epoch": 0.49795983678694294,
      "grad_norm": 0.7526703476905823,
      "learning_rate": 3.955154231222134e-05,
      "loss": 0.7907,
      "step": 389
    },
    {
      "epoch": 0.4992399391951356,
      "grad_norm": 0.8451778292655945,
      "learning_rate": 3.9548228243383e-05,
      "loss": 0.7927,
      "step": 390
    },
    {
      "epoch": 0.4992399391951356,
      "eval_loss": 0.01254909485578537,
      "eval_runtime": 10.8303,
      "eval_samples_per_second": 45.797,
      "eval_steps_per_second": 5.725,
      "step": 390
    },
    {
      "epoch": 0.5005200416033283,
      "grad_norm": 0.7430022954940796,
      "learning_rate": 3.9544902129834945e-05,
      "loss": 0.7952,
      "step": 391
    },
    {
      "epoch": 0.501800144011521,
      "grad_norm": 0.7509642243385315,
      "learning_rate": 3.9541563973860146e-05,
      "loss": 0.7917,
      "step": 392
    },
    {
      "epoch": 0.5030802464197136,
      "grad_norm": 0.8211520910263062,
      "learning_rate": 3.953821377774984e-05,
      "loss": 0.8173,
      "step": 393
    },
    {
      "epoch": 0.5043603488279063,
      "grad_norm": 0.7656919360160828,
      "learning_rate": 3.953485154380352e-05,
      "loss": 0.782,
      "step": 394
    },
    {
      "epoch": 0.5056404512360989,
      "grad_norm": 0.9075120687484741,
      "learning_rate": 3.953147727432896e-05,
      "loss": 0.7809,
      "step": 395
    },
    {
      "epoch": 0.5069205536442916,
      "grad_norm": 0.8644618988037109,
      "learning_rate": 3.952809097164216e-05,
      "loss": 0.7783,
      "step": 396
    },
    {
      "epoch": 0.5082006560524842,
      "grad_norm": 0.8745702505111694,
      "learning_rate": 3.952469263806742e-05,
      "loss": 0.772,
      "step": 397
    },
    {
      "epoch": 0.5094807584606769,
      "grad_norm": 0.7140446305274963,
      "learning_rate": 3.952128227593728e-05,
      "loss": 0.7552,
      "step": 398
    },
    {
      "epoch": 0.5107608608688695,
      "grad_norm": 0.7114863991737366,
      "learning_rate": 3.951785988759253e-05,
      "loss": 0.8154,
      "step": 399
    },
    {
      "epoch": 0.5120409632770622,
      "grad_norm": 0.8365476131439209,
      "learning_rate": 3.9514425475382216e-05,
      "loss": 0.8243,
      "step": 400
    },
    {
      "epoch": 0.5133210656852548,
      "grad_norm": 0.7395104765892029,
      "learning_rate": 3.951097904166366e-05,
      "loss": 0.8199,
      "step": 401
    },
    {
      "epoch": 0.5146011680934475,
      "grad_norm": 0.8931131958961487,
      "learning_rate": 3.9507520588802386e-05,
      "loss": 0.7991,
      "step": 402
    },
    {
      "epoch": 0.5158812705016401,
      "grad_norm": 0.7250818014144897,
      "learning_rate": 3.950405011917223e-05,
      "loss": 0.8229,
      "step": 403
    },
    {
      "epoch": 0.5171613729098328,
      "grad_norm": 0.8582420349121094,
      "learning_rate": 3.9500567635155234e-05,
      "loss": 0.7623,
      "step": 404
    },
    {
      "epoch": 0.5184414753180254,
      "grad_norm": 0.7450950145721436,
      "learning_rate": 3.949707313914169e-05,
      "loss": 0.7681,
      "step": 405
    },
    {
      "epoch": 0.5197215777262181,
      "grad_norm": 0.8141098022460938,
      "learning_rate": 3.9493566633530146e-05,
      "loss": 0.8331,
      "step": 406
    },
    {
      "epoch": 0.5210016801344107,
      "grad_norm": 0.7822496294975281,
      "learning_rate": 3.949004812072738e-05,
      "loss": 0.7366,
      "step": 407
    },
    {
      "epoch": 0.5222817825426034,
      "grad_norm": 0.6154834628105164,
      "learning_rate": 3.948651760314844e-05,
      "loss": 0.7496,
      "step": 408
    },
    {
      "epoch": 0.523561884950796,
      "grad_norm": 0.7177313566207886,
      "learning_rate": 3.9482975083216575e-05,
      "loss": 0.8059,
      "step": 409
    },
    {
      "epoch": 0.5248419873589887,
      "grad_norm": 0.6349936723709106,
      "learning_rate": 3.94794205633633e-05,
      "loss": 0.7811,
      "step": 410
    },
    {
      "epoch": 0.5261220897671813,
      "grad_norm": 0.727734386920929,
      "learning_rate": 3.947585404602836e-05,
      "loss": 0.7975,
      "step": 411
    },
    {
      "epoch": 0.527402192175374,
      "grad_norm": 0.6441000699996948,
      "learning_rate": 3.947227553365973e-05,
      "loss": 0.7693,
      "step": 412
    },
    {
      "epoch": 0.5286822945835666,
      "grad_norm": 0.7431919574737549,
      "learning_rate": 3.9468685028713607e-05,
      "loss": 0.8235,
      "step": 413
    },
    {
      "epoch": 0.5299623969917593,
      "grad_norm": 0.651664137840271,
      "learning_rate": 3.9465082533654453e-05,
      "loss": 0.7452,
      "step": 414
    },
    {
      "epoch": 0.531242499399952,
      "grad_norm": 0.7347097396850586,
      "learning_rate": 3.9461468050954935e-05,
      "loss": 0.8066,
      "step": 415
    },
    {
      "epoch": 0.5325226018081447,
      "grad_norm": 0.8070315718650818,
      "learning_rate": 3.945784158309594e-05,
      "loss": 0.8177,
      "step": 416
    },
    {
      "epoch": 0.5338027042163374,
      "grad_norm": 0.7046098709106445,
      "learning_rate": 3.945420313256661e-05,
      "loss": 0.8039,
      "step": 417
    },
    {
      "epoch": 0.53508280662453,
      "grad_norm": 0.7942793965339661,
      "learning_rate": 3.9450552701864294e-05,
      "loss": 0.7839,
      "step": 418
    },
    {
      "epoch": 0.5363629090327227,
      "grad_norm": 0.720222532749176,
      "learning_rate": 3.9446890293494554e-05,
      "loss": 0.7682,
      "step": 419
    },
    {
      "epoch": 0.5376430114409153,
      "grad_norm": 0.6810816526412964,
      "learning_rate": 3.944321590997119e-05,
      "loss": 0.77,
      "step": 420
    },
    {
      "epoch": 0.5376430114409153,
      "eval_loss": 0.012503801845014095,
      "eval_runtime": 10.8277,
      "eval_samples_per_second": 45.808,
      "eval_steps_per_second": 5.726,
      "step": 420
    },
    {
      "epoch": 0.538923113849108,
      "grad_norm": 0.8036742210388184,
      "learning_rate": 3.943952955381622e-05,
      "loss": 0.8102,
      "step": 421
    },
    {
      "epoch": 0.5402032162573006,
      "grad_norm": 0.6643345952033997,
      "learning_rate": 3.943583122755987e-05,
      "loss": 0.7684,
      "step": 422
    },
    {
      "epoch": 0.5414833186654933,
      "grad_norm": 0.8585495352745056,
      "learning_rate": 3.94321209337406e-05,
      "loss": 0.7443,
      "step": 423
    },
    {
      "epoch": 0.5427634210736859,
      "grad_norm": 0.6907642483711243,
      "learning_rate": 3.942839867490506e-05,
      "loss": 0.7432,
      "step": 424
    },
    {
      "epoch": 0.5440435234818786,
      "grad_norm": 0.7666288614273071,
      "learning_rate": 3.942466445360813e-05,
      "loss": 0.7892,
      "step": 425
    },
    {
      "epoch": 0.5453236258900712,
      "grad_norm": 0.7625541687011719,
      "learning_rate": 3.9420918272412894e-05,
      "loss": 0.7922,
      "step": 426
    },
    {
      "epoch": 0.5466037282982639,
      "grad_norm": 0.8049483895301819,
      "learning_rate": 3.941716013389065e-05,
      "loss": 0.7801,
      "step": 427
    },
    {
      "epoch": 0.5478838307064565,
      "grad_norm": 0.7327088713645935,
      "learning_rate": 3.941339004062089e-05,
      "loss": 0.7857,
      "step": 428
    },
    {
      "epoch": 0.5491639331146492,
      "grad_norm": 0.7284425497055054,
      "learning_rate": 3.940960799519134e-05,
      "loss": 0.7463,
      "step": 429
    },
    {
      "epoch": 0.5504440355228418,
      "grad_norm": 0.7273387312889099,
      "learning_rate": 3.9405814000197894e-05,
      "loss": 0.8092,
      "step": 430
    },
    {
      "epoch": 0.5517241379310345,
      "grad_norm": 0.7149500250816345,
      "learning_rate": 3.940200805824468e-05,
      "loss": 0.7614,
      "step": 431
    },
    {
      "epoch": 0.5530042403392271,
      "grad_norm": 0.7388850450515747,
      "learning_rate": 3.9398190171943986e-05,
      "loss": 0.8254,
      "step": 432
    },
    {
      "epoch": 0.5542843427474198,
      "grad_norm": 0.7056301832199097,
      "learning_rate": 3.939436034391634e-05,
      "loss": 0.7872,
      "step": 433
    },
    {
      "epoch": 0.5555644451556124,
      "grad_norm": 0.7976992130279541,
      "learning_rate": 3.939051857679046e-05,
      "loss": 0.7727,
      "step": 434
    },
    {
      "epoch": 0.5568445475638051,
      "grad_norm": 0.8013225793838501,
      "learning_rate": 3.938666487320323e-05,
      "loss": 0.7981,
      "step": 435
    },
    {
      "epoch": 0.5581246499719977,
      "grad_norm": 0.7863363027572632,
      "learning_rate": 3.938279923579976e-05,
      "loss": 0.7918,
      "step": 436
    },
    {
      "epoch": 0.5594047523801904,
      "grad_norm": 0.7414101958274841,
      "learning_rate": 3.937892166723332e-05,
      "loss": 0.8134,
      "step": 437
    },
    {
      "epoch": 0.560684854788383,
      "grad_norm": 0.7970485687255859,
      "learning_rate": 3.9375032170165405e-05,
      "loss": 0.78,
      "step": 438
    },
    {
      "epoch": 0.5619649571965757,
      "grad_norm": 0.8212478160858154,
      "learning_rate": 3.937113074726567e-05,
      "loss": 0.7787,
      "step": 439
    },
    {
      "epoch": 0.5632450596047683,
      "grad_norm": 0.6914942264556885,
      "learning_rate": 3.936721740121196e-05,
      "loss": 0.7886,
      "step": 440
    },
    {
      "epoch": 0.564525162012961,
      "grad_norm": 0.8115800023078918,
      "learning_rate": 3.9363292134690315e-05,
      "loss": 0.7783,
      "step": 441
    },
    {
      "epoch": 0.5658052644211536,
      "grad_norm": 0.787767231464386,
      "learning_rate": 3.935935495039494e-05,
      "loss": 0.796,
      "step": 442
    },
    {
      "epoch": 0.5670853668293463,
      "grad_norm": 0.7171950340270996,
      "learning_rate": 3.935540585102824e-05,
      "loss": 0.8083,
      "step": 443
    },
    {
      "epoch": 0.5683654692375391,
      "grad_norm": 0.8266027569770813,
      "learning_rate": 3.935144483930078e-05,
      "loss": 0.8096,
      "step": 444
    },
    {
      "epoch": 0.5696455716457317,
      "grad_norm": 0.7655935287475586,
      "learning_rate": 3.9347471917931316e-05,
      "loss": 0.8342,
      "step": 445
    },
    {
      "epoch": 0.5709256740539244,
      "grad_norm": 0.8506702184677124,
      "learning_rate": 3.9343487089646774e-05,
      "loss": 0.8265,
      "step": 446
    },
    {
      "epoch": 0.572205776462117,
      "grad_norm": 0.7414137125015259,
      "learning_rate": 3.933949035718224e-05,
      "loss": 0.811,
      "step": 447
    },
    {
      "epoch": 0.5734858788703097,
      "grad_norm": 0.756046712398529,
      "learning_rate": 3.933548172328099e-05,
      "loss": 0.8081,
      "step": 448
    },
    {
      "epoch": 0.5747659812785023,
      "grad_norm": 0.7636081576347351,
      "learning_rate": 3.933146119069446e-05,
      "loss": 0.7879,
      "step": 449
    },
    {
      "epoch": 0.576046083686695,
      "grad_norm": 0.8503948450088501,
      "learning_rate": 3.9327428762182244e-05,
      "loss": 0.8393,
      "step": 450
    },
    {
      "epoch": 0.576046083686695,
      "eval_loss": 0.012458461336791515,
      "eval_runtime": 10.8963,
      "eval_samples_per_second": 45.52,
      "eval_steps_per_second": 5.69,
      "step": 450
    },
    {
      "epoch": 0.5773261860948876,
      "grad_norm": 0.7647830843925476,
      "learning_rate": 3.932338444051213e-05,
      "loss": 0.7619,
      "step": 451
    },
    {
      "epoch": 0.5786062885030803,
      "grad_norm": 0.7900508046150208,
      "learning_rate": 3.9319328228460024e-05,
      "loss": 0.7497,
      "step": 452
    },
    {
      "epoch": 0.5798863909112729,
      "grad_norm": 0.7502211332321167,
      "learning_rate": 3.931526012881004e-05,
      "loss": 0.7866,
      "step": 453
    },
    {
      "epoch": 0.5811664933194656,
      "grad_norm": 0.7293073534965515,
      "learning_rate": 3.931118014435442e-05,
      "loss": 0.754,
      "step": 454
    },
    {
      "epoch": 0.5824465957276582,
      "grad_norm": 0.7336457371711731,
      "learning_rate": 3.930708827789357e-05,
      "loss": 0.7793,
      "step": 455
    },
    {
      "epoch": 0.5837266981358509,
      "grad_norm": 0.7212786078453064,
      "learning_rate": 3.930298453223607e-05,
      "loss": 0.7615,
      "step": 456
    },
    {
      "epoch": 0.5850068005440435,
      "grad_norm": 0.6969265937805176,
      "learning_rate": 3.929886891019862e-05,
      "loss": 0.7404,
      "step": 457
    },
    {
      "epoch": 0.5862869029522362,
      "grad_norm": 0.8226010799407959,
      "learning_rate": 3.929474141460611e-05,
      "loss": 0.8168,
      "step": 458
    },
    {
      "epoch": 0.5875670053604288,
      "grad_norm": 0.6723736524581909,
      "learning_rate": 3.929060204829155e-05,
      "loss": 0.7588,
      "step": 459
    },
    {
      "epoch": 0.5888471077686215,
      "grad_norm": 0.7979745268821716,
      "learning_rate": 3.9286450814096106e-05,
      "loss": 0.8326,
      "step": 460
    },
    {
      "epoch": 0.5901272101768141,
      "grad_norm": 0.7417559027671814,
      "learning_rate": 3.928228771486909e-05,
      "loss": 0.7576,
      "step": 461
    },
    {
      "epoch": 0.5914073125850068,
      "grad_norm": 0.8069484829902649,
      "learning_rate": 3.927811275346797e-05,
      "loss": 0.8097,
      "step": 462
    },
    {
      "epoch": 0.5926874149931994,
      "grad_norm": 0.7181478142738342,
      "learning_rate": 3.927392593275834e-05,
      "loss": 0.7734,
      "step": 463
    },
    {
      "epoch": 0.5939675174013921,
      "grad_norm": 0.8048413991928101,
      "learning_rate": 3.926972725561394e-05,
      "loss": 0.7619,
      "step": 464
    },
    {
      "epoch": 0.5952476198095847,
      "grad_norm": 0.7611750960350037,
      "learning_rate": 3.926551672491665e-05,
      "loss": 0.8193,
      "step": 465
    },
    {
      "epoch": 0.5965277222177774,
      "grad_norm": 0.7776764631271362,
      "learning_rate": 3.926129434355647e-05,
      "loss": 0.8004,
      "step": 466
    },
    {
      "epoch": 0.59780782462597,
      "grad_norm": 0.7161852717399597,
      "learning_rate": 3.925706011443157e-05,
      "loss": 0.7896,
      "step": 467
    },
    {
      "epoch": 0.5990879270341627,
      "grad_norm": 0.7222532033920288,
      "learning_rate": 3.925281404044821e-05,
      "loss": 0.7605,
      "step": 468
    },
    {
      "epoch": 0.6003680294423553,
      "grad_norm": 0.6803270578384399,
      "learning_rate": 3.924855612452082e-05,
      "loss": 0.7784,
      "step": 469
    },
    {
      "epoch": 0.601648131850548,
      "grad_norm": 0.6683339476585388,
      "learning_rate": 3.924428636957192e-05,
      "loss": 0.8033,
      "step": 470
    },
    {
      "epoch": 0.6029282342587406,
      "grad_norm": 0.6430296301841736,
      "learning_rate": 3.924000477853218e-05,
      "loss": 0.7567,
      "step": 471
    },
    {
      "epoch": 0.6042083366669334,
      "grad_norm": 0.6537981033325195,
      "learning_rate": 3.923571135434039e-05,
      "loss": 0.7519,
      "step": 472
    },
    {
      "epoch": 0.6054884390751261,
      "grad_norm": 0.7790893912315369,
      "learning_rate": 3.923140609994346e-05,
      "loss": 0.7947,
      "step": 473
    },
    {
      "epoch": 0.6067685414833187,
      "grad_norm": 0.6284059286117554,
      "learning_rate": 3.922708901829641e-05,
      "loss": 0.7399,
      "step": 474
    },
    {
      "epoch": 0.6080486438915114,
      "grad_norm": 0.6981340050697327,
      "learning_rate": 3.92227601123624e-05,
      "loss": 0.7789,
      "step": 475
    },
    {
      "epoch": 0.609328746299704,
      "grad_norm": 0.8239121437072754,
      "learning_rate": 3.9218419385112693e-05,
      "loss": 0.7848,
      "step": 476
    },
    {
      "epoch": 0.6106088487078967,
      "grad_norm": 0.6836838126182556,
      "learning_rate": 3.921406683952666e-05,
      "loss": 0.785,
      "step": 477
    },
    {
      "epoch": 0.6118889511160893,
      "grad_norm": 0.8668432235717773,
      "learning_rate": 3.920970247859179e-05,
      "loss": 0.8173,
      "step": 478
    },
    {
      "epoch": 0.613169053524282,
      "grad_norm": 0.685899555683136,
      "learning_rate": 3.9205326305303695e-05,
      "loss": 0.7923,
      "step": 479
    },
    {
      "epoch": 0.6144491559324746,
      "grad_norm": 0.7608736753463745,
      "learning_rate": 3.920093832266607e-05,
      "loss": 0.8111,
      "step": 480
    },
    {
      "epoch": 0.6144491559324746,
      "eval_loss": 0.012420989573001862,
      "eval_runtime": 10.8235,
      "eval_samples_per_second": 45.826,
      "eval_steps_per_second": 5.728,
      "step": 480
    },
    {
      "epoch": 0.6157292583406673,
      "grad_norm": 0.722259521484375,
      "learning_rate": 3.9196538533690734e-05,
      "loss": 0.7814,
      "step": 481
    },
    {
      "epoch": 0.6170093607488599,
      "grad_norm": 0.7831443548202515,
      "learning_rate": 3.91921269413976e-05,
      "loss": 0.7865,
      "step": 482
    },
    {
      "epoch": 0.6182894631570526,
      "grad_norm": 0.7355871796607971,
      "learning_rate": 3.918770354881468e-05,
      "loss": 0.8156,
      "step": 483
    },
    {
      "epoch": 0.6195695655652452,
      "grad_norm": 0.6596747040748596,
      "learning_rate": 3.9183268358978107e-05,
      "loss": 0.7984,
      "step": 484
    },
    {
      "epoch": 0.6208496679734379,
      "grad_norm": 0.8083442449569702,
      "learning_rate": 3.9178821374932076e-05,
      "loss": 0.7642,
      "step": 485
    },
    {
      "epoch": 0.6221297703816305,
      "grad_norm": 0.7155700325965881,
      "learning_rate": 3.917436259972891e-05,
      "loss": 0.7798,
      "step": 486
    },
    {
      "epoch": 0.6234098727898232,
      "grad_norm": 0.7327741384506226,
      "learning_rate": 3.9169892036429014e-05,
      "loss": 0.7979,
      "step": 487
    },
    {
      "epoch": 0.6246899751980158,
      "grad_norm": 0.7384740710258484,
      "learning_rate": 3.916540968810087e-05,
      "loss": 0.7683,
      "step": 488
    },
    {
      "epoch": 0.6259700776062085,
      "grad_norm": 0.7598046660423279,
      "learning_rate": 3.916091555782108e-05,
      "loss": 0.7885,
      "step": 489
    },
    {
      "epoch": 0.6272501800144011,
      "grad_norm": 0.7548463344573975,
      "learning_rate": 3.915640964867429e-05,
      "loss": 0.783,
      "step": 490
    },
    {
      "epoch": 0.6285302824225938,
      "grad_norm": 0.7831941843032837,
      "learning_rate": 3.915189196375327e-05,
      "loss": 0.8326,
      "step": 491
    },
    {
      "epoch": 0.6298103848307864,
      "grad_norm": 0.6835455298423767,
      "learning_rate": 3.914736250615886e-05,
      "loss": 0.7877,
      "step": 492
    },
    {
      "epoch": 0.6310904872389791,
      "grad_norm": 0.7209146618843079,
      "learning_rate": 3.914282127899998e-05,
      "loss": 0.7708,
      "step": 493
    },
    {
      "epoch": 0.6323705896471717,
      "grad_norm": 0.641525149345398,
      "learning_rate": 3.9138268285393606e-05,
      "loss": 0.7663,
      "step": 494
    },
    {
      "epoch": 0.6336506920553644,
      "grad_norm": 0.8103832602500916,
      "learning_rate": 3.913370352846483e-05,
      "loss": 0.775,
      "step": 495
    },
    {
      "epoch": 0.634930794463557,
      "grad_norm": 0.7506917119026184,
      "learning_rate": 3.91291270113468e-05,
      "loss": 0.8159,
      "step": 496
    },
    {
      "epoch": 0.6362108968717497,
      "grad_norm": 0.7250286936759949,
      "learning_rate": 3.9124538737180714e-05,
      "loss": 0.8283,
      "step": 497
    },
    {
      "epoch": 0.6374909992799423,
      "grad_norm": 0.6864186525344849,
      "learning_rate": 3.911993870911588e-05,
      "loss": 0.7507,
      "step": 498
    },
    {
      "epoch": 0.638771101688135,
      "grad_norm": 0.7688778042793274,
      "learning_rate": 3.911532693030965e-05,
      "loss": 0.8236,
      "step": 499
    },
    {
      "epoch": 0.6400512040963277,
      "grad_norm": 0.6764448285102844,
      "learning_rate": 3.9110703403927434e-05,
      "loss": 0.7597,
      "step": 500
    },
    {
      "epoch": 0.6413313065045204,
      "grad_norm": 0.7230846285820007,
      "learning_rate": 3.910606813314273e-05,
      "loss": 0.7737,
      "step": 501
    },
    {
      "epoch": 0.6426114089127131,
      "grad_norm": 0.7076696157455444,
      "learning_rate": 3.9101421121137084e-05,
      "loss": 0.7603,
      "step": 502
    },
    {
      "epoch": 0.6438915113209057,
      "grad_norm": 0.7059433460235596,
      "learning_rate": 3.9096762371100094e-05,
      "loss": 0.8008,
      "step": 503
    },
    {
      "epoch": 0.6451716137290984,
      "grad_norm": 0.7637941837310791,
      "learning_rate": 3.909209188622942e-05,
      "loss": 0.7825,
      "step": 504
    },
    {
      "epoch": 0.646451716137291,
      "grad_norm": 0.7858322262763977,
      "learning_rate": 3.9087409669730777e-05,
      "loss": 0.7736,
      "step": 505
    },
    {
      "epoch": 0.6477318185454837,
      "grad_norm": 0.7069405317306519,
      "learning_rate": 3.908271572481795e-05,
      "loss": 0.809,
      "step": 506
    },
    {
      "epoch": 0.6490119209536763,
      "grad_norm": 0.7629151344299316,
      "learning_rate": 3.907801005471273e-05,
      "loss": 0.8079,
      "step": 507
    },
    {
      "epoch": 0.650292023361869,
      "grad_norm": 0.7711296081542969,
      "learning_rate": 3.907329266264501e-05,
      "loss": 0.7799,
      "step": 508
    },
    {
      "epoch": 0.6515721257700616,
      "grad_norm": 0.6990572810173035,
      "learning_rate": 3.906856355185269e-05,
      "loss": 0.8338,
      "step": 509
    },
    {
      "epoch": 0.6528522281782543,
      "grad_norm": 0.7535392642021179,
      "learning_rate": 3.906382272558172e-05,
      "loss": 0.7969,
      "step": 510
    },
    {
      "epoch": 0.6528522281782543,
      "eval_loss": 0.012368489056825638,
      "eval_runtime": 10.82,
      "eval_samples_per_second": 45.841,
      "eval_steps_per_second": 5.73,
      "step": 510
    },
    {
      "epoch": 0.6541323305864469,
      "grad_norm": 0.6653261184692383,
      "learning_rate": 3.90590701870861e-05,
      "loss": 0.8127,
      "step": 511
    },
    {
      "epoch": 0.6554124329946396,
      "grad_norm": 0.7311388850212097,
      "learning_rate": 3.9054305939627875e-05,
      "loss": 0.7701,
      "step": 512
    },
    {
      "epoch": 0.6566925354028322,
      "grad_norm": 0.7191159725189209,
      "learning_rate": 3.904952998647711e-05,
      "loss": 0.8356,
      "step": 513
    },
    {
      "epoch": 0.6579726378110249,
      "grad_norm": 0.7642592787742615,
      "learning_rate": 3.9044742330911904e-05,
      "loss": 0.7971,
      "step": 514
    },
    {
      "epoch": 0.6592527402192175,
      "grad_norm": 0.7250025272369385,
      "learning_rate": 3.9039942976218415e-05,
      "loss": 0.7625,
      "step": 515
    },
    {
      "epoch": 0.6605328426274102,
      "grad_norm": 0.8104845881462097,
      "learning_rate": 3.903513192569079e-05,
      "loss": 0.8328,
      "step": 516
    },
    {
      "epoch": 0.6618129450356028,
      "grad_norm": 0.7267283201217651,
      "learning_rate": 3.903030918263124e-05,
      "loss": 0.751,
      "step": 517
    },
    {
      "epoch": 0.6630930474437955,
      "grad_norm": 0.8060623407363892,
      "learning_rate": 3.9025474750349994e-05,
      "loss": 0.8218,
      "step": 518
    },
    {
      "epoch": 0.6643731498519881,
      "grad_norm": 0.7860483527183533,
      "learning_rate": 3.902062863216528e-05,
      "loss": 0.7816,
      "step": 519
    },
    {
      "epoch": 0.6656532522601808,
      "grad_norm": 0.7551194429397583,
      "learning_rate": 3.9015770831403385e-05,
      "loss": 0.7528,
      "step": 520
    },
    {
      "epoch": 0.6669333546683734,
      "grad_norm": 0.7541205883026123,
      "learning_rate": 3.9010901351398576e-05,
      "loss": 0.7909,
      "step": 521
    },
    {
      "epoch": 0.6682134570765661,
      "grad_norm": 0.7642291188240051,
      "learning_rate": 3.900602019549316e-05,
      "loss": 0.8077,
      "step": 522
    },
    {
      "epoch": 0.6694935594847587,
      "grad_norm": 0.8029669523239136,
      "learning_rate": 3.900112736703746e-05,
      "loss": 0.8171,
      "step": 523
    },
    {
      "epoch": 0.6707736618929514,
      "grad_norm": 0.7413590550422668,
      "learning_rate": 3.8996222869389784e-05,
      "loss": 0.8278,
      "step": 524
    },
    {
      "epoch": 0.672053764301144,
      "grad_norm": 0.7315031886100769,
      "learning_rate": 3.89913067059165e-05,
      "loss": 0.757,
      "step": 525
    },
    {
      "epoch": 0.6733338667093367,
      "grad_norm": 0.8058844208717346,
      "learning_rate": 3.8986378879991924e-05,
      "loss": 0.8072,
      "step": 526
    },
    {
      "epoch": 0.6746139691175294,
      "grad_norm": 0.6936185359954834,
      "learning_rate": 3.8981439394998425e-05,
      "loss": 0.7543,
      "step": 527
    },
    {
      "epoch": 0.675894071525722,
      "grad_norm": 0.7407493591308594,
      "learning_rate": 3.897648825432634e-05,
      "loss": 0.7685,
      "step": 528
    },
    {
      "epoch": 0.6771741739339147,
      "grad_norm": 0.6896268725395203,
      "learning_rate": 3.897152546137403e-05,
      "loss": 0.7794,
      "step": 529
    },
    {
      "epoch": 0.6784542763421074,
      "grad_norm": 0.716063916683197,
      "learning_rate": 3.896655101954783e-05,
      "loss": 0.7964,
      "step": 530
    },
    {
      "epoch": 0.6797343787503001,
      "grad_norm": 0.6614677309989929,
      "learning_rate": 3.896156493226211e-05,
      "loss": 0.8156,
      "step": 531
    },
    {
      "epoch": 0.6810144811584927,
      "grad_norm": 0.6699355840682983,
      "learning_rate": 3.8956567202939196e-05,
      "loss": 0.7828,
      "step": 532
    },
    {
      "epoch": 0.6822945835666854,
      "grad_norm": 0.6638008952140808,
      "learning_rate": 3.895155783500941e-05,
      "loss": 0.7758,
      "step": 533
    },
    {
      "epoch": 0.683574685974878,
      "grad_norm": 0.68021160364151,
      "learning_rate": 3.8946536831911074e-05,
      "loss": 0.815,
      "step": 534
    },
    {
      "epoch": 0.6848547883830707,
      "grad_norm": 0.6995747685432434,
      "learning_rate": 3.89415041970905e-05,
      "loss": 0.8041,
      "step": 535
    },
    {
      "epoch": 0.6861348907912633,
      "grad_norm": 0.6517994403839111,
      "learning_rate": 3.893645993400196e-05,
      "loss": 0.7734,
      "step": 536
    },
    {
      "epoch": 0.687414993199456,
      "grad_norm": 0.7438444495201111,
      "learning_rate": 3.893140404610773e-05,
      "loss": 0.7829,
      "step": 537
    },
    {
      "epoch": 0.6886950956076486,
      "grad_norm": 0.640255868434906,
      "learning_rate": 3.892633653687807e-05,
      "loss": 0.7597,
      "step": 538
    },
    {
      "epoch": 0.6899751980158413,
      "grad_norm": 0.7098150849342346,
      "learning_rate": 3.892125740979119e-05,
      "loss": 0.798,
      "step": 539
    },
    {
      "epoch": 0.6912553004240339,
      "grad_norm": 0.7514739036560059,
      "learning_rate": 3.891616666833329e-05,
      "loss": 0.823,
      "step": 540
    },
    {
      "epoch": 0.6912553004240339,
      "eval_loss": 0.012333170510828495,
      "eval_runtime": 10.8058,
      "eval_samples_per_second": 45.901,
      "eval_steps_per_second": 5.738,
      "step": 540
    },
    {
      "epoch": 0.6925354028322266,
      "grad_norm": 0.7085390090942383,
      "learning_rate": 3.891106431599854e-05,
      "loss": 0.7909,
      "step": 541
    },
    {
      "epoch": 0.6938155052404192,
      "grad_norm": 0.693387508392334,
      "learning_rate": 3.8905950356289095e-05,
      "loss": 0.7578,
      "step": 542
    },
    {
      "epoch": 0.6950956076486119,
      "grad_norm": 0.7129430174827576,
      "learning_rate": 3.890082479271504e-05,
      "loss": 0.8122,
      "step": 543
    },
    {
      "epoch": 0.6963757100568045,
      "grad_norm": 0.6622198224067688,
      "learning_rate": 3.889568762879446e-05,
      "loss": 0.7629,
      "step": 544
    },
    {
      "epoch": 0.6976558124649972,
      "grad_norm": 0.7511202692985535,
      "learning_rate": 3.889053886805339e-05,
      "loss": 0.79,
      "step": 545
    },
    {
      "epoch": 0.6989359148731898,
      "grad_norm": 0.7332901954650879,
      "learning_rate": 3.888537851402582e-05,
      "loss": 0.8003,
      "step": 546
    },
    {
      "epoch": 0.7002160172813825,
      "grad_norm": 0.7890613079071045,
      "learning_rate": 3.8880206570253694e-05,
      "loss": 0.7777,
      "step": 547
    },
    {
      "epoch": 0.7014961196895751,
      "grad_norm": 0.7728927135467529,
      "learning_rate": 3.8875023040286926e-05,
      "loss": 0.7968,
      "step": 548
    },
    {
      "epoch": 0.7027762220977678,
      "grad_norm": 0.7917543649673462,
      "learning_rate": 3.886982792768338e-05,
      "loss": 0.7955,
      "step": 549
    },
    {
      "epoch": 0.7040563245059605,
      "grad_norm": 0.7046210169792175,
      "learning_rate": 3.886462123600885e-05,
      "loss": 0.7961,
      "step": 550
    },
    {
      "epoch": 0.7053364269141531,
      "grad_norm": 0.8072237372398376,
      "learning_rate": 3.8859402968837106e-05,
      "loss": 0.7975,
      "step": 551
    },
    {
      "epoch": 0.7066165293223458,
      "grad_norm": 0.6598737835884094,
      "learning_rate": 3.8854173129749846e-05,
      "loss": 0.729,
      "step": 552
    },
    {
      "epoch": 0.7078966317305384,
      "grad_norm": 0.8321033120155334,
      "learning_rate": 3.8848931722336705e-05,
      "loss": 0.8004,
      "step": 553
    },
    {
      "epoch": 0.7091767341387311,
      "grad_norm": 0.6804755330085754,
      "learning_rate": 3.884367875019528e-05,
      "loss": 0.7632,
      "step": 554
    },
    {
      "epoch": 0.7104568365469237,
      "grad_norm": 0.8645675778388977,
      "learning_rate": 3.8838414216931076e-05,
      "loss": 0.813,
      "step": 555
    },
    {
      "epoch": 0.7117369389551164,
      "grad_norm": 0.6690579652786255,
      "learning_rate": 3.883313812615757e-05,
      "loss": 0.7636,
      "step": 556
    },
    {
      "epoch": 0.713017041363309,
      "grad_norm": 0.8966508507728577,
      "learning_rate": 3.882785048149615e-05,
      "loss": 0.7623,
      "step": 557
    },
    {
      "epoch": 0.7142971437715018,
      "grad_norm": 0.7796165943145752,
      "learning_rate": 3.882255128657612e-05,
      "loss": 0.7908,
      "step": 558
    },
    {
      "epoch": 0.7155772461796944,
      "grad_norm": 0.7847517132759094,
      "learning_rate": 3.881724054503474e-05,
      "loss": 0.7773,
      "step": 559
    },
    {
      "epoch": 0.7168573485878871,
      "grad_norm": 0.7085537314414978,
      "learning_rate": 3.881191826051719e-05,
      "loss": 0.7696,
      "step": 560
    },
    {
      "epoch": 0.7181374509960797,
      "grad_norm": 0.7186529636383057,
      "learning_rate": 3.880658443667655e-05,
      "loss": 0.7908,
      "step": 561
    },
    {
      "epoch": 0.7194175534042724,
      "grad_norm": 0.6572086215019226,
      "learning_rate": 3.8801239077173847e-05,
      "loss": 0.7448,
      "step": 562
    },
    {
      "epoch": 0.720697655812465,
      "grad_norm": 0.7846883535385132,
      "learning_rate": 3.879588218567802e-05,
      "loss": 0.8326,
      "step": 563
    },
    {
      "epoch": 0.7219777582206577,
      "grad_norm": 0.6568480730056763,
      "learning_rate": 3.879051376586591e-05,
      "loss": 0.752,
      "step": 564
    },
    {
      "epoch": 0.7232578606288503,
      "grad_norm": 0.7430264949798584,
      "learning_rate": 3.878513382142228e-05,
      "loss": 0.7381,
      "step": 565
    },
    {
      "epoch": 0.724537963037043,
      "grad_norm": 0.6743800044059753,
      "learning_rate": 3.8779742356039815e-05,
      "loss": 0.7667,
      "step": 566
    },
    {
      "epoch": 0.7258180654452356,
      "grad_norm": 0.7817213535308838,
      "learning_rate": 3.877433937341908e-05,
      "loss": 0.7419,
      "step": 567
    },
    {
      "epoch": 0.7270981678534283,
      "grad_norm": 0.7134994268417358,
      "learning_rate": 3.876892487726857e-05,
      "loss": 0.7696,
      "step": 568
    },
    {
      "epoch": 0.7283782702616209,
      "grad_norm": 0.8109943866729736,
      "learning_rate": 3.876349887130467e-05,
      "loss": 0.802,
      "step": 569
    },
    {
      "epoch": 0.7296583726698136,
      "grad_norm": 0.7262176871299744,
      "learning_rate": 3.8758061359251675e-05,
      "loss": 0.7646,
      "step": 570
    },
    {
      "epoch": 0.7296583726698136,
      "eval_loss": 0.012286221608519554,
      "eval_runtime": 10.8217,
      "eval_samples_per_second": 45.834,
      "eval_steps_per_second": 5.729,
      "step": 570
    },
    {
      "epoch": 0.7309384750780062,
      "grad_norm": 0.6469229459762573,
      "learning_rate": 3.875261234484176e-05,
      "loss": 0.7652,
      "step": 571
    },
    {
      "epoch": 0.7322185774861989,
      "grad_norm": 0.7295794486999512,
      "learning_rate": 3.874715183181502e-05,
      "loss": 0.7786,
      "step": 572
    },
    {
      "epoch": 0.7334986798943915,
      "grad_norm": 0.7332335710525513,
      "learning_rate": 3.8741679823919415e-05,
      "loss": 0.7654,
      "step": 573
    },
    {
      "epoch": 0.7347787823025842,
      "grad_norm": 0.6600735783576965,
      "learning_rate": 3.873619632491081e-05,
      "loss": 0.7617,
      "step": 574
    },
    {
      "epoch": 0.7360588847107769,
      "grad_norm": 0.7789995074272156,
      "learning_rate": 3.8730701338552965e-05,
      "loss": 0.7876,
      "step": 575
    },
    {
      "epoch": 0.7373389871189695,
      "grad_norm": 0.6553863286972046,
      "learning_rate": 3.872519486861752e-05,
      "loss": 0.7613,
      "step": 576
    },
    {
      "epoch": 0.7386190895271622,
      "grad_norm": 0.6660206317901611,
      "learning_rate": 3.871967691888397e-05,
      "loss": 0.78,
      "step": 577
    },
    {
      "epoch": 0.7398991919353548,
      "grad_norm": 0.7108450531959534,
      "learning_rate": 3.871414749313972e-05,
      "loss": 0.7856,
      "step": 578
    },
    {
      "epoch": 0.7411792943435475,
      "grad_norm": 0.7628849744796753,
      "learning_rate": 3.870860659518006e-05,
      "loss": 0.7903,
      "step": 579
    },
    {
      "epoch": 0.7424593967517401,
      "grad_norm": 0.7231595516204834,
      "learning_rate": 3.870305422880812e-05,
      "loss": 0.7782,
      "step": 580
    },
    {
      "epoch": 0.7437394991599328,
      "grad_norm": 0.6817286610603333,
      "learning_rate": 3.869749039783492e-05,
      "loss": 0.7999,
      "step": 581
    },
    {
      "epoch": 0.7450196015681254,
      "grad_norm": 0.7689202427864075,
      "learning_rate": 3.869191510607936e-05,
      "loss": 0.7923,
      "step": 582
    },
    {
      "epoch": 0.7462997039763181,
      "grad_norm": 0.7295532822608948,
      "learning_rate": 3.868632835736819e-05,
      "loss": 0.781,
      "step": 583
    },
    {
      "epoch": 0.7475798063845107,
      "grad_norm": 0.8823017477989197,
      "learning_rate": 3.8680730155536024e-05,
      "loss": 0.8224,
      "step": 584
    },
    {
      "epoch": 0.7488599087927034,
      "grad_norm": 0.7479209899902344,
      "learning_rate": 3.8675120504425345e-05,
      "loss": 0.7673,
      "step": 585
    },
    {
      "epoch": 0.750140011200896,
      "grad_norm": 0.7687907218933105,
      "learning_rate": 3.86694994078865e-05,
      "loss": 0.7566,
      "step": 586
    },
    {
      "epoch": 0.7514201136090888,
      "grad_norm": 0.7190302014350891,
      "learning_rate": 3.866386686977766e-05,
      "loss": 0.7997,
      "step": 587
    },
    {
      "epoch": 0.7527002160172814,
      "grad_norm": 0.8372753262519836,
      "learning_rate": 3.86582228939649e-05,
      "loss": 0.8182,
      "step": 588
    },
    {
      "epoch": 0.7539803184254741,
      "grad_norm": 0.659578800201416,
      "learning_rate": 3.86525674843221e-05,
      "loss": 0.7934,
      "step": 589
    },
    {
      "epoch": 0.7552604208336667,
      "grad_norm": 0.7682313919067383,
      "learning_rate": 3.864690064473102e-05,
      "loss": 0.8022,
      "step": 590
    },
    {
      "epoch": 0.7565405232418594,
      "grad_norm": 0.6671510338783264,
      "learning_rate": 3.864122237908123e-05,
      "loss": 0.7892,
      "step": 591
    },
    {
      "epoch": 0.757820625650052,
      "grad_norm": 0.7912404537200928,
      "learning_rate": 3.8635532691270196e-05,
      "loss": 0.7975,
      "step": 592
    },
    {
      "epoch": 0.7591007280582447,
      "grad_norm": 0.6808180212974548,
      "learning_rate": 3.862983158520316e-05,
      "loss": 0.7717,
      "step": 593
    },
    {
      "epoch": 0.7603808304664373,
      "grad_norm": 0.6629878282546997,
      "learning_rate": 3.862411906479326e-05,
      "loss": 0.7576,
      "step": 594
    },
    {
      "epoch": 0.76166093287463,
      "grad_norm": 0.6980240345001221,
      "learning_rate": 3.861839513396142e-05,
      "loss": 0.7656,
      "step": 595
    },
    {
      "epoch": 0.7629410352828226,
      "grad_norm": 0.6688075065612793,
      "learning_rate": 3.861265979663643e-05,
      "loss": 0.7729,
      "step": 596
    },
    {
      "epoch": 0.7642211376910153,
      "grad_norm": 0.734492301940918,
      "learning_rate": 3.860691305675489e-05,
      "loss": 0.7548,
      "step": 597
    },
    {
      "epoch": 0.765501240099208,
      "grad_norm": 0.6866258382797241,
      "learning_rate": 3.8601154918261235e-05,
      "loss": 0.7937,
      "step": 598
    },
    {
      "epoch": 0.7667813425074006,
      "grad_norm": 0.6859091520309448,
      "learning_rate": 3.859538538510772e-05,
      "loss": 0.8092,
      "step": 599
    },
    {
      "epoch": 0.7680614449155932,
      "grad_norm": 0.6623457074165344,
      "learning_rate": 3.858960446125443e-05,
      "loss": 0.738,
      "step": 600
    },
    {
      "epoch": 0.7680614449155932,
      "eval_loss": 0.01225673034787178,
      "eval_runtime": 10.8122,
      "eval_samples_per_second": 45.874,
      "eval_steps_per_second": 5.734,
      "step": 600
    },
    {
      "epoch": 0.7693415473237859,
      "grad_norm": 0.7174913883209229,
      "learning_rate": 3.858381215066926e-05,
      "loss": 0.7629,
      "step": 601
    },
    {
      "epoch": 0.7706216497319786,
      "grad_norm": 0.6979793906211853,
      "learning_rate": 3.857800845732792e-05,
      "loss": 0.8092,
      "step": 602
    },
    {
      "epoch": 0.7719017521401712,
      "grad_norm": 0.8007774949073792,
      "learning_rate": 3.857219338521393e-05,
      "loss": 0.7728,
      "step": 603
    },
    {
      "epoch": 0.7731818545483639,
      "grad_norm": 0.6879320740699768,
      "learning_rate": 3.856636693831863e-05,
      "loss": 0.7655,
      "step": 604
    },
    {
      "epoch": 0.7744619569565565,
      "grad_norm": 0.727526068687439,
      "learning_rate": 3.856052912064116e-05,
      "loss": 0.7675,
      "step": 605
    },
    {
      "epoch": 0.7757420593647492,
      "grad_norm": 0.678472638130188,
      "learning_rate": 3.855467993618847e-05,
      "loss": 0.7878,
      "step": 606
    },
    {
      "epoch": 0.7770221617729418,
      "grad_norm": 0.7199679017066956,
      "learning_rate": 3.854881938897531e-05,
      "loss": 0.7834,
      "step": 607
    },
    {
      "epoch": 0.7783022641811345,
      "grad_norm": 0.6436090469360352,
      "learning_rate": 3.854294748302422e-05,
      "loss": 0.7722,
      "step": 608
    },
    {
      "epoch": 0.7795823665893271,
      "grad_norm": 0.6526851654052734,
      "learning_rate": 3.8537064222365545e-05,
      "loss": 0.8065,
      "step": 609
    },
    {
      "epoch": 0.7808624689975198,
      "grad_norm": 0.6819186806678772,
      "learning_rate": 3.853116961103744e-05,
      "loss": 0.8207,
      "step": 610
    },
    {
      "epoch": 0.7821425714057124,
      "grad_norm": 0.6532469987869263,
      "learning_rate": 3.852526365308581e-05,
      "loss": 0.7678,
      "step": 611
    },
    {
      "epoch": 0.7834226738139051,
      "grad_norm": 0.6011775135993958,
      "learning_rate": 3.851934635256438e-05,
      "loss": 0.7885,
      "step": 612
    },
    {
      "epoch": 0.7847027762220977,
      "grad_norm": 0.5763460993766785,
      "learning_rate": 3.8513417713534676e-05,
      "loss": 0.7549,
      "step": 613
    },
    {
      "epoch": 0.7859828786302904,
      "grad_norm": 0.5785894989967346,
      "learning_rate": 3.850747774006594e-05,
      "loss": 0.8032,
      "step": 614
    },
    {
      "epoch": 0.7872629810384831,
      "grad_norm": 0.6389304399490356,
      "learning_rate": 3.850152643623527e-05,
      "loss": 0.7706,
      "step": 615
    },
    {
      "epoch": 0.7885430834466758,
      "grad_norm": 0.5873603820800781,
      "learning_rate": 3.849556380612749e-05,
      "loss": 0.7763,
      "step": 616
    },
    {
      "epoch": 0.7898231858548684,
      "grad_norm": 0.614728569984436,
      "learning_rate": 3.848958985383522e-05,
      "loss": 0.7387,
      "step": 617
    },
    {
      "epoch": 0.7911032882630611,
      "grad_norm": 0.6720883846282959,
      "learning_rate": 3.848360458345885e-05,
      "loss": 0.7901,
      "step": 618
    },
    {
      "epoch": 0.7923833906712537,
      "grad_norm": 0.6720901131629944,
      "learning_rate": 3.847760799910654e-05,
      "loss": 0.802,
      "step": 619
    },
    {
      "epoch": 0.7936634930794464,
      "grad_norm": 0.671968936920166,
      "learning_rate": 3.847160010489419e-05,
      "loss": 0.761,
      "step": 620
    },
    {
      "epoch": 0.794943595487639,
      "grad_norm": 0.7273372411727905,
      "learning_rate": 3.84655809049455e-05,
      "loss": 0.7881,
      "step": 621
    },
    {
      "epoch": 0.7962236978958317,
      "grad_norm": 0.7307608723640442,
      "learning_rate": 3.8459550403391916e-05,
      "loss": 0.7909,
      "step": 622
    },
    {
      "epoch": 0.7975038003040243,
      "grad_norm": 0.7239599227905273,
      "learning_rate": 3.8453508604372627e-05,
      "loss": 0.8082,
      "step": 623
    },
    {
      "epoch": 0.798783902712217,
      "grad_norm": 0.673768162727356,
      "learning_rate": 3.84474555120346e-05,
      "loss": 0.7329,
      "step": 624
    },
    {
      "epoch": 0.8000640051204096,
      "grad_norm": 0.6614078283309937,
      "learning_rate": 3.844139113053253e-05,
      "loss": 0.7377,
      "step": 625
    },
    {
      "epoch": 0.8013441075286023,
      "grad_norm": 0.6718165874481201,
      "learning_rate": 3.843531546402889e-05,
      "loss": 0.7872,
      "step": 626
    },
    {
      "epoch": 0.802624209936795,
      "grad_norm": 0.7179350852966309,
      "learning_rate": 3.8429228516693856e-05,
      "loss": 0.7596,
      "step": 627
    },
    {
      "epoch": 0.8039043123449876,
      "grad_norm": 0.6812868714332581,
      "learning_rate": 3.842313029270539e-05,
      "loss": 0.8254,
      "step": 628
    },
    {
      "epoch": 0.8051844147531803,
      "grad_norm": 0.6353902816772461,
      "learning_rate": 3.841702079624918e-05,
      "loss": 0.7819,
      "step": 629
    },
    {
      "epoch": 0.8064645171613729,
      "grad_norm": 0.6739655137062073,
      "learning_rate": 3.841090003151863e-05,
      "loss": 0.7659,
      "step": 630
    },
    {
      "epoch": 0.8064645171613729,
      "eval_loss": 0.0122321592643857,
      "eval_runtime": 10.8138,
      "eval_samples_per_second": 45.867,
      "eval_steps_per_second": 5.733,
      "step": 630
    },
    {
      "epoch": 0.8077446195695656,
      "grad_norm": 0.7184500694274902,
      "learning_rate": 3.8404768002714904e-05,
      "loss": 0.7462,
      "step": 631
    },
    {
      "epoch": 0.8090247219777582,
      "grad_norm": 0.6388519406318665,
      "learning_rate": 3.839862471404689e-05,
      "loss": 0.7681,
      "step": 632
    },
    {
      "epoch": 0.8103048243859509,
      "grad_norm": 0.7300659418106079,
      "learning_rate": 3.8392470169731206e-05,
      "loss": 0.785,
      "step": 633
    },
    {
      "epoch": 0.8115849267941435,
      "grad_norm": 0.678228497505188,
      "learning_rate": 3.8386304373992185e-05,
      "loss": 0.7745,
      "step": 634
    },
    {
      "epoch": 0.8128650292023362,
      "grad_norm": 0.781303346157074,
      "learning_rate": 3.8380127331061895e-05,
      "loss": 0.7683,
      "step": 635
    },
    {
      "epoch": 0.8141451316105288,
      "grad_norm": 0.6938972473144531,
      "learning_rate": 3.837393904518012e-05,
      "loss": 0.798,
      "step": 636
    },
    {
      "epoch": 0.8154252340187215,
      "grad_norm": 0.7247546911239624,
      "learning_rate": 3.836773952059436e-05,
      "loss": 0.7478,
      "step": 637
    },
    {
      "epoch": 0.8167053364269141,
      "grad_norm": 0.7374678254127502,
      "learning_rate": 3.8361528761559834e-05,
      "loss": 0.7816,
      "step": 638
    },
    {
      "epoch": 0.8179854388351068,
      "grad_norm": 0.6233804225921631,
      "learning_rate": 3.835530677233946e-05,
      "loss": 0.7543,
      "step": 639
    },
    {
      "epoch": 0.8192655412432994,
      "grad_norm": 0.6578785181045532,
      "learning_rate": 3.8349073557203875e-05,
      "loss": 0.7197,
      "step": 640
    },
    {
      "epoch": 0.8205456436514921,
      "grad_norm": 0.6945452094078064,
      "learning_rate": 3.834282912043141e-05,
      "loss": 0.7486,
      "step": 641
    },
    {
      "epoch": 0.8218257460596847,
      "grad_norm": 0.6540107727050781,
      "learning_rate": 3.833657346630812e-05,
      "loss": 0.7944,
      "step": 642
    },
    {
      "epoch": 0.8231058484678774,
      "grad_norm": 0.8248779773712158,
      "learning_rate": 3.833030659912774e-05,
      "loss": 0.79,
      "step": 643
    },
    {
      "epoch": 0.8243859508760701,
      "grad_norm": 0.669266402721405,
      "learning_rate": 3.8324028523191703e-05,
      "loss": 0.7307,
      "step": 644
    },
    {
      "epoch": 0.8256660532842628,
      "grad_norm": 0.7554711699485779,
      "learning_rate": 3.8317739242809144e-05,
      "loss": 0.777,
      "step": 645
    },
    {
      "epoch": 0.8269461556924554,
      "grad_norm": 0.720956027507782,
      "learning_rate": 3.831143876229688e-05,
      "loss": 0.7577,
      "step": 646
    },
    {
      "epoch": 0.8282262581006481,
      "grad_norm": 0.6820900440216064,
      "learning_rate": 3.830512708597942e-05,
      "loss": 0.7694,
      "step": 647
    },
    {
      "epoch": 0.8295063605088407,
      "grad_norm": 0.809729814529419,
      "learning_rate": 3.8298804218188956e-05,
      "loss": 0.848,
      "step": 648
    },
    {
      "epoch": 0.8307864629170334,
      "grad_norm": 0.7035766839981079,
      "learning_rate": 3.8292470163265375e-05,
      "loss": 0.7782,
      "step": 649
    },
    {
      "epoch": 0.832066565325226,
      "grad_norm": 0.7612035274505615,
      "learning_rate": 3.8286124925556206e-05,
      "loss": 0.7292,
      "step": 650
    },
    {
      "epoch": 0.8333466677334187,
      "grad_norm": 0.7206668257713318,
      "learning_rate": 3.8279768509416705e-05,
      "loss": 0.7793,
      "step": 651
    },
    {
      "epoch": 0.8346267701416114,
      "grad_norm": 0.8308065533638,
      "learning_rate": 3.827340091920975e-05,
      "loss": 0.7912,
      "step": 652
    },
    {
      "epoch": 0.835906872549804,
      "grad_norm": 0.66724693775177,
      "learning_rate": 3.826702215930593e-05,
      "loss": 0.7879,
      "step": 653
    },
    {
      "epoch": 0.8371869749579967,
      "grad_norm": 0.8065431714057922,
      "learning_rate": 3.826063223408346e-05,
      "loss": 0.7904,
      "step": 654
    },
    {
      "epoch": 0.8384670773661893,
      "grad_norm": 0.7159976959228516,
      "learning_rate": 3.825423114792826e-05,
      "loss": 0.8066,
      "step": 655
    },
    {
      "epoch": 0.839747179774382,
      "grad_norm": 0.7225691080093384,
      "learning_rate": 3.8247818905233895e-05,
      "loss": 0.7569,
      "step": 656
    },
    {
      "epoch": 0.8410272821825746,
      "grad_norm": 0.7047813534736633,
      "learning_rate": 3.824139551040157e-05,
      "loss": 0.7929,
      "step": 657
    },
    {
      "epoch": 0.8423073845907673,
      "grad_norm": 0.6967841982841492,
      "learning_rate": 3.823496096784017e-05,
      "loss": 0.7872,
      "step": 658
    },
    {
      "epoch": 0.8435874869989599,
      "grad_norm": 0.7850909233093262,
      "learning_rate": 3.822851528196621e-05,
      "loss": 0.7926,
      "step": 659
    },
    {
      "epoch": 0.8448675894071526,
      "grad_norm": 0.7023453712463379,
      "learning_rate": 3.8222058457203875e-05,
      "loss": 0.7993,
      "step": 660
    },
    {
      "epoch": 0.8448675894071526,
      "eval_loss": 0.012208767235279083,
      "eval_runtime": 10.8086,
      "eval_samples_per_second": 45.889,
      "eval_steps_per_second": 5.736,
      "step": 660
    },
    {
      "epoch": 0.8461476918153452,
      "grad_norm": 0.7789046168327332,
      "learning_rate": 3.821559049798497e-05,
      "loss": 0.7373,
      "step": 661
    },
    {
      "epoch": 0.8474277942235379,
      "grad_norm": 0.7462376356124878,
      "learning_rate": 3.820911140874897e-05,
      "loss": 0.7695,
      "step": 662
    },
    {
      "epoch": 0.8487078966317305,
      "grad_norm": 0.7990427017211914,
      "learning_rate": 3.8202621193942976e-05,
      "loss": 0.7879,
      "step": 663
    },
    {
      "epoch": 0.8499879990399232,
      "grad_norm": 0.8021045923233032,
      "learning_rate": 3.819611985802172e-05,
      "loss": 0.7441,
      "step": 664
    },
    {
      "epoch": 0.8512681014481158,
      "grad_norm": 0.7579212188720703,
      "learning_rate": 3.818960740544757e-05,
      "loss": 0.7852,
      "step": 665
    },
    {
      "epoch": 0.8525482038563085,
      "grad_norm": 0.7370229959487915,
      "learning_rate": 3.818308384069054e-05,
      "loss": 0.7548,
      "step": 666
    },
    {
      "epoch": 0.8538283062645011,
      "grad_norm": 0.7963458299636841,
      "learning_rate": 3.817654916822825e-05,
      "loss": 0.754,
      "step": 667
    },
    {
      "epoch": 0.8551084086726938,
      "grad_norm": 0.7232759594917297,
      "learning_rate": 3.8170003392545944e-05,
      "loss": 0.7608,
      "step": 668
    },
    {
      "epoch": 0.8563885110808864,
      "grad_norm": 0.7259448766708374,
      "learning_rate": 3.816344651813651e-05,
      "loss": 0.786,
      "step": 669
    },
    {
      "epoch": 0.8576686134890791,
      "grad_norm": 0.7260249257087708,
      "learning_rate": 3.815687854950044e-05,
      "loss": 0.8142,
      "step": 670
    },
    {
      "epoch": 0.8589487158972717,
      "grad_norm": 0.6957441568374634,
      "learning_rate": 3.8150299491145824e-05,
      "loss": 0.7082,
      "step": 671
    },
    {
      "epoch": 0.8602288183054644,
      "grad_norm": 0.7644979953765869,
      "learning_rate": 3.8143709347588396e-05,
      "loss": 0.7566,
      "step": 672
    },
    {
      "epoch": 0.8615089207136571,
      "grad_norm": 0.6766453385353088,
      "learning_rate": 3.813710812335147e-05,
      "loss": 0.7714,
      "step": 673
    },
    {
      "epoch": 0.8627890231218498,
      "grad_norm": 0.7125437259674072,
      "learning_rate": 3.8130495822965987e-05,
      "loss": 0.7721,
      "step": 674
    },
    {
      "epoch": 0.8640691255300424,
      "grad_norm": 0.7135685682296753,
      "learning_rate": 3.812387245097048e-05,
      "loss": 0.7604,
      "step": 675
    },
    {
      "epoch": 0.8653492279382351,
      "grad_norm": 0.6936878561973572,
      "learning_rate": 3.811723801191108e-05,
      "loss": 0.7587,
      "step": 676
    },
    {
      "epoch": 0.8666293303464278,
      "grad_norm": 0.6780898571014404,
      "learning_rate": 3.811059251034152e-05,
      "loss": 0.8206,
      "step": 677
    },
    {
      "epoch": 0.8679094327546204,
      "grad_norm": 0.6613956093788147,
      "learning_rate": 3.8103935950823115e-05,
      "loss": 0.7532,
      "step": 678
    },
    {
      "epoch": 0.869189535162813,
      "grad_norm": 0.6378298997879028,
      "learning_rate": 3.809726833792479e-05,
      "loss": 0.7565,
      "step": 679
    },
    {
      "epoch": 0.8704696375710057,
      "grad_norm": 0.718760073184967,
      "learning_rate": 3.8090589676223035e-05,
      "loss": 0.7802,
      "step": 680
    },
    {
      "epoch": 0.8717497399791984,
      "grad_norm": 0.6480719447135925,
      "learning_rate": 3.8083899970301944e-05,
      "loss": 0.7628,
      "step": 681
    },
    {
      "epoch": 0.873029842387391,
      "grad_norm": 0.6628484129905701,
      "learning_rate": 3.807719922475317e-05,
      "loss": 0.7781,
      "step": 682
    },
    {
      "epoch": 0.8743099447955837,
      "grad_norm": 0.6903721690177917,
      "learning_rate": 3.8070487444175965e-05,
      "loss": 0.7525,
      "step": 683
    },
    {
      "epoch": 0.8755900472037763,
      "grad_norm": 0.6006175875663757,
      "learning_rate": 3.8063764633177126e-05,
      "loss": 0.7547,
      "step": 684
    },
    {
      "epoch": 0.876870149611969,
      "grad_norm": 0.701867938041687,
      "learning_rate": 3.805703079637106e-05,
      "loss": 0.7759,
      "step": 685
    },
    {
      "epoch": 0.8781502520201616,
      "grad_norm": 0.5799329876899719,
      "learning_rate": 3.8050285938379714e-05,
      "loss": 0.7503,
      "step": 686
    },
    {
      "epoch": 0.8794303544283543,
      "grad_norm": 0.5676674842834473,
      "learning_rate": 3.80435300638326e-05,
      "loss": 0.6808,
      "step": 687
    },
    {
      "epoch": 0.8807104568365469,
      "grad_norm": 0.7126132249832153,
      "learning_rate": 3.803676317736681e-05,
      "loss": 0.7371,
      "step": 688
    },
    {
      "epoch": 0.8819905592447396,
      "grad_norm": 0.6570621728897095,
      "learning_rate": 3.802998528362697e-05,
      "loss": 0.7344,
      "step": 689
    },
    {
      "epoch": 0.8832706616529322,
      "grad_norm": 0.6264975070953369,
      "learning_rate": 3.802319638726528e-05,
      "loss": 0.7435,
      "step": 690
    },
    {
      "epoch": 0.8832706616529322,
      "eval_loss": 0.012167807668447495,
      "eval_runtime": 10.8349,
      "eval_samples_per_second": 45.778,
      "eval_steps_per_second": 5.722,
      "step": 690
    },
    {
      "epoch": 0.8845507640611249,
      "grad_norm": 0.6723735332489014,
      "learning_rate": 3.8016396492941496e-05,
      "loss": 0.7388,
      "step": 691
    },
    {
      "epoch": 0.8858308664693175,
      "grad_norm": 0.6698789596557617,
      "learning_rate": 3.80095856053229e-05,
      "loss": 0.7449,
      "step": 692
    },
    {
      "epoch": 0.8871109688775102,
      "grad_norm": 0.66045081615448,
      "learning_rate": 3.8002763729084335e-05,
      "loss": 0.7579,
      "step": 693
    },
    {
      "epoch": 0.8883910712857028,
      "grad_norm": 0.7423791289329529,
      "learning_rate": 3.799593086890819e-05,
      "loss": 0.8204,
      "step": 694
    },
    {
      "epoch": 0.8896711736938955,
      "grad_norm": 0.6909918189048767,
      "learning_rate": 3.798908702948437e-05,
      "loss": 0.7848,
      "step": 695
    },
    {
      "epoch": 0.8909512761020881,
      "grad_norm": 0.7587515711784363,
      "learning_rate": 3.798223221551036e-05,
      "loss": 0.7894,
      "step": 696
    },
    {
      "epoch": 0.8922313785102808,
      "grad_norm": 0.634644091129303,
      "learning_rate": 3.7975366431691124e-05,
      "loss": 0.7697,
      "step": 697
    },
    {
      "epoch": 0.8935114809184734,
      "grad_norm": 0.704616904258728,
      "learning_rate": 3.7968489682739195e-05,
      "loss": 0.7816,
      "step": 698
    },
    {
      "epoch": 0.8947915833266661,
      "grad_norm": 0.6776732206344604,
      "learning_rate": 3.796160197337462e-05,
      "loss": 0.7722,
      "step": 699
    },
    {
      "epoch": 0.8960716857348587,
      "grad_norm": 0.7427555322647095,
      "learning_rate": 3.795470330832496e-05,
      "loss": 0.7988,
      "step": 700
    },
    {
      "epoch": 0.8973517881430515,
      "grad_norm": 0.6761599183082581,
      "learning_rate": 3.794779369232531e-05,
      "loss": 0.7748,
      "step": 701
    },
    {
      "epoch": 0.8986318905512442,
      "grad_norm": 0.7047790884971619,
      "learning_rate": 3.7940873130118275e-05,
      "loss": 0.7686,
      "step": 702
    },
    {
      "epoch": 0.8999119929594368,
      "grad_norm": 0.7226253151893616,
      "learning_rate": 3.7933941626453965e-05,
      "loss": 0.7851,
      "step": 703
    },
    {
      "epoch": 0.9011920953676295,
      "grad_norm": 0.657839298248291,
      "learning_rate": 3.792699918609002e-05,
      "loss": 0.7873,
      "step": 704
    },
    {
      "epoch": 0.9024721977758221,
      "grad_norm": 0.6704640984535217,
      "learning_rate": 3.792004581379157e-05,
      "loss": 0.7602,
      "step": 705
    },
    {
      "epoch": 0.9037523001840148,
      "grad_norm": 0.7041623592376709,
      "learning_rate": 3.7913081514331256e-05,
      "loss": 0.7737,
      "step": 706
    },
    {
      "epoch": 0.9050324025922074,
      "grad_norm": 0.7038724422454834,
      "learning_rate": 3.7906106292489214e-05,
      "loss": 0.7381,
      "step": 707
    },
    {
      "epoch": 0.9063125050004001,
      "grad_norm": 0.7266661524772644,
      "learning_rate": 3.789912015305308e-05,
      "loss": 0.7721,
      "step": 708
    },
    {
      "epoch": 0.9075926074085927,
      "grad_norm": 0.6591942310333252,
      "learning_rate": 3.789212310081798e-05,
      "loss": 0.779,
      "step": 709
    },
    {
      "epoch": 0.9088727098167854,
      "grad_norm": 0.7500007748603821,
      "learning_rate": 3.7885115140586546e-05,
      "loss": 0.7791,
      "step": 710
    },
    {
      "epoch": 0.910152812224978,
      "grad_norm": 0.673807680606842,
      "learning_rate": 3.787809627716887e-05,
      "loss": 0.7729,
      "step": 711
    },
    {
      "epoch": 0.9114329146331707,
      "grad_norm": 0.6827871799468994,
      "learning_rate": 3.787106651538254e-05,
      "loss": 0.799,
      "step": 712
    },
    {
      "epoch": 0.9127130170413633,
      "grad_norm": 0.6365512609481812,
      "learning_rate": 3.786402586005264e-05,
      "loss": 0.7772,
      "step": 713
    },
    {
      "epoch": 0.913993119449556,
      "grad_norm": 0.7164957523345947,
      "learning_rate": 3.785697431601172e-05,
      "loss": 0.7439,
      "step": 714
    },
    {
      "epoch": 0.9152732218577486,
      "grad_norm": 0.6707044243812561,
      "learning_rate": 3.784991188809979e-05,
      "loss": 0.7779,
      "step": 715
    },
    {
      "epoch": 0.9165533242659413,
      "grad_norm": 0.7647178769111633,
      "learning_rate": 3.784283858116434e-05,
      "loss": 0.8092,
      "step": 716
    },
    {
      "epoch": 0.9178334266741339,
      "grad_norm": 0.720170795917511,
      "learning_rate": 3.783575440006033e-05,
      "loss": 0.7153,
      "step": 717
    },
    {
      "epoch": 0.9191135290823266,
      "grad_norm": 0.6896149516105652,
      "learning_rate": 3.782865934965019e-05,
      "loss": 0.8097,
      "step": 718
    },
    {
      "epoch": 0.9203936314905192,
      "grad_norm": 0.6445088982582092,
      "learning_rate": 3.7821553434803806e-05,
      "loss": 0.7758,
      "step": 719
    },
    {
      "epoch": 0.9216737338987119,
      "grad_norm": 0.6135826706886292,
      "learning_rate": 3.7814436660398514e-05,
      "loss": 0.7878,
      "step": 720
    },
    {
      "epoch": 0.9216737338987119,
      "eval_loss": 0.012132863514125347,
      "eval_runtime": 10.8159,
      "eval_samples_per_second": 45.859,
      "eval_steps_per_second": 5.732,
      "step": 720
    },
    {
      "epoch": 0.9229538363069045,
      "grad_norm": 0.6798313856124878,
      "learning_rate": 3.78073090313191e-05,
      "loss": 0.8094,
      "step": 721
    },
    {
      "epoch": 0.9242339387150972,
      "grad_norm": 0.6502001881599426,
      "learning_rate": 3.780017055245782e-05,
      "loss": 0.8123,
      "step": 722
    },
    {
      "epoch": 0.9255140411232898,
      "grad_norm": 0.6290013194084167,
      "learning_rate": 3.7793021228714366e-05,
      "loss": 0.7683,
      "step": 723
    },
    {
      "epoch": 0.9267941435314825,
      "grad_norm": 0.6015080809593201,
      "learning_rate": 3.7785861064995854e-05,
      "loss": 0.7337,
      "step": 724
    },
    {
      "epoch": 0.9280742459396751,
      "grad_norm": 0.6684856414794922,
      "learning_rate": 3.7778690066216886e-05,
      "loss": 0.7629,
      "step": 725
    },
    {
      "epoch": 0.9293543483478678,
      "grad_norm": 0.6500918865203857,
      "learning_rate": 3.7771508237299456e-05,
      "loss": 0.786,
      "step": 726
    },
    {
      "epoch": 0.9306344507560604,
      "grad_norm": 0.6887004971504211,
      "learning_rate": 3.776431558317302e-05,
      "loss": 0.7584,
      "step": 727
    },
    {
      "epoch": 0.9319145531642531,
      "grad_norm": 0.629284143447876,
      "learning_rate": 3.775711210877445e-05,
      "loss": 0.7254,
      "step": 728
    },
    {
      "epoch": 0.9331946555724457,
      "grad_norm": 0.7328817248344421,
      "learning_rate": 3.7749897819048044e-05,
      "loss": 0.7683,
      "step": 729
    },
    {
      "epoch": 0.9344747579806385,
      "grad_norm": 0.6202633380889893,
      "learning_rate": 3.774267271894554e-05,
      "loss": 0.8149,
      "step": 730
    },
    {
      "epoch": 0.9357548603888312,
      "grad_norm": 0.6682559847831726,
      "learning_rate": 3.773543681342607e-05,
      "loss": 0.7787,
      "step": 731
    },
    {
      "epoch": 0.9370349627970238,
      "grad_norm": 0.5899592638015747,
      "learning_rate": 3.7728190107456214e-05,
      "loss": 0.7462,
      "step": 732
    },
    {
      "epoch": 0.9383150652052165,
      "grad_norm": 0.6884704232215881,
      "learning_rate": 3.772093260600993e-05,
      "loss": 0.7163,
      "step": 733
    },
    {
      "epoch": 0.9395951676134091,
      "grad_norm": 0.7006421089172363,
      "learning_rate": 3.771366431406863e-05,
      "loss": 0.802,
      "step": 734
    },
    {
      "epoch": 0.9408752700216018,
      "grad_norm": 0.8256950974464417,
      "learning_rate": 3.7706385236621074e-05,
      "loss": 0.7762,
      "step": 735
    },
    {
      "epoch": 0.9421553724297944,
      "grad_norm": 0.7171242833137512,
      "learning_rate": 3.7699095378663476e-05,
      "loss": 0.7976,
      "step": 736
    },
    {
      "epoch": 0.9434354748379871,
      "grad_norm": 0.7153730392456055,
      "learning_rate": 3.7691794745199435e-05,
      "loss": 0.7723,
      "step": 737
    },
    {
      "epoch": 0.9447155772461797,
      "grad_norm": 0.6709647178649902,
      "learning_rate": 3.768448334123993e-05,
      "loss": 0.7783,
      "step": 738
    },
    {
      "epoch": 0.9459956796543724,
      "grad_norm": 0.7212246060371399,
      "learning_rate": 3.7677161171803346e-05,
      "loss": 0.7928,
      "step": 739
    },
    {
      "epoch": 0.947275782062565,
      "grad_norm": 0.6821660995483398,
      "learning_rate": 3.766982824191547e-05,
      "loss": 0.767,
      "step": 740
    },
    {
      "epoch": 0.9485558844707577,
      "grad_norm": 0.6796236634254456,
      "learning_rate": 3.766248455660944e-05,
      "loss": 0.7837,
      "step": 741
    },
    {
      "epoch": 0.9498359868789503,
      "grad_norm": 0.6628891825675964,
      "learning_rate": 3.7655130120925805e-05,
      "loss": 0.8115,
      "step": 742
    },
    {
      "epoch": 0.951116089287143,
      "grad_norm": 0.6210861206054688,
      "learning_rate": 3.764776493991249e-05,
      "loss": 0.7428,
      "step": 743
    },
    {
      "epoch": 0.9523961916953356,
      "grad_norm": 0.6889374256134033,
      "learning_rate": 3.764038901862478e-05,
      "loss": 0.7808,
      "step": 744
    },
    {
      "epoch": 0.9536762941035283,
      "grad_norm": 0.627668023109436,
      "learning_rate": 3.763300236212535e-05,
      "loss": 0.7605,
      "step": 745
    },
    {
      "epoch": 0.9549563965117209,
      "grad_norm": 0.6655675172805786,
      "learning_rate": 3.762560497548424e-05,
      "loss": 0.7738,
      "step": 746
    },
    {
      "epoch": 0.9562364989199136,
      "grad_norm": 0.634579598903656,
      "learning_rate": 3.761819686377882e-05,
      "loss": 0.7781,
      "step": 747
    },
    {
      "epoch": 0.9575166013281062,
      "grad_norm": 0.6562496423721313,
      "learning_rate": 3.761077803209389e-05,
      "loss": 0.7877,
      "step": 748
    },
    {
      "epoch": 0.9587967037362989,
      "grad_norm": 0.5922711491584778,
      "learning_rate": 3.7603348485521546e-05,
      "loss": 0.7515,
      "step": 749
    },
    {
      "epoch": 0.9600768061444915,
      "grad_norm": 0.6168217062950134,
      "learning_rate": 3.759590822916128e-05,
      "loss": 0.7377,
      "step": 750
    },
    {
      "epoch": 0.9600768061444915,
      "eval_loss": 0.012109671719372272,
      "eval_runtime": 10.8247,
      "eval_samples_per_second": 45.821,
      "eval_steps_per_second": 5.728,
      "step": 750
    },
    {
      "epoch": 0.9613569085526842,
      "grad_norm": 0.6015171408653259,
      "learning_rate": 3.75884572681199e-05,
      "loss": 0.7448,
      "step": 751
    },
    {
      "epoch": 0.9626370109608768,
      "grad_norm": 0.6879626512527466,
      "learning_rate": 3.7580995607511584e-05,
      "loss": 0.7926,
      "step": 752
    },
    {
      "epoch": 0.9639171133690695,
      "grad_norm": 0.598674476146698,
      "learning_rate": 3.757352325245784e-05,
      "loss": 0.7109,
      "step": 753
    },
    {
      "epoch": 0.9651972157772621,
      "grad_norm": 0.7455387711524963,
      "learning_rate": 3.7566040208087544e-05,
      "loss": 0.7873,
      "step": 754
    },
    {
      "epoch": 0.9664773181854548,
      "grad_norm": 0.6520408987998962,
      "learning_rate": 3.755854647953687e-05,
      "loss": 0.7519,
      "step": 755
    },
    {
      "epoch": 0.9677574205936474,
      "grad_norm": 0.6713144779205322,
      "learning_rate": 3.755104207194936e-05,
      "loss": 0.7474,
      "step": 756
    },
    {
      "epoch": 0.9690375230018401,
      "grad_norm": 0.6542277932167053,
      "learning_rate": 3.7543526990475864e-05,
      "loss": 0.7349,
      "step": 757
    },
    {
      "epoch": 0.9703176254100329,
      "grad_norm": 0.734278678894043,
      "learning_rate": 3.753600124027456e-05,
      "loss": 0.793,
      "step": 758
    },
    {
      "epoch": 0.9715977278182255,
      "grad_norm": 0.640709638595581,
      "learning_rate": 3.7528464826510965e-05,
      "loss": 0.7578,
      "step": 759
    },
    {
      "epoch": 0.9728778302264182,
      "grad_norm": 0.6655123829841614,
      "learning_rate": 3.7520917754357895e-05,
      "loss": 0.7886,
      "step": 760
    },
    {
      "epoch": 0.9741579326346108,
      "grad_norm": 0.6819617748260498,
      "learning_rate": 3.751336002899549e-05,
      "loss": 0.7638,
      "step": 761
    },
    {
      "epoch": 0.9754380350428035,
      "grad_norm": 0.6399703025817871,
      "learning_rate": 3.750579165561121e-05,
      "loss": 0.7718,
      "step": 762
    },
    {
      "epoch": 0.9767181374509961,
      "grad_norm": 0.6448771357536316,
      "learning_rate": 3.7498212639399814e-05,
      "loss": 0.7403,
      "step": 763
    },
    {
      "epoch": 0.9779982398591888,
      "grad_norm": 0.6541329026222229,
      "learning_rate": 3.749062298556337e-05,
      "loss": 0.737,
      "step": 764
    },
    {
      "epoch": 0.9792783422673814,
      "grad_norm": 0.6602521538734436,
      "learning_rate": 3.748302269931124e-05,
      "loss": 0.7742,
      "step": 765
    },
    {
      "epoch": 0.9805584446755741,
      "grad_norm": 0.6031312346458435,
      "learning_rate": 3.74754117858601e-05,
      "loss": 0.7204,
      "step": 766
    },
    {
      "epoch": 0.9818385470837667,
      "grad_norm": 0.645257830619812,
      "learning_rate": 3.7467790250433903e-05,
      "loss": 0.7431,
      "step": 767
    },
    {
      "epoch": 0.9831186494919594,
      "grad_norm": 0.6873139142990112,
      "learning_rate": 3.7460158098263895e-05,
      "loss": 0.8201,
      "step": 768
    },
    {
      "epoch": 0.984398751900152,
      "grad_norm": 0.679110050201416,
      "learning_rate": 3.745251533458863e-05,
      "loss": 0.7408,
      "step": 769
    },
    {
      "epoch": 0.9856788543083447,
      "grad_norm": 0.6790152192115784,
      "learning_rate": 3.7444861964653905e-05,
      "loss": 0.7395,
      "step": 770
    },
    {
      "epoch": 0.9869589567165373,
      "grad_norm": 0.69802325963974,
      "learning_rate": 3.743719799371284e-05,
      "loss": 0.7828,
      "step": 771
    },
    {
      "epoch": 0.98823905912473,
      "grad_norm": 0.6704466342926025,
      "learning_rate": 3.7429523427025813e-05,
      "loss": 0.7735,
      "step": 772
    },
    {
      "epoch": 0.9895191615329226,
      "grad_norm": 0.6800136566162109,
      "learning_rate": 3.742183826986046e-05,
      "loss": 0.7531,
      "step": 773
    },
    {
      "epoch": 0.9907992639411153,
      "grad_norm": 0.649958074092865,
      "learning_rate": 3.741414252749171e-05,
      "loss": 0.7642,
      "step": 774
    },
    {
      "epoch": 0.9920793663493079,
      "grad_norm": 0.6794930100440979,
      "learning_rate": 3.7406436205201755e-05,
      "loss": 0.7957,
      "step": 775
    },
    {
      "epoch": 0.9933594687575006,
      "grad_norm": 0.6961868405342102,
      "learning_rate": 3.739871930828002e-05,
      "loss": 0.7478,
      "step": 776
    },
    {
      "epoch": 0.9946395711656932,
      "grad_norm": 0.7181962132453918,
      "learning_rate": 3.739099184202324e-05,
      "loss": 0.7827,
      "step": 777
    },
    {
      "epoch": 0.9959196735738859,
      "grad_norm": 0.7640824913978577,
      "learning_rate": 3.7383253811735346e-05,
      "loss": 0.7335,
      "step": 778
    },
    {
      "epoch": 0.9971997759820785,
      "grad_norm": 0.624038577079773,
      "learning_rate": 3.737550522272756e-05,
      "loss": 0.7184,
      "step": 779
    },
    {
      "epoch": 0.9984798783902712,
      "grad_norm": 0.7637789249420166,
      "learning_rate": 3.736774608031834e-05,
      "loss": 0.7586,
      "step": 780
    },
    {
      "epoch": 0.9984798783902712,
      "eval_loss": 0.012082226574420929,
      "eval_runtime": 10.8094,
      "eval_samples_per_second": 45.886,
      "eval_steps_per_second": 5.736,
      "step": 780
    },
    {
      "epoch": 0.9997599807984638,
      "grad_norm": 0.6540350317955017,
      "learning_rate": 3.7359976389833384e-05,
      "loss": 0.7477,
      "step": 781
    },
    {
      "epoch": 1.0,
      "grad_norm": 1.015478253364563,
      "learning_rate": 3.735219615660564e-05,
      "loss": 0.7103,
      "step": 782
    },
    {
      "epoch": 1.0012801024081928,
      "grad_norm": 0.9592230916023254,
      "learning_rate": 3.734440538597528e-05,
      "loss": 0.7656,
      "step": 783
    },
    {
      "epoch": 1.0025602048163853,
      "grad_norm": 0.7201738357543945,
      "learning_rate": 3.733660408328971e-05,
      "loss": 0.7175,
      "step": 784
    },
    {
      "epoch": 1.003840307224578,
      "grad_norm": 0.747381865978241,
      "learning_rate": 3.732879225390357e-05,
      "loss": 0.6981,
      "step": 785
    },
    {
      "epoch": 1.0051204096327706,
      "grad_norm": 0.7670899629592896,
      "learning_rate": 3.7320969903178735e-05,
      "loss": 0.74,
      "step": 786
    },
    {
      "epoch": 1.0064005120409634,
      "grad_norm": 0.7124491333961487,
      "learning_rate": 3.731313703648427e-05,
      "loss": 0.7294,
      "step": 787
    },
    {
      "epoch": 1.007680614449156,
      "grad_norm": 0.677145779132843,
      "learning_rate": 3.730529365919651e-05,
      "loss": 0.7084,
      "step": 788
    },
    {
      "epoch": 1.0089607168573487,
      "grad_norm": 0.7383764982223511,
      "learning_rate": 3.729743977669895e-05,
      "loss": 0.7607,
      "step": 789
    },
    {
      "epoch": 1.0102408192655412,
      "grad_norm": 0.658332347869873,
      "learning_rate": 3.728957539438231e-05,
      "loss": 0.7378,
      "step": 790
    },
    {
      "epoch": 1.011520921673734,
      "grad_norm": 0.6816548109054565,
      "learning_rate": 3.7281700517644554e-05,
      "loss": 0.7041,
      "step": 791
    },
    {
      "epoch": 1.0128010240819265,
      "grad_norm": 0.6481317281723022,
      "learning_rate": 3.72738151518908e-05,
      "loss": 0.7524,
      "step": 792
    },
    {
      "epoch": 1.0140811264901193,
      "grad_norm": 0.7398667335510254,
      "learning_rate": 3.726591930253339e-05,
      "loss": 0.7284,
      "step": 793
    },
    {
      "epoch": 1.0153612288983118,
      "grad_norm": 0.6366782188415527,
      "learning_rate": 3.7258012974991864e-05,
      "loss": 0.734,
      "step": 794
    },
    {
      "epoch": 1.0166413313065046,
      "grad_norm": 0.7185632586479187,
      "learning_rate": 3.7250096174692936e-05,
      "loss": 0.7509,
      "step": 795
    },
    {
      "epoch": 1.0179214337146971,
      "grad_norm": 0.6839074492454529,
      "learning_rate": 3.7242168907070534e-05,
      "loss": 0.7403,
      "step": 796
    },
    {
      "epoch": 1.0192015361228899,
      "grad_norm": 0.6224228739738464,
      "learning_rate": 3.723423117756574e-05,
      "loss": 0.7084,
      "step": 797
    },
    {
      "epoch": 1.0204816385310824,
      "grad_norm": 0.7044923305511475,
      "learning_rate": 3.722628299162684e-05,
      "loss": 0.7213,
      "step": 798
    },
    {
      "epoch": 1.0217617409392752,
      "grad_norm": 0.6783655285835266,
      "learning_rate": 3.7218324354709305e-05,
      "loss": 0.7465,
      "step": 799
    },
    {
      "epoch": 1.0230418433474677,
      "grad_norm": 0.6528795957565308,
      "learning_rate": 3.721035527227575e-05,
      "loss": 0.7334,
      "step": 800
    },
    {
      "epoch": 1.0243219457556605,
      "grad_norm": 0.6916537880897522,
      "learning_rate": 3.720237574979597e-05,
      "loss": 0.7314,
      "step": 801
    },
    {
      "epoch": 1.025602048163853,
      "grad_norm": 0.6323184370994568,
      "learning_rate": 3.7194385792746934e-05,
      "loss": 0.7349,
      "step": 802
    },
    {
      "epoch": 1.0268821505720458,
      "grad_norm": 0.6384755373001099,
      "learning_rate": 3.718638540661278e-05,
      "loss": 0.7202,
      "step": 803
    },
    {
      "epoch": 1.0281622529802383,
      "grad_norm": 0.698417067527771,
      "learning_rate": 3.7178374596884784e-05,
      "loss": 0.7491,
      "step": 804
    },
    {
      "epoch": 1.029442355388431,
      "grad_norm": 0.6580883860588074,
      "learning_rate": 3.717035336906138e-05,
      "loss": 0.7492,
      "step": 805
    },
    {
      "epoch": 1.0307224577966236,
      "grad_norm": 0.6361373662948608,
      "learning_rate": 3.716232172864817e-05,
      "loss": 0.7284,
      "step": 806
    },
    {
      "epoch": 1.0320025602048164,
      "grad_norm": 0.6777246594429016,
      "learning_rate": 3.715427968115789e-05,
      "loss": 0.71,
      "step": 807
    },
    {
      "epoch": 1.033282662613009,
      "grad_norm": 0.7357933521270752,
      "learning_rate": 3.714622723211041e-05,
      "loss": 0.7579,
      "step": 808
    },
    {
      "epoch": 1.0345627650212017,
      "grad_norm": 0.6447516083717346,
      "learning_rate": 3.713816438703276e-05,
      "loss": 0.7091,
      "step": 809
    },
    {
      "epoch": 1.0358428674293942,
      "grad_norm": 0.6505489945411682,
      "learning_rate": 3.7130091151459094e-05,
      "loss": 0.7135,
      "step": 810
    },
    {
      "epoch": 1.0358428674293942,
      "eval_loss": 0.012068789452314377,
      "eval_runtime": 10.8218,
      "eval_samples_per_second": 45.833,
      "eval_steps_per_second": 5.729,
      "step": 810
    },
    {
      "epoch": 1.037122969837587,
      "grad_norm": 0.6926580667495728,
      "learning_rate": 3.71220075309307e-05,
      "loss": 0.7419,
      "step": 811
    },
    {
      "epoch": 1.0384030722457798,
      "grad_norm": 0.6922308206558228,
      "learning_rate": 3.711391353099599e-05,
      "loss": 0.7549,
      "step": 812
    },
    {
      "epoch": 1.0396831746539723,
      "grad_norm": 0.7106603384017944,
      "learning_rate": 3.710580915721051e-05,
      "loss": 0.7637,
      "step": 813
    },
    {
      "epoch": 1.040963277062165,
      "grad_norm": 0.6899227499961853,
      "learning_rate": 3.709769441513691e-05,
      "loss": 0.7337,
      "step": 814
    },
    {
      "epoch": 1.0422433794703576,
      "grad_norm": 0.6479454636573792,
      "learning_rate": 3.708956931034498e-05,
      "loss": 0.6827,
      "step": 815
    },
    {
      "epoch": 1.0435234818785504,
      "grad_norm": 0.6753072142601013,
      "learning_rate": 3.7081433848411596e-05,
      "loss": 0.7404,
      "step": 816
    },
    {
      "epoch": 1.044803584286743,
      "grad_norm": 0.6539455056190491,
      "learning_rate": 3.707328803492077e-05,
      "loss": 0.6972,
      "step": 817
    },
    {
      "epoch": 1.0460836866949357,
      "grad_norm": 0.6149129867553711,
      "learning_rate": 3.70651318754636e-05,
      "loss": 0.7423,
      "step": 818
    },
    {
      "epoch": 1.0473637891031282,
      "grad_norm": 0.6942073106765747,
      "learning_rate": 3.705696537563828e-05,
      "loss": 0.7324,
      "step": 819
    },
    {
      "epoch": 1.048643891511321,
      "grad_norm": 0.6388072371482849,
      "learning_rate": 3.7048788541050126e-05,
      "loss": 0.7452,
      "step": 820
    },
    {
      "epoch": 1.0499239939195135,
      "grad_norm": 0.6410319209098816,
      "learning_rate": 3.7040601377311526e-05,
      "loss": 0.715,
      "step": 821
    },
    {
      "epoch": 1.0512040963277063,
      "grad_norm": 0.6240140795707703,
      "learning_rate": 3.703240389004197e-05,
      "loss": 0.7068,
      "step": 822
    },
    {
      "epoch": 1.0524841987358988,
      "grad_norm": 0.7288002967834473,
      "learning_rate": 3.702419608486803e-05,
      "loss": 0.7451,
      "step": 823
    },
    {
      "epoch": 1.0537643011440916,
      "grad_norm": 0.6330709457397461,
      "learning_rate": 3.7015977967423356e-05,
      "loss": 0.7175,
      "step": 824
    },
    {
      "epoch": 1.0550444035522841,
      "grad_norm": 0.7527199387550354,
      "learning_rate": 3.700774954334868e-05,
      "loss": 0.725,
      "step": 825
    },
    {
      "epoch": 1.056324505960477,
      "grad_norm": 0.5856966376304626,
      "learning_rate": 3.69995108182918e-05,
      "loss": 0.7131,
      "step": 826
    },
    {
      "epoch": 1.0576046083686694,
      "grad_norm": 0.7010636925697327,
      "learning_rate": 3.6991261797907604e-05,
      "loss": 0.7244,
      "step": 827
    },
    {
      "epoch": 1.0588847107768622,
      "grad_norm": 0.6359009742736816,
      "learning_rate": 3.698300248785803e-05,
      "loss": 0.7058,
      "step": 828
    },
    {
      "epoch": 1.0601648131850547,
      "grad_norm": 0.8455493450164795,
      "learning_rate": 3.6974732893812074e-05,
      "loss": 0.7228,
      "step": 829
    },
    {
      "epoch": 1.0614449155932475,
      "grad_norm": 0.6704761385917664,
      "learning_rate": 3.696645302144582e-05,
      "loss": 0.7454,
      "step": 830
    },
    {
      "epoch": 1.06272501800144,
      "grad_norm": 0.8195145726203918,
      "learning_rate": 3.695816287644236e-05,
      "loss": 0.7227,
      "step": 831
    },
    {
      "epoch": 1.0640051204096328,
      "grad_norm": 0.7179127931594849,
      "learning_rate": 3.694986246449189e-05,
      "loss": 0.7338,
      "step": 832
    },
    {
      "epoch": 1.0652852228178253,
      "grad_norm": 0.7122701406478882,
      "learning_rate": 3.6941551791291604e-05,
      "loss": 0.7565,
      "step": 833
    },
    {
      "epoch": 1.066565325226018,
      "grad_norm": 0.720400333404541,
      "learning_rate": 3.693323086254578e-05,
      "loss": 0.7649,
      "step": 834
    },
    {
      "epoch": 1.0678454276342106,
      "grad_norm": 0.7331603169441223,
      "learning_rate": 3.69248996839657e-05,
      "loss": 0.7458,
      "step": 835
    },
    {
      "epoch": 1.0691255300424034,
      "grad_norm": 0.8550188541412354,
      "learning_rate": 3.691655826126971e-05,
      "loss": 0.7268,
      "step": 836
    },
    {
      "epoch": 1.070405632450596,
      "grad_norm": 0.6874879002571106,
      "learning_rate": 3.690820660018317e-05,
      "loss": 0.7141,
      "step": 837
    },
    {
      "epoch": 1.0716857348587887,
      "grad_norm": 0.7743686437606812,
      "learning_rate": 3.689984470643847e-05,
      "loss": 0.7338,
      "step": 838
    },
    {
      "epoch": 1.0729658372669815,
      "grad_norm": 0.7543725371360779,
      "learning_rate": 3.6891472585775034e-05,
      "loss": 0.7459,
      "step": 839
    },
    {
      "epoch": 1.074245939675174,
      "grad_norm": 0.6992446184158325,
      "learning_rate": 3.688309024393929e-05,
      "loss": 0.7203,
      "step": 840
    },
    {
      "epoch": 1.074245939675174,
      "eval_loss": 0.012044671922922134,
      "eval_runtime": 10.8226,
      "eval_samples_per_second": 45.83,
      "eval_steps_per_second": 5.729,
      "step": 840
    },
    {
      "epoch": 1.0755260420833668,
      "grad_norm": 0.7683684825897217,
      "learning_rate": 3.687469768668469e-05,
      "loss": 0.731,
      "step": 841
    },
    {
      "epoch": 1.0768061444915593,
      "grad_norm": 0.7468374967575073,
      "learning_rate": 3.6866294919771705e-05,
      "loss": 0.7385,
      "step": 842
    },
    {
      "epoch": 1.078086246899752,
      "grad_norm": 0.6611967086791992,
      "learning_rate": 3.685788194896779e-05,
      "loss": 0.7384,
      "step": 843
    },
    {
      "epoch": 1.0793663493079446,
      "grad_norm": 0.7011679410934448,
      "learning_rate": 3.684945878004744e-05,
      "loss": 0.7376,
      "step": 844
    },
    {
      "epoch": 1.0806464517161374,
      "grad_norm": 0.6693490743637085,
      "learning_rate": 3.6841025418792114e-05,
      "loss": 0.7127,
      "step": 845
    },
    {
      "epoch": 1.08192655412433,
      "grad_norm": 0.6240278482437134,
      "learning_rate": 3.683258187099028e-05,
      "loss": 0.7288,
      "step": 846
    },
    {
      "epoch": 1.0832066565325227,
      "grad_norm": 0.7663692235946655,
      "learning_rate": 3.682412814243741e-05,
      "loss": 0.7449,
      "step": 847
    },
    {
      "epoch": 1.0844867589407152,
      "grad_norm": 0.6547810435295105,
      "learning_rate": 3.681566423893594e-05,
      "loss": 0.7392,
      "step": 848
    },
    {
      "epoch": 1.085766861348908,
      "grad_norm": 0.6520044803619385,
      "learning_rate": 3.6807190166295324e-05,
      "loss": 0.7451,
      "step": 849
    },
    {
      "epoch": 1.0870469637571005,
      "grad_norm": 0.6523797512054443,
      "learning_rate": 3.6798705930331954e-05,
      "loss": 0.7039,
      "step": 850
    },
    {
      "epoch": 1.0883270661652933,
      "grad_norm": 0.6763718128204346,
      "learning_rate": 3.679021153686923e-05,
      "loss": 0.7569,
      "step": 851
    },
    {
      "epoch": 1.0896071685734858,
      "grad_norm": 0.5969076156616211,
      "learning_rate": 3.678170699173751e-05,
      "loss": 0.7088,
      "step": 852
    },
    {
      "epoch": 1.0908872709816786,
      "grad_norm": 0.7176002860069275,
      "learning_rate": 3.677319230077412e-05,
      "loss": 0.7037,
      "step": 853
    },
    {
      "epoch": 1.0921673733898711,
      "grad_norm": 0.6345864534378052,
      "learning_rate": 3.676466746982336e-05,
      "loss": 0.749,
      "step": 854
    },
    {
      "epoch": 1.093447475798064,
      "grad_norm": 0.7272902131080627,
      "learning_rate": 3.675613250473648e-05,
      "loss": 0.7444,
      "step": 855
    },
    {
      "epoch": 1.0947275782062564,
      "grad_norm": 0.6915029287338257,
      "learning_rate": 3.674758741137168e-05,
      "loss": 0.7381,
      "step": 856
    },
    {
      "epoch": 1.0960076806144492,
      "grad_norm": 0.7332066893577576,
      "learning_rate": 3.6739032195594134e-05,
      "loss": 0.7536,
      "step": 857
    },
    {
      "epoch": 1.0972877830226417,
      "grad_norm": 0.5940446853637695,
      "learning_rate": 3.673046686327594e-05,
      "loss": 0.7253,
      "step": 858
    },
    {
      "epoch": 1.0985678854308345,
      "grad_norm": 0.7006335258483887,
      "learning_rate": 3.672189142029614e-05,
      "loss": 0.6761,
      "step": 859
    },
    {
      "epoch": 1.099847987839027,
      "grad_norm": 0.6557371020317078,
      "learning_rate": 3.671330587254076e-05,
      "loss": 0.7012,
      "step": 860
    },
    {
      "epoch": 1.1011280902472198,
      "grad_norm": 0.6992865800857544,
      "learning_rate": 3.670471022590269e-05,
      "loss": 0.7275,
      "step": 861
    },
    {
      "epoch": 1.1024081926554123,
      "grad_norm": 0.6820812225341797,
      "learning_rate": 3.669610448628181e-05,
      "loss": 0.7375,
      "step": 862
    },
    {
      "epoch": 1.103688295063605,
      "grad_norm": 0.7431970834732056,
      "learning_rate": 3.66874886595849e-05,
      "loss": 0.7301,
      "step": 863
    },
    {
      "epoch": 1.1049683974717976,
      "grad_norm": 0.6357002258300781,
      "learning_rate": 3.667886275172567e-05,
      "loss": 0.7378,
      "step": 864
    },
    {
      "epoch": 1.1062484998799904,
      "grad_norm": 0.7715697884559631,
      "learning_rate": 3.6670226768624745e-05,
      "loss": 0.7484,
      "step": 865
    },
    {
      "epoch": 1.107528602288183,
      "grad_norm": 0.621955394744873,
      "learning_rate": 3.6661580716209684e-05,
      "loss": 0.7159,
      "step": 866
    },
    {
      "epoch": 1.1088087046963757,
      "grad_norm": 0.8410272598266602,
      "learning_rate": 3.665292460041492e-05,
      "loss": 0.7438,
      "step": 867
    },
    {
      "epoch": 1.1100888071045683,
      "grad_norm": 0.7019923329353333,
      "learning_rate": 3.6644258427181845e-05,
      "loss": 0.7053,
      "step": 868
    },
    {
      "epoch": 1.111368909512761,
      "grad_norm": 0.7673763036727905,
      "learning_rate": 3.66355822024587e-05,
      "loss": 0.7165,
      "step": 869
    },
    {
      "epoch": 1.1126490119209538,
      "grad_norm": 0.794391393661499,
      "learning_rate": 3.662689593220066e-05,
      "loss": 0.7703,
      "step": 870
    },
    {
      "epoch": 1.1126490119209538,
      "eval_loss": 0.012025858275592327,
      "eval_runtime": 10.8162,
      "eval_samples_per_second": 45.857,
      "eval_steps_per_second": 5.732,
      "step": 870
    },
    {
      "epoch": 1.1139291143291463,
      "grad_norm": 0.652980387210846,
      "learning_rate": 3.66181996223698e-05,
      "loss": 0.7104,
      "step": 871
    },
    {
      "epoch": 1.115209216737339,
      "grad_norm": 0.7875044941902161,
      "learning_rate": 3.660949327893505e-05,
      "loss": 0.6948,
      "step": 872
    },
    {
      "epoch": 1.1164893191455316,
      "grad_norm": 0.7187885642051697,
      "learning_rate": 3.660077690787226e-05,
      "loss": 0.7869,
      "step": 873
    },
    {
      "epoch": 1.1177694215537244,
      "grad_norm": 0.6242810487747192,
      "learning_rate": 3.659205051516414e-05,
      "loss": 0.6933,
      "step": 874
    },
    {
      "epoch": 1.119049523961917,
      "grad_norm": 0.697650134563446,
      "learning_rate": 3.658331410680031e-05,
      "loss": 0.7347,
      "step": 875
    },
    {
      "epoch": 1.1203296263701097,
      "grad_norm": 0.6302863955497742,
      "learning_rate": 3.657456768877723e-05,
      "loss": 0.7231,
      "step": 876
    },
    {
      "epoch": 1.1216097287783022,
      "grad_norm": 0.6891476511955261,
      "learning_rate": 3.6565811267098245e-05,
      "loss": 0.7689,
      "step": 877
    },
    {
      "epoch": 1.122889831186495,
      "grad_norm": 0.5999673008918762,
      "learning_rate": 3.6557044847773575e-05,
      "loss": 0.723,
      "step": 878
    },
    {
      "epoch": 1.1241699335946875,
      "grad_norm": 0.6499205231666565,
      "learning_rate": 3.654826843682028e-05,
      "loss": 0.7098,
      "step": 879
    },
    {
      "epoch": 1.1254500360028803,
      "grad_norm": 0.642432689666748,
      "learning_rate": 3.6539482040262304e-05,
      "loss": 0.7362,
      "step": 880
    },
    {
      "epoch": 1.1267301384110728,
      "grad_norm": 0.6041407585144043,
      "learning_rate": 3.653068566413043e-05,
      "loss": 0.7276,
      "step": 881
    },
    {
      "epoch": 1.1280102408192656,
      "grad_norm": 0.6853089928627014,
      "learning_rate": 3.6521879314462285e-05,
      "loss": 0.734,
      "step": 882
    },
    {
      "epoch": 1.1292903432274581,
      "grad_norm": 0.6265656352043152,
      "learning_rate": 3.6513062997302356e-05,
      "loss": 0.7148,
      "step": 883
    },
    {
      "epoch": 1.130570445635651,
      "grad_norm": 0.6284116506576538,
      "learning_rate": 3.650423671870197e-05,
      "loss": 0.7134,
      "step": 884
    },
    {
      "epoch": 1.1318505480438434,
      "grad_norm": 0.6319745779037476,
      "learning_rate": 3.649540048471927e-05,
      "loss": 0.7082,
      "step": 885
    },
    {
      "epoch": 1.1331306504520362,
      "grad_norm": 0.6498191356658936,
      "learning_rate": 3.648655430141927e-05,
      "loss": 0.747,
      "step": 886
    },
    {
      "epoch": 1.1344107528602287,
      "grad_norm": 0.6267776489257812,
      "learning_rate": 3.647769817487377e-05,
      "loss": 0.7125,
      "step": 887
    },
    {
      "epoch": 1.1356908552684215,
      "grad_norm": 0.6035043001174927,
      "learning_rate": 3.646883211116143e-05,
      "loss": 0.7423,
      "step": 888
    },
    {
      "epoch": 1.136970957676614,
      "grad_norm": 0.6620738506317139,
      "learning_rate": 3.645995611636772e-05,
      "loss": 0.7157,
      "step": 889
    },
    {
      "epoch": 1.1382510600848068,
      "grad_norm": 0.6425696611404419,
      "learning_rate": 3.645107019658491e-05,
      "loss": 0.718,
      "step": 890
    },
    {
      "epoch": 1.1395311624929993,
      "grad_norm": 0.5747361183166504,
      "learning_rate": 3.64421743579121e-05,
      "loss": 0.7198,
      "step": 891
    },
    {
      "epoch": 1.1408112649011921,
      "grad_norm": 0.5982975363731384,
      "learning_rate": 3.643326860645521e-05,
      "loss": 0.7085,
      "step": 892
    },
    {
      "epoch": 1.1420913673093849,
      "grad_norm": 0.6179444789886475,
      "learning_rate": 3.642435294832692e-05,
      "loss": 0.7244,
      "step": 893
    },
    {
      "epoch": 1.1433714697175774,
      "grad_norm": 0.6015804409980774,
      "learning_rate": 3.641542738964676e-05,
      "loss": 0.7372,
      "step": 894
    },
    {
      "epoch": 1.14465157212577,
      "grad_norm": 0.6817703247070312,
      "learning_rate": 3.640649193654102e-05,
      "loss": 0.7725,
      "step": 895
    },
    {
      "epoch": 1.1459316745339627,
      "grad_norm": 0.6058518290519714,
      "learning_rate": 3.639754659514281e-05,
      "loss": 0.7433,
      "step": 896
    },
    {
      "epoch": 1.1472117769421555,
      "grad_norm": 0.7060476541519165,
      "learning_rate": 3.638859137159199e-05,
      "loss": 0.7336,
      "step": 897
    },
    {
      "epoch": 1.148491879350348,
      "grad_norm": 0.6474248170852661,
      "learning_rate": 3.6379626272035233e-05,
      "loss": 0.7502,
      "step": 898
    },
    {
      "epoch": 1.1497719817585406,
      "grad_norm": 0.7358543872833252,
      "learning_rate": 3.637065130262599e-05,
      "loss": 0.7155,
      "step": 899
    },
    {
      "epoch": 1.1510520841667333,
      "grad_norm": 0.7459300756454468,
      "learning_rate": 3.636166646952448e-05,
      "loss": 0.7739,
      "step": 900
    },
    {
      "epoch": 1.1510520841667333,
      "eval_loss": 0.01201329194009304,
      "eval_runtime": 10.8486,
      "eval_samples_per_second": 45.72,
      "eval_steps_per_second": 5.715,
      "step": 900
    },
    {
      "epoch": 1.152332186574926,
      "grad_norm": 0.6569747924804688,
      "learning_rate": 3.635267177889767e-05,
      "loss": 0.7462,
      "step": 901
    },
    {
      "epoch": 1.1536122889831186,
      "grad_norm": 0.6367251873016357,
      "learning_rate": 3.6343667236919336e-05,
      "loss": 0.7315,
      "step": 902
    },
    {
      "epoch": 1.1548923913913114,
      "grad_norm": 0.6607640981674194,
      "learning_rate": 3.633465284976998e-05,
      "loss": 0.7316,
      "step": 903
    },
    {
      "epoch": 1.156172493799504,
      "grad_norm": 0.6306076645851135,
      "learning_rate": 3.632562862363688e-05,
      "loss": 0.7328,
      "step": 904
    },
    {
      "epoch": 1.1574525962076967,
      "grad_norm": 0.6793370842933655,
      "learning_rate": 3.6316594564714055e-05,
      "loss": 0.7533,
      "step": 905
    },
    {
      "epoch": 1.1587326986158892,
      "grad_norm": 0.7186464667320251,
      "learning_rate": 3.630755067920229e-05,
      "loss": 0.735,
      "step": 906
    },
    {
      "epoch": 1.160012801024082,
      "grad_norm": 0.6464977264404297,
      "learning_rate": 3.629849697330911e-05,
      "loss": 0.7214,
      "step": 907
    },
    {
      "epoch": 1.1612929034322745,
      "grad_norm": 0.6474738121032715,
      "learning_rate": 3.6289433453248755e-05,
      "loss": 0.7394,
      "step": 908
    },
    {
      "epoch": 1.1625730058404673,
      "grad_norm": 0.7385799884796143,
      "learning_rate": 3.6280360125242234e-05,
      "loss": 0.7385,
      "step": 909
    },
    {
      "epoch": 1.1638531082486598,
      "grad_norm": 0.6468353271484375,
      "learning_rate": 3.6271276995517275e-05,
      "loss": 0.7333,
      "step": 910
    },
    {
      "epoch": 1.1651332106568526,
      "grad_norm": 0.6115081310272217,
      "learning_rate": 3.626218407030834e-05,
      "loss": 0.7124,
      "step": 911
    },
    {
      "epoch": 1.1664133130650451,
      "grad_norm": 0.6233425140380859,
      "learning_rate": 3.62530813558566e-05,
      "loss": 0.7274,
      "step": 912
    },
    {
      "epoch": 1.167693415473238,
      "grad_norm": 0.6084201335906982,
      "learning_rate": 3.6243968858409955e-05,
      "loss": 0.6875,
      "step": 913
    },
    {
      "epoch": 1.1689735178814304,
      "grad_norm": 0.6146586537361145,
      "learning_rate": 3.623484658422302e-05,
      "loss": 0.746,
      "step": 914
    },
    {
      "epoch": 1.1702536202896232,
      "grad_norm": 0.6179696321487427,
      "learning_rate": 3.622571453955712e-05,
      "loss": 0.715,
      "step": 915
    },
    {
      "epoch": 1.1715337226978157,
      "grad_norm": 0.6410030722618103,
      "learning_rate": 3.6216572730680285e-05,
      "loss": 0.7224,
      "step": 916
    },
    {
      "epoch": 1.1728138251060085,
      "grad_norm": 0.706678032875061,
      "learning_rate": 3.620742116386724e-05,
      "loss": 0.7428,
      "step": 917
    },
    {
      "epoch": 1.174093927514201,
      "grad_norm": 0.5854217410087585,
      "learning_rate": 3.6198259845399424e-05,
      "loss": 0.7092,
      "step": 918
    },
    {
      "epoch": 1.1753740299223938,
      "grad_norm": 0.712439775466919,
      "learning_rate": 3.618908878156496e-05,
      "loss": 0.7472,
      "step": 919
    },
    {
      "epoch": 1.1766541323305864,
      "grad_norm": 0.6230387687683105,
      "learning_rate": 3.617990797865866e-05,
      "loss": 0.7331,
      "step": 920
    },
    {
      "epoch": 1.1779342347387791,
      "grad_norm": 0.6859497427940369,
      "learning_rate": 3.617071744298201e-05,
      "loss": 0.741,
      "step": 921
    },
    {
      "epoch": 1.1792143371469717,
      "grad_norm": 0.6405500173568726,
      "learning_rate": 3.6161517180843204e-05,
      "loss": 0.755,
      "step": 922
    },
    {
      "epoch": 1.1804944395551644,
      "grad_norm": 0.6109355688095093,
      "learning_rate": 3.615230719855709e-05,
      "loss": 0.7578,
      "step": 923
    },
    {
      "epoch": 1.1817745419633572,
      "grad_norm": 0.6397739052772522,
      "learning_rate": 3.614308750244518e-05,
      "loss": 0.7348,
      "step": 924
    },
    {
      "epoch": 1.1830546443715497,
      "grad_norm": 0.596720814704895,
      "learning_rate": 3.613385809883569e-05,
      "loss": 0.7053,
      "step": 925
    },
    {
      "epoch": 1.1843347467797423,
      "grad_norm": 0.6623912453651428,
      "learning_rate": 3.6124618994063446e-05,
      "loss": 0.739,
      "step": 926
    },
    {
      "epoch": 1.185614849187935,
      "grad_norm": 0.6404293775558472,
      "learning_rate": 3.611537019446999e-05,
      "loss": 0.7206,
      "step": 927
    },
    {
      "epoch": 1.1868949515961278,
      "grad_norm": 0.6415302157402039,
      "learning_rate": 3.610611170640347e-05,
      "loss": 0.7599,
      "step": 928
    },
    {
      "epoch": 1.1881750540043203,
      "grad_norm": 0.6325163245201111,
      "learning_rate": 3.6096843536218714e-05,
      "loss": 0.7226,
      "step": 929
    },
    {
      "epoch": 1.189455156412513,
      "grad_norm": 0.6508901715278625,
      "learning_rate": 3.608756569027719e-05,
      "loss": 0.7508,
      "step": 930
    },
    {
      "epoch": 1.189455156412513,
      "eval_loss": 0.011997492983937263,
      "eval_runtime": 10.831,
      "eval_samples_per_second": 45.794,
      "eval_steps_per_second": 5.724,
      "step": 930
    },
    {
      "epoch": 1.1907352588207056,
      "grad_norm": 0.653530478477478,
      "learning_rate": 3.607827817494699e-05,
      "loss": 0.7326,
      "step": 931
    },
    {
      "epoch": 1.1920153612288984,
      "grad_norm": 0.6009781360626221,
      "learning_rate": 3.6068980996602874e-05,
      "loss": 0.7414,
      "step": 932
    },
    {
      "epoch": 1.193295463637091,
      "grad_norm": 0.6281853318214417,
      "learning_rate": 3.605967416162619e-05,
      "loss": 0.7062,
      "step": 933
    },
    {
      "epoch": 1.1945755660452837,
      "grad_norm": 0.6202437877655029,
      "learning_rate": 3.605035767640497e-05,
      "loss": 0.6771,
      "step": 934
    },
    {
      "epoch": 1.1958556684534762,
      "grad_norm": 0.6261670589447021,
      "learning_rate": 3.604103154733382e-05,
      "loss": 0.7125,
      "step": 935
    },
    {
      "epoch": 1.197135770861669,
      "grad_norm": 0.6848472356796265,
      "learning_rate": 3.6031695780813994e-05,
      "loss": 0.7413,
      "step": 936
    },
    {
      "epoch": 1.1984158732698615,
      "grad_norm": 0.6595590710639954,
      "learning_rate": 3.602235038325335e-05,
      "loss": 0.7272,
      "step": 937
    },
    {
      "epoch": 1.1996959756780543,
      "grad_norm": 0.6647142171859741,
      "learning_rate": 3.601299536106636e-05,
      "loss": 0.6955,
      "step": 938
    },
    {
      "epoch": 1.2009760780862468,
      "grad_norm": 0.6085090637207031,
      "learning_rate": 3.60036307206741e-05,
      "loss": 0.7304,
      "step": 939
    },
    {
      "epoch": 1.2022561804944396,
      "grad_norm": 0.7456724643707275,
      "learning_rate": 3.5994256468504245e-05,
      "loss": 0.7545,
      "step": 940
    },
    {
      "epoch": 1.2035362829026321,
      "grad_norm": 0.6394507884979248,
      "learning_rate": 3.598487261099108e-05,
      "loss": 0.7044,
      "step": 941
    },
    {
      "epoch": 1.204816385310825,
      "grad_norm": 0.5928666591644287,
      "learning_rate": 3.597547915457547e-05,
      "loss": 0.7002,
      "step": 942
    },
    {
      "epoch": 1.2060964877190175,
      "grad_norm": 0.6246052980422974,
      "learning_rate": 3.596607610570487e-05,
      "loss": 0.7465,
      "step": 943
    },
    {
      "epoch": 1.2073765901272102,
      "grad_norm": 0.6777945160865784,
      "learning_rate": 3.5956663470833326e-05,
      "loss": 0.7406,
      "step": 944
    },
    {
      "epoch": 1.2086566925354028,
      "grad_norm": 0.6257613897323608,
      "learning_rate": 3.594724125642145e-05,
      "loss": 0.6978,
      "step": 945
    },
    {
      "epoch": 1.2099367949435955,
      "grad_norm": 0.6669995784759521,
      "learning_rate": 3.593780946893645e-05,
      "loss": 0.7464,
      "step": 946
    },
    {
      "epoch": 1.211216897351788,
      "grad_norm": 0.6547150015830994,
      "learning_rate": 3.592836811485209e-05,
      "loss": 0.7196,
      "step": 947
    },
    {
      "epoch": 1.2124969997599808,
      "grad_norm": 0.5924673080444336,
      "learning_rate": 3.59189172006487e-05,
      "loss": 0.7623,
      "step": 948
    },
    {
      "epoch": 1.2137771021681734,
      "grad_norm": 0.5863516330718994,
      "learning_rate": 3.5909456732813175e-05,
      "loss": 0.7276,
      "step": 949
    },
    {
      "epoch": 1.2150572045763661,
      "grad_norm": 0.6385282278060913,
      "learning_rate": 3.589998671783897e-05,
      "loss": 0.7209,
      "step": 950
    },
    {
      "epoch": 1.2163373069845589,
      "grad_norm": 0.6384835839271545,
      "learning_rate": 3.589050716222609e-05,
      "loss": 0.7634,
      "step": 951
    },
    {
      "epoch": 1.2176174093927514,
      "grad_norm": 0.6587206125259399,
      "learning_rate": 3.588101807248109e-05,
      "loss": 0.7446,
      "step": 952
    },
    {
      "epoch": 1.218897511800944,
      "grad_norm": 0.6326109170913696,
      "learning_rate": 3.587151945511707e-05,
      "loss": 0.6902,
      "step": 953
    },
    {
      "epoch": 1.2201776142091367,
      "grad_norm": 0.6003281474113464,
      "learning_rate": 3.586201131665366e-05,
      "loss": 0.7207,
      "step": 954
    },
    {
      "epoch": 1.2214577166173295,
      "grad_norm": 0.605811595916748,
      "learning_rate": 3.585249366361703e-05,
      "loss": 0.727,
      "step": 955
    },
    {
      "epoch": 1.222737819025522,
      "grad_norm": 0.6090505123138428,
      "learning_rate": 3.58429665025399e-05,
      "loss": 0.7127,
      "step": 956
    },
    {
      "epoch": 1.2240179214337146,
      "grad_norm": 0.6559271216392517,
      "learning_rate": 3.583342983996149e-05,
      "loss": 0.7165,
      "step": 957
    },
    {
      "epoch": 1.2252980238419073,
      "grad_norm": 0.6495171189308167,
      "learning_rate": 3.5823883682427545e-05,
      "loss": 0.715,
      "step": 958
    },
    {
      "epoch": 1.2265781262501,
      "grad_norm": 0.6958083510398865,
      "learning_rate": 3.5814328036490346e-05,
      "loss": 0.7607,
      "step": 959
    },
    {
      "epoch": 1.2278582286582926,
      "grad_norm": 0.6588174104690552,
      "learning_rate": 3.5804762908708665e-05,
      "loss": 0.7227,
      "step": 960
    },
    {
      "epoch": 1.2278582286582926,
      "eval_loss": 0.011964739300310612,
      "eval_runtime": 10.8252,
      "eval_samples_per_second": 45.819,
      "eval_steps_per_second": 5.727,
      "step": 960
    },
    {
      "epoch": 1.2291383310664854,
      "grad_norm": 0.6300966739654541,
      "learning_rate": 3.5795188305647786e-05,
      "loss": 0.6902,
      "step": 961
    },
    {
      "epoch": 1.230418433474678,
      "grad_norm": 0.6609377861022949,
      "learning_rate": 3.5785604233879515e-05,
      "loss": 0.7238,
      "step": 962
    },
    {
      "epoch": 1.2316985358828707,
      "grad_norm": 0.6768269538879395,
      "learning_rate": 3.577601069998214e-05,
      "loss": 0.7112,
      "step": 963
    },
    {
      "epoch": 1.2329786382910632,
      "grad_norm": 0.6381678581237793,
      "learning_rate": 3.5766407710540435e-05,
      "loss": 0.7304,
      "step": 964
    },
    {
      "epoch": 1.234258740699256,
      "grad_norm": 0.6308141350746155,
      "learning_rate": 3.575679527214569e-05,
      "loss": 0.7418,
      "step": 965
    },
    {
      "epoch": 1.2355388431074485,
      "grad_norm": 0.6880090832710266,
      "learning_rate": 3.5747173391395665e-05,
      "loss": 0.7645,
      "step": 966
    },
    {
      "epoch": 1.2368189455156413,
      "grad_norm": 0.6165257692337036,
      "learning_rate": 3.5737542074894595e-05,
      "loss": 0.7046,
      "step": 967
    },
    {
      "epoch": 1.2380990479238339,
      "grad_norm": 0.6735715866088867,
      "learning_rate": 3.57279013292532e-05,
      "loss": 0.7337,
      "step": 968
    },
    {
      "epoch": 1.2393791503320266,
      "grad_norm": 0.6188624501228333,
      "learning_rate": 3.571825116108868e-05,
      "loss": 0.7185,
      "step": 969
    },
    {
      "epoch": 1.2406592527402192,
      "grad_norm": 0.7257860898971558,
      "learning_rate": 3.570859157702469e-05,
      "loss": 0.7116,
      "step": 970
    },
    {
      "epoch": 1.241939355148412,
      "grad_norm": 0.700437068939209,
      "learning_rate": 3.569892258369135e-05,
      "loss": 0.7252,
      "step": 971
    },
    {
      "epoch": 1.2432194575566045,
      "grad_norm": 0.7349070906639099,
      "learning_rate": 3.568924418772524e-05,
      "loss": 0.718,
      "step": 972
    },
    {
      "epoch": 1.2444995599647972,
      "grad_norm": 0.637186586856842,
      "learning_rate": 3.567955639576939e-05,
      "loss": 0.7161,
      "step": 973
    },
    {
      "epoch": 1.2457796623729898,
      "grad_norm": 0.6284525990486145,
      "learning_rate": 3.56698592144733e-05,
      "loss": 0.7388,
      "step": 974
    },
    {
      "epoch": 1.2470597647811825,
      "grad_norm": 0.7157219648361206,
      "learning_rate": 3.5660152650492874e-05,
      "loss": 0.7247,
      "step": 975
    },
    {
      "epoch": 1.248339867189375,
      "grad_norm": 0.7407729625701904,
      "learning_rate": 3.565043671049049e-05,
      "loss": 0.7587,
      "step": 976
    },
    {
      "epoch": 1.2496199695975678,
      "grad_norm": 0.6328476667404175,
      "learning_rate": 3.564071140113495e-05,
      "loss": 0.6988,
      "step": 977
    },
    {
      "epoch": 1.2509000720057606,
      "grad_norm": 0.6791459918022156,
      "learning_rate": 3.563097672910149e-05,
      "loss": 0.7544,
      "step": 978
    },
    {
      "epoch": 1.2521801744139531,
      "grad_norm": 0.6767594814300537,
      "learning_rate": 3.562123270107177e-05,
      "loss": 0.7559,
      "step": 979
    },
    {
      "epoch": 1.2534602768221457,
      "grad_norm": 0.6203661561012268,
      "learning_rate": 3.561147932373387e-05,
      "loss": 0.724,
      "step": 980
    },
    {
      "epoch": 1.2547403792303384,
      "grad_norm": 0.6236318349838257,
      "learning_rate": 3.5601716603782295e-05,
      "loss": 0.7418,
      "step": 981
    },
    {
      "epoch": 1.2560204816385312,
      "grad_norm": 0.6532002091407776,
      "learning_rate": 3.559194454791795e-05,
      "loss": 0.7387,
      "step": 982
    },
    {
      "epoch": 1.2573005840467237,
      "grad_norm": 0.6847975254058838,
      "learning_rate": 3.558216316284815e-05,
      "loss": 0.7529,
      "step": 983
    },
    {
      "epoch": 1.2585806864549163,
      "grad_norm": 0.6952961087226868,
      "learning_rate": 3.557237245528662e-05,
      "loss": 0.7284,
      "step": 984
    },
    {
      "epoch": 1.259860788863109,
      "grad_norm": 0.6524568796157837,
      "learning_rate": 3.556257243195349e-05,
      "loss": 0.7111,
      "step": 985
    },
    {
      "epoch": 1.2611408912713018,
      "grad_norm": 0.7332265973091125,
      "learning_rate": 3.555276309957528e-05,
      "loss": 0.733,
      "step": 986
    },
    {
      "epoch": 1.2624209936794943,
      "grad_norm": 0.6487500071525574,
      "learning_rate": 3.554294446488488e-05,
      "loss": 0.7473,
      "step": 987
    },
    {
      "epoch": 1.2637010960876869,
      "grad_norm": 0.6780820488929749,
      "learning_rate": 3.55331165346216e-05,
      "loss": 0.7353,
      "step": 988
    },
    {
      "epoch": 1.2649811984958796,
      "grad_norm": 0.6837990283966064,
      "learning_rate": 3.5523279315531084e-05,
      "loss": 0.7117,
      "step": 989
    },
    {
      "epoch": 1.2662613009040724,
      "grad_norm": 0.641559898853302,
      "learning_rate": 3.55134328143654e-05,
      "loss": 0.6989,
      "step": 990
    },
    {
      "epoch": 1.2662613009040724,
      "eval_loss": 0.011943795718252659,
      "eval_runtime": 10.8281,
      "eval_samples_per_second": 45.807,
      "eval_steps_per_second": 5.726,
      "step": 990
    },
    {
      "epoch": 1.267541403312265,
      "grad_norm": 0.7196192741394043,
      "learning_rate": 3.550357703788295e-05,
      "loss": 0.7338,
      "step": 991
    },
    {
      "epoch": 1.2688215057204577,
      "grad_norm": 0.6231567859649658,
      "learning_rate": 3.5493711992848536e-05,
      "loss": 0.7572,
      "step": 992
    },
    {
      "epoch": 1.2701016081286503,
      "grad_norm": 0.666607141494751,
      "learning_rate": 3.5483837686033284e-05,
      "loss": 0.709,
      "step": 993
    },
    {
      "epoch": 1.271381710536843,
      "grad_norm": 0.6255402565002441,
      "learning_rate": 3.547395412421471e-05,
      "loss": 0.71,
      "step": 994
    },
    {
      "epoch": 1.2726618129450356,
      "grad_norm": 0.6895260810852051,
      "learning_rate": 3.546406131417666e-05,
      "loss": 0.7424,
      "step": 995
    },
    {
      "epoch": 1.2739419153532283,
      "grad_norm": 0.7269278168678284,
      "learning_rate": 3.5454159262709354e-05,
      "loss": 0.7179,
      "step": 996
    },
    {
      "epoch": 1.2752220177614209,
      "grad_norm": 0.6350796222686768,
      "learning_rate": 3.544424797660931e-05,
      "loss": 0.7251,
      "step": 997
    },
    {
      "epoch": 1.2765021201696136,
      "grad_norm": 0.6415387988090515,
      "learning_rate": 3.543432746267943e-05,
      "loss": 0.7502,
      "step": 998
    },
    {
      "epoch": 1.2777822225778062,
      "grad_norm": 0.6302523612976074,
      "learning_rate": 3.542439772772893e-05,
      "loss": 0.7094,
      "step": 999
    },
    {
      "epoch": 1.279062324985999,
      "grad_norm": 0.6089184284210205,
      "learning_rate": 3.541445877857335e-05,
      "loss": 0.7144,
      "step": 1000
    },
    {
      "epoch": 1.2803424273941915,
      "grad_norm": 0.639612078666687,
      "learning_rate": 3.540451062203457e-05,
      "loss": 0.6652,
      "step": 1001
    },
    {
      "epoch": 1.2816225298023842,
      "grad_norm": 0.7329027652740479,
      "learning_rate": 3.539455326494078e-05,
      "loss": 0.7331,
      "step": 1002
    },
    {
      "epoch": 1.2829026322105768,
      "grad_norm": 0.5882426500320435,
      "learning_rate": 3.5384586714126486e-05,
      "loss": 0.6926,
      "step": 1003
    },
    {
      "epoch": 1.2841827346187695,
      "grad_norm": 0.6871622204780579,
      "learning_rate": 3.5374610976432496e-05,
      "loss": 0.7433,
      "step": 1004
    },
    {
      "epoch": 1.2854628370269623,
      "grad_norm": 0.6158824563026428,
      "learning_rate": 3.5364626058705944e-05,
      "loss": 0.7034,
      "step": 1005
    },
    {
      "epoch": 1.2867429394351548,
      "grad_norm": 0.6488764882087708,
      "learning_rate": 3.535463196780025e-05,
      "loss": 0.752,
      "step": 1006
    },
    {
      "epoch": 1.2880230418433474,
      "grad_norm": 0.6519235372543335,
      "learning_rate": 3.534462871057514e-05,
      "loss": 0.7448,
      "step": 1007
    },
    {
      "epoch": 1.2893031442515401,
      "grad_norm": 0.6193985342979431,
      "learning_rate": 3.5334616293896596e-05,
      "loss": 0.7201,
      "step": 1008
    },
    {
      "epoch": 1.290583246659733,
      "grad_norm": 0.587981641292572,
      "learning_rate": 3.532459472463695e-05,
      "loss": 0.7323,
      "step": 1009
    },
    {
      "epoch": 1.2918633490679254,
      "grad_norm": 0.59144526720047,
      "learning_rate": 3.531456400967477e-05,
      "loss": 0.7284,
      "step": 1010
    },
    {
      "epoch": 1.293143451476118,
      "grad_norm": 0.6190255880355835,
      "learning_rate": 3.530452415589491e-05,
      "loss": 0.7427,
      "step": 1011
    },
    {
      "epoch": 1.2944235538843107,
      "grad_norm": 0.6405760645866394,
      "learning_rate": 3.52944751701885e-05,
      "loss": 0.7389,
      "step": 1012
    },
    {
      "epoch": 1.2957036562925035,
      "grad_norm": 0.6301907896995544,
      "learning_rate": 3.528441705945294e-05,
      "loss": 0.7452,
      "step": 1013
    },
    {
      "epoch": 1.296983758700696,
      "grad_norm": 0.610808253288269,
      "learning_rate": 3.52743498305919e-05,
      "loss": 0.7202,
      "step": 1014
    },
    {
      "epoch": 1.2982638611088886,
      "grad_norm": 0.6241909861564636,
      "learning_rate": 3.526427349051528e-05,
      "loss": 0.7354,
      "step": 1015
    },
    {
      "epoch": 1.2995439635170813,
      "grad_norm": 0.5971837639808655,
      "learning_rate": 3.525418804613928e-05,
      "loss": 0.6881,
      "step": 1016
    },
    {
      "epoch": 1.300824065925274,
      "grad_norm": 0.5995813012123108,
      "learning_rate": 3.52440935043863e-05,
      "loss": 0.6979,
      "step": 1017
    },
    {
      "epoch": 1.3021041683334666,
      "grad_norm": 0.703272819519043,
      "learning_rate": 3.523398987218502e-05,
      "loss": 0.7622,
      "step": 1018
    },
    {
      "epoch": 1.3033842707416594,
      "grad_norm": 0.6928847432136536,
      "learning_rate": 3.5223877156470333e-05,
      "loss": 0.7324,
      "step": 1019
    },
    {
      "epoch": 1.304664373149852,
      "grad_norm": 0.746273934841156,
      "learning_rate": 3.521375536418339e-05,
      "loss": 0.7471,
      "step": 1020
    },
    {
      "epoch": 1.304664373149852,
      "eval_loss": 0.011940889991819859,
      "eval_runtime": 10.8334,
      "eval_samples_per_second": 45.784,
      "eval_steps_per_second": 5.723,
      "step": 1020
    },
    {
      "epoch": 1.3059444755580447,
      "grad_norm": 0.6955692768096924,
      "learning_rate": 3.520362450227155e-05,
      "loss": 0.7192,
      "step": 1021
    },
    {
      "epoch": 1.3072245779662373,
      "grad_norm": 0.6236073970794678,
      "learning_rate": 3.519348457768844e-05,
      "loss": 0.7056,
      "step": 1022
    },
    {
      "epoch": 1.30850468037443,
      "grad_norm": 0.6319799423217773,
      "learning_rate": 3.518333559739384e-05,
      "loss": 0.7278,
      "step": 1023
    },
    {
      "epoch": 1.3097847827826226,
      "grad_norm": 0.6991655826568604,
      "learning_rate": 3.5173177568353796e-05,
      "loss": 0.7334,
      "step": 1024
    },
    {
      "epoch": 1.3110648851908153,
      "grad_norm": 0.6507253646850586,
      "learning_rate": 3.5163010497540556e-05,
      "loss": 0.7064,
      "step": 1025
    },
    {
      "epoch": 1.3123449875990079,
      "grad_norm": 0.6370417475700378,
      "learning_rate": 3.515283439193257e-05,
      "loss": 0.71,
      "step": 1026
    },
    {
      "epoch": 1.3136250900072006,
      "grad_norm": 0.6877231597900391,
      "learning_rate": 3.514264925851448e-05,
      "loss": 0.7485,
      "step": 1027
    },
    {
      "epoch": 1.3149051924153932,
      "grad_norm": 0.6662914752960205,
      "learning_rate": 3.5132455104277135e-05,
      "loss": 0.743,
      "step": 1028
    },
    {
      "epoch": 1.316185294823586,
      "grad_norm": 0.6527225375175476,
      "learning_rate": 3.5122251936217575e-05,
      "loss": 0.7124,
      "step": 1029
    },
    {
      "epoch": 1.3174653972317785,
      "grad_norm": 0.6729642152786255,
      "learning_rate": 3.511203976133903e-05,
      "loss": 0.726,
      "step": 1030
    },
    {
      "epoch": 1.3187454996399712,
      "grad_norm": 0.6522324681282043,
      "learning_rate": 3.51018185866509e-05,
      "loss": 0.7289,
      "step": 1031
    },
    {
      "epoch": 1.3200256020481638,
      "grad_norm": 0.663841187953949,
      "learning_rate": 3.509158841916877e-05,
      "loss": 0.7223,
      "step": 1032
    },
    {
      "epoch": 1.3213057044563565,
      "grad_norm": 0.636076033115387,
      "learning_rate": 3.50813492659144e-05,
      "loss": 0.7264,
      "step": 1033
    },
    {
      "epoch": 1.322585806864549,
      "grad_norm": 0.666597843170166,
      "learning_rate": 3.507110113391571e-05,
      "loss": 0.7054,
      "step": 1034
    },
    {
      "epoch": 1.3238659092727418,
      "grad_norm": 0.6378565430641174,
      "learning_rate": 3.50608440302068e-05,
      "loss": 0.7035,
      "step": 1035
    },
    {
      "epoch": 1.3251460116809346,
      "grad_norm": 0.650204598903656,
      "learning_rate": 3.5050577961827904e-05,
      "loss": 0.7078,
      "step": 1036
    },
    {
      "epoch": 1.3264261140891271,
      "grad_norm": 0.6551227569580078,
      "learning_rate": 3.504030293582543e-05,
      "loss": 0.7501,
      "step": 1037
    },
    {
      "epoch": 1.3277062164973197,
      "grad_norm": 0.6497569680213928,
      "learning_rate": 3.5030018959251915e-05,
      "loss": 0.7048,
      "step": 1038
    },
    {
      "epoch": 1.3289863189055124,
      "grad_norm": 0.703510046005249,
      "learning_rate": 3.5019726039166055e-05,
      "loss": 0.7359,
      "step": 1039
    },
    {
      "epoch": 1.3302664213137052,
      "grad_norm": 0.6817246675491333,
      "learning_rate": 3.500942418263267e-05,
      "loss": 0.7005,
      "step": 1040
    },
    {
      "epoch": 1.3315465237218977,
      "grad_norm": 0.7863104939460754,
      "learning_rate": 3.499911339672274e-05,
      "loss": 0.7522,
      "step": 1041
    },
    {
      "epoch": 1.3328266261300903,
      "grad_norm": 0.638771653175354,
      "learning_rate": 3.498879368851335e-05,
      "loss": 0.7106,
      "step": 1042
    },
    {
      "epoch": 1.334106728538283,
      "grad_norm": 0.6441402435302734,
      "learning_rate": 3.497846506508771e-05,
      "loss": 0.7382,
      "step": 1043
    },
    {
      "epoch": 1.3353868309464758,
      "grad_norm": 0.5955315232276917,
      "learning_rate": 3.496812753353516e-05,
      "loss": 0.732,
      "step": 1044
    },
    {
      "epoch": 1.3366669333546684,
      "grad_norm": 0.6294196248054504,
      "learning_rate": 3.495778110095114e-05,
      "loss": 0.7579,
      "step": 1045
    },
    {
      "epoch": 1.337947035762861,
      "grad_norm": 0.643982470035553,
      "learning_rate": 3.494742577443723e-05,
      "loss": 0.7588,
      "step": 1046
    },
    {
      "epoch": 1.3392271381710537,
      "grad_norm": 0.6264019012451172,
      "learning_rate": 3.493706156110107e-05,
      "loss": 0.7379,
      "step": 1047
    },
    {
      "epoch": 1.3405072405792464,
      "grad_norm": 0.6815388202667236,
      "learning_rate": 3.492668846805644e-05,
      "loss": 0.7185,
      "step": 1048
    },
    {
      "epoch": 1.341787342987439,
      "grad_norm": 0.6314342021942139,
      "learning_rate": 3.491630650242319e-05,
      "loss": 0.7261,
      "step": 1049
    },
    {
      "epoch": 1.3430674453956317,
      "grad_norm": 0.5777370929718018,
      "learning_rate": 3.4905915671327274e-05,
      "loss": 0.742,
      "step": 1050
    },
    {
      "epoch": 1.3430674453956317,
      "eval_loss": 0.011894307099282742,
      "eval_runtime": 10.8212,
      "eval_samples_per_second": 45.836,
      "eval_steps_per_second": 5.73,
      "step": 1050
    },
    {
      "epoch": 1.3443475478038243,
      "grad_norm": 0.5540257096290588,
      "learning_rate": 3.489551598190072e-05,
      "loss": 0.7038,
      "step": 1051
    },
    {
      "epoch": 1.345627650212017,
      "grad_norm": 0.633642315864563,
      "learning_rate": 3.488510744128164e-05,
      "loss": 0.7181,
      "step": 1052
    },
    {
      "epoch": 1.3469077526202096,
      "grad_norm": 0.5957219004631042,
      "learning_rate": 3.4874690056614225e-05,
      "loss": 0.7617,
      "step": 1053
    },
    {
      "epoch": 1.3481878550284023,
      "grad_norm": 0.6198498606681824,
      "learning_rate": 3.4864263835048735e-05,
      "loss": 0.7107,
      "step": 1054
    },
    {
      "epoch": 1.3494679574365949,
      "grad_norm": 0.6253262162208557,
      "learning_rate": 3.48538287837415e-05,
      "loss": 0.7729,
      "step": 1055
    },
    {
      "epoch": 1.3507480598447876,
      "grad_norm": 0.6212433576583862,
      "learning_rate": 3.484338490985489e-05,
      "loss": 0.7112,
      "step": 1056
    },
    {
      "epoch": 1.3520281622529802,
      "grad_norm": 0.5720719695091248,
      "learning_rate": 3.4832932220557355e-05,
      "loss": 0.7236,
      "step": 1057
    },
    {
      "epoch": 1.353308264661173,
      "grad_norm": 0.6214074492454529,
      "learning_rate": 3.4822470723023385e-05,
      "loss": 0.7011,
      "step": 1058
    },
    {
      "epoch": 1.3545883670693655,
      "grad_norm": 0.621953547000885,
      "learning_rate": 3.481200042443352e-05,
      "loss": 0.7842,
      "step": 1059
    },
    {
      "epoch": 1.3558684694775582,
      "grad_norm": 0.601488471031189,
      "learning_rate": 3.480152133197433e-05,
      "loss": 0.7494,
      "step": 1060
    },
    {
      "epoch": 1.3571485718857508,
      "grad_norm": 0.6324599981307983,
      "learning_rate": 3.479103345283843e-05,
      "loss": 0.7281,
      "step": 1061
    },
    {
      "epoch": 1.3584286742939435,
      "grad_norm": 0.5631807446479797,
      "learning_rate": 3.478053679422447e-05,
      "loss": 0.6959,
      "step": 1062
    },
    {
      "epoch": 1.3597087767021363,
      "grad_norm": 0.6149842739105225,
      "learning_rate": 3.4770031363337126e-05,
      "loss": 0.7374,
      "step": 1063
    },
    {
      "epoch": 1.3609888791103288,
      "grad_norm": 0.6333210468292236,
      "learning_rate": 3.475951716738708e-05,
      "loss": 0.7033,
      "step": 1064
    },
    {
      "epoch": 1.3622689815185214,
      "grad_norm": 0.6083963513374329,
      "learning_rate": 3.474899421359103e-05,
      "loss": 0.7021,
      "step": 1065
    },
    {
      "epoch": 1.3635490839267141,
      "grad_norm": 0.6221580505371094,
      "learning_rate": 3.473846250917172e-05,
      "loss": 0.7283,
      "step": 1066
    },
    {
      "epoch": 1.364829186334907,
      "grad_norm": 0.6208587288856506,
      "learning_rate": 3.472792206135786e-05,
      "loss": 0.7337,
      "step": 1067
    },
    {
      "epoch": 1.3661092887430994,
      "grad_norm": 0.660995602607727,
      "learning_rate": 3.4717372877384175e-05,
      "loss": 0.7227,
      "step": 1068
    },
    {
      "epoch": 1.367389391151292,
      "grad_norm": 0.633240282535553,
      "learning_rate": 3.4706814964491395e-05,
      "loss": 0.7324,
      "step": 1069
    },
    {
      "epoch": 1.3686694935594848,
      "grad_norm": 0.740360677242279,
      "learning_rate": 3.469624832992623e-05,
      "loss": 0.7286,
      "step": 1070
    },
    {
      "epoch": 1.3699495959676775,
      "grad_norm": 0.6104412078857422,
      "learning_rate": 3.4685672980941384e-05,
      "loss": 0.7155,
      "step": 1071
    },
    {
      "epoch": 1.37122969837587,
      "grad_norm": 0.6899016499519348,
      "learning_rate": 3.4675088924795535e-05,
      "loss": 0.698,
      "step": 1072
    },
    {
      "epoch": 1.3725098007840626,
      "grad_norm": 0.6214810013771057,
      "learning_rate": 3.466449616875335e-05,
      "loss": 0.768,
      "step": 1073
    },
    {
      "epoch": 1.3737899031922554,
      "grad_norm": 0.7415969371795654,
      "learning_rate": 3.4653894720085443e-05,
      "loss": 0.7356,
      "step": 1074
    },
    {
      "epoch": 1.3750700056004481,
      "grad_norm": 0.6360346078872681,
      "learning_rate": 3.464328458606843e-05,
      "loss": 0.7339,
      "step": 1075
    },
    {
      "epoch": 1.3763501080086407,
      "grad_norm": 0.7460790276527405,
      "learning_rate": 3.463266577398485e-05,
      "loss": 0.7524,
      "step": 1076
    },
    {
      "epoch": 1.3776302104168334,
      "grad_norm": 0.6799301505088806,
      "learning_rate": 3.462203829112323e-05,
      "loss": 0.7217,
      "step": 1077
    },
    {
      "epoch": 1.378910312825026,
      "grad_norm": 0.7293802499771118,
      "learning_rate": 3.4611402144778027e-05,
      "loss": 0.7501,
      "step": 1078
    },
    {
      "epoch": 1.3801904152332187,
      "grad_norm": 0.6283996105194092,
      "learning_rate": 3.460075734224966e-05,
      "loss": 0.7062,
      "step": 1079
    },
    {
      "epoch": 1.3814705176414113,
      "grad_norm": 0.744900643825531,
      "learning_rate": 3.459010389084448e-05,
      "loss": 0.7323,
      "step": 1080
    },
    {
      "epoch": 1.3814705176414113,
      "eval_loss": 0.011888692155480385,
      "eval_runtime": 10.8204,
      "eval_samples_per_second": 45.839,
      "eval_steps_per_second": 5.73,
      "step": 1080
    },
    {
      "epoch": 1.382750620049604,
      "grad_norm": 0.6258221864700317,
      "learning_rate": 3.457944179787477e-05,
      "loss": 0.732,
      "step": 1081
    },
    {
      "epoch": 1.3840307224577966,
      "grad_norm": 0.7455996870994568,
      "learning_rate": 3.4568771070658764e-05,
      "loss": 0.7384,
      "step": 1082
    },
    {
      "epoch": 1.3853108248659893,
      "grad_norm": 0.6790741682052612,
      "learning_rate": 3.45580917165206e-05,
      "loss": 0.7187,
      "step": 1083
    },
    {
      "epoch": 1.3865909272741819,
      "grad_norm": 0.7180479168891907,
      "learning_rate": 3.4547403742790346e-05,
      "loss": 0.763,
      "step": 1084
    },
    {
      "epoch": 1.3878710296823746,
      "grad_norm": 0.6763840913772583,
      "learning_rate": 3.4536707156803996e-05,
      "loss": 0.7108,
      "step": 1085
    },
    {
      "epoch": 1.3891511320905672,
      "grad_norm": 0.7159371972084045,
      "learning_rate": 3.452600196590344e-05,
      "loss": 0.7035,
      "step": 1086
    },
    {
      "epoch": 1.39043123449876,
      "grad_norm": 0.64292311668396,
      "learning_rate": 3.4515288177436474e-05,
      "loss": 0.7282,
      "step": 1087
    },
    {
      "epoch": 1.3917113369069525,
      "grad_norm": 0.708309531211853,
      "learning_rate": 3.450456579875682e-05,
      "loss": 0.73,
      "step": 1088
    },
    {
      "epoch": 1.3929914393151452,
      "grad_norm": 0.6651067137718201,
      "learning_rate": 3.4493834837224067e-05,
      "loss": 0.7334,
      "step": 1089
    },
    {
      "epoch": 1.394271541723338,
      "grad_norm": 0.6737949848175049,
      "learning_rate": 3.44830953002037e-05,
      "loss": 0.71,
      "step": 1090
    },
    {
      "epoch": 1.3955516441315305,
      "grad_norm": 0.579788088798523,
      "learning_rate": 3.44723471950671e-05,
      "loss": 0.7185,
      "step": 1091
    },
    {
      "epoch": 1.396831746539723,
      "grad_norm": 0.7158629298210144,
      "learning_rate": 3.446159052919153e-05,
      "loss": 0.7259,
      "step": 1092
    },
    {
      "epoch": 1.3981118489479158,
      "grad_norm": 0.6628416180610657,
      "learning_rate": 3.445082530996012e-05,
      "loss": 0.7808,
      "step": 1093
    },
    {
      "epoch": 1.3993919513561086,
      "grad_norm": 0.6353318691253662,
      "learning_rate": 3.4440051544761876e-05,
      "loss": 0.6881,
      "step": 1094
    },
    {
      "epoch": 1.4006720537643012,
      "grad_norm": 0.7098138928413391,
      "learning_rate": 3.442926924099167e-05,
      "loss": 0.7411,
      "step": 1095
    },
    {
      "epoch": 1.4019521561724937,
      "grad_norm": 0.6623754501342773,
      "learning_rate": 3.441847840605023e-05,
      "loss": 0.7424,
      "step": 1096
    },
    {
      "epoch": 1.4032322585806865,
      "grad_norm": 0.7072277665138245,
      "learning_rate": 3.440767904734414e-05,
      "loss": 0.7579,
      "step": 1097
    },
    {
      "epoch": 1.4045123609888792,
      "grad_norm": 0.6239615678787231,
      "learning_rate": 3.439687117228585e-05,
      "loss": 0.7213,
      "step": 1098
    },
    {
      "epoch": 1.4057924633970718,
      "grad_norm": 0.6581880450248718,
      "learning_rate": 3.4386054788293625e-05,
      "loss": 0.7252,
      "step": 1099
    },
    {
      "epoch": 1.4070725658052643,
      "grad_norm": 0.5721520781517029,
      "learning_rate": 3.437522990279161e-05,
      "loss": 0.7231,
      "step": 1100
    },
    {
      "epoch": 1.408352668213457,
      "grad_norm": 0.6812596917152405,
      "learning_rate": 3.436439652320973e-05,
      "loss": 0.7259,
      "step": 1101
    },
    {
      "epoch": 1.4096327706216498,
      "grad_norm": 0.6836157441139221,
      "learning_rate": 3.435355465698381e-05,
      "loss": 0.7804,
      "step": 1102
    },
    {
      "epoch": 1.4109128730298424,
      "grad_norm": 0.6757557392120361,
      "learning_rate": 3.434270431155544e-05,
      "loss": 0.7438,
      "step": 1103
    },
    {
      "epoch": 1.4121929754380351,
      "grad_norm": 0.6978136897087097,
      "learning_rate": 3.433184549437206e-05,
      "loss": 0.7584,
      "step": 1104
    },
    {
      "epoch": 1.4134730778462277,
      "grad_norm": 0.6466197967529297,
      "learning_rate": 3.432097821288692e-05,
      "loss": 0.7506,
      "step": 1105
    },
    {
      "epoch": 1.4147531802544204,
      "grad_norm": 0.6686610579490662,
      "learning_rate": 3.431010247455908e-05,
      "loss": 0.7326,
      "step": 1106
    },
    {
      "epoch": 1.416033282662613,
      "grad_norm": 0.6493745446205139,
      "learning_rate": 3.4299218286853394e-05,
      "loss": 0.7103,
      "step": 1107
    },
    {
      "epoch": 1.4173133850708057,
      "grad_norm": 0.6098056435585022,
      "learning_rate": 3.428832565724053e-05,
      "loss": 0.7693,
      "step": 1108
    },
    {
      "epoch": 1.4185934874789983,
      "grad_norm": 0.6214213967323303,
      "learning_rate": 3.427742459319694e-05,
      "loss": 0.7364,
      "step": 1109
    },
    {
      "epoch": 1.419873589887191,
      "grad_norm": 0.6468144655227661,
      "learning_rate": 3.426651510220488e-05,
      "loss": 0.7149,
      "step": 1110
    },
    {
      "epoch": 1.419873589887191,
      "eval_loss": 0.011858620680868626,
      "eval_runtime": 10.8144,
      "eval_samples_per_second": 45.865,
      "eval_steps_per_second": 5.733,
      "step": 1110
    },
    {
      "epoch": 1.4211536922953836,
      "grad_norm": 0.5441794395446777,
      "learning_rate": 3.4255597191752365e-05,
      "loss": 0.6997,
      "step": 1111
    },
    {
      "epoch": 1.4224337947035763,
      "grad_norm": 0.6870185732841492,
      "learning_rate": 3.424467086933322e-05,
      "loss": 0.6852,
      "step": 1112
    },
    {
      "epoch": 1.4237138971117689,
      "grad_norm": 0.6588074564933777,
      "learning_rate": 3.423373614244702e-05,
      "loss": 0.7463,
      "step": 1113
    },
    {
      "epoch": 1.4249939995199616,
      "grad_norm": 0.6720722913742065,
      "learning_rate": 3.42227930185991e-05,
      "loss": 0.6804,
      "step": 1114
    },
    {
      "epoch": 1.4262741019281542,
      "grad_norm": 0.6890459060668945,
      "learning_rate": 3.42118415053006e-05,
      "loss": 0.7552,
      "step": 1115
    },
    {
      "epoch": 1.427554204336347,
      "grad_norm": 0.6325091123580933,
      "learning_rate": 3.420088161006837e-05,
      "loss": 0.7315,
      "step": 1116
    },
    {
      "epoch": 1.4288343067445395,
      "grad_norm": 0.6969271898269653,
      "learning_rate": 3.4189913340425054e-05,
      "loss": 0.7108,
      "step": 1117
    },
    {
      "epoch": 1.4301144091527322,
      "grad_norm": 0.6649100184440613,
      "learning_rate": 3.417893670389902e-05,
      "loss": 0.7129,
      "step": 1118
    },
    {
      "epoch": 1.4313945115609248,
      "grad_norm": 0.6117990016937256,
      "learning_rate": 3.4167951708024386e-05,
      "loss": 0.7243,
      "step": 1119
    },
    {
      "epoch": 1.4326746139691175,
      "grad_norm": 0.6294746398925781,
      "learning_rate": 3.4156958360341006e-05,
      "loss": 0.7066,
      "step": 1120
    },
    {
      "epoch": 1.4339547163773103,
      "grad_norm": 0.6256268620491028,
      "learning_rate": 3.414595666839446e-05,
      "loss": 0.7359,
      "step": 1121
    },
    {
      "epoch": 1.4352348187855029,
      "grad_norm": 0.5756694078445435,
      "learning_rate": 3.413494663973606e-05,
      "loss": 0.7462,
      "step": 1122
    },
    {
      "epoch": 1.4365149211936954,
      "grad_norm": 0.7117789387702942,
      "learning_rate": 3.4123928281922865e-05,
      "loss": 0.7292,
      "step": 1123
    },
    {
      "epoch": 1.4377950236018882,
      "grad_norm": 0.652121365070343,
      "learning_rate": 3.41129016025176e-05,
      "loss": 0.6941,
      "step": 1124
    },
    {
      "epoch": 1.439075126010081,
      "grad_norm": 0.6118738055229187,
      "learning_rate": 3.4101866609088755e-05,
      "loss": 0.6771,
      "step": 1125
    },
    {
      "epoch": 1.4403552284182735,
      "grad_norm": 0.7171832919120789,
      "learning_rate": 3.409082330921048e-05,
      "loss": 0.7454,
      "step": 1126
    },
    {
      "epoch": 1.441635330826466,
      "grad_norm": 0.7150013446807861,
      "learning_rate": 3.407977171046267e-05,
      "loss": 0.751,
      "step": 1127
    },
    {
      "epoch": 1.4429154332346588,
      "grad_norm": 0.6304430961608887,
      "learning_rate": 3.4068711820430866e-05,
      "loss": 0.7218,
      "step": 1128
    },
    {
      "epoch": 1.4441955356428515,
      "grad_norm": 0.6315369009971619,
      "learning_rate": 3.405764364670635e-05,
      "loss": 0.6788,
      "step": 1129
    },
    {
      "epoch": 1.445475638051044,
      "grad_norm": 0.6245869994163513,
      "learning_rate": 3.4046567196886076e-05,
      "loss": 0.7266,
      "step": 1130
    },
    {
      "epoch": 1.4467557404592366,
      "grad_norm": 0.6458489298820496,
      "learning_rate": 3.403548247857264e-05,
      "loss": 0.7156,
      "step": 1131
    },
    {
      "epoch": 1.4480358428674294,
      "grad_norm": 0.5823201537132263,
      "learning_rate": 3.402438949937436e-05,
      "loss": 0.7041,
      "step": 1132
    },
    {
      "epoch": 1.4493159452756221,
      "grad_norm": 0.6566415429115295,
      "learning_rate": 3.4013288266905214e-05,
      "loss": 0.747,
      "step": 1133
    },
    {
      "epoch": 1.4505960476838147,
      "grad_norm": 0.6050204634666443,
      "learning_rate": 3.4002178788784834e-05,
      "loss": 0.7443,
      "step": 1134
    },
    {
      "epoch": 1.4518761500920074,
      "grad_norm": 0.690113365650177,
      "learning_rate": 3.399106107263851e-05,
      "loss": 0.7656,
      "step": 1135
    },
    {
      "epoch": 1.4531562525002,
      "grad_norm": 0.634456217288971,
      "learning_rate": 3.39799351260972e-05,
      "loss": 0.7138,
      "step": 1136
    },
    {
      "epoch": 1.4544363549083927,
      "grad_norm": 0.6726987957954407,
      "learning_rate": 3.396880095679749e-05,
      "loss": 0.7415,
      "step": 1137
    },
    {
      "epoch": 1.4557164573165853,
      "grad_norm": 0.6116801500320435,
      "learning_rate": 3.3957658572381654e-05,
      "loss": 0.7422,
      "step": 1138
    },
    {
      "epoch": 1.456996559724778,
      "grad_norm": 0.6155052185058594,
      "learning_rate": 3.394650798049755e-05,
      "loss": 0.7494,
      "step": 1139
    },
    {
      "epoch": 1.4582766621329706,
      "grad_norm": 0.6293736696243286,
      "learning_rate": 3.393534918879869e-05,
      "loss": 0.7214,
      "step": 1140
    },
    {
      "epoch": 1.4582766621329706,
      "eval_loss": 0.011847763322293758,
      "eval_runtime": 10.8318,
      "eval_samples_per_second": 45.791,
      "eval_steps_per_second": 5.724,
      "step": 1140
    },
    {
      "epoch": 1.4595567645411633,
      "grad_norm": 0.6018232703208923,
      "learning_rate": 3.392418220494423e-05,
      "loss": 0.6992,
      "step": 1141
    },
    {
      "epoch": 1.4608368669493559,
      "grad_norm": 0.6392772793769836,
      "learning_rate": 3.3913007036598936e-05,
      "loss": 0.7247,
      "step": 1142
    },
    {
      "epoch": 1.4621169693575486,
      "grad_norm": 0.5587735176086426,
      "learning_rate": 3.390182369143319e-05,
      "loss": 0.7088,
      "step": 1143
    },
    {
      "epoch": 1.4633970717657412,
      "grad_norm": 0.6419893503189087,
      "learning_rate": 3.389063217712299e-05,
      "loss": 0.6749,
      "step": 1144
    },
    {
      "epoch": 1.464677174173934,
      "grad_norm": 0.5953524708747864,
      "learning_rate": 3.3879432501349945e-05,
      "loss": 0.7109,
      "step": 1145
    },
    {
      "epoch": 1.4659572765821265,
      "grad_norm": 0.6553881168365479,
      "learning_rate": 3.386822467180124e-05,
      "loss": 0.7357,
      "step": 1146
    },
    {
      "epoch": 1.4672373789903193,
      "grad_norm": 0.601859986782074,
      "learning_rate": 3.3857008696169705e-05,
      "loss": 0.7364,
      "step": 1147
    },
    {
      "epoch": 1.468517481398512,
      "grad_norm": 0.6729167699813843,
      "learning_rate": 3.384578458215371e-05,
      "loss": 0.7491,
      "step": 1148
    },
    {
      "epoch": 1.4697975838067046,
      "grad_norm": 0.6336567401885986,
      "learning_rate": 3.383455233745727e-05,
      "loss": 0.7416,
      "step": 1149
    },
    {
      "epoch": 1.471077686214897,
      "grad_norm": 0.6387529969215393,
      "learning_rate": 3.38233119697899e-05,
      "loss": 0.7185,
      "step": 1150
    },
    {
      "epoch": 1.4723577886230899,
      "grad_norm": 0.5772356390953064,
      "learning_rate": 3.3812063486866766e-05,
      "loss": 0.7306,
      "step": 1151
    },
    {
      "epoch": 1.4736378910312826,
      "grad_norm": 0.6451202034950256,
      "learning_rate": 3.3800806896408564e-05,
      "loss": 0.714,
      "step": 1152
    },
    {
      "epoch": 1.4749179934394752,
      "grad_norm": 0.6001375913619995,
      "learning_rate": 3.378954220614156e-05,
      "loss": 0.7366,
      "step": 1153
    },
    {
      "epoch": 1.4761980958476677,
      "grad_norm": 0.6709045171737671,
      "learning_rate": 3.3778269423797604e-05,
      "loss": 0.7475,
      "step": 1154
    },
    {
      "epoch": 1.4774781982558605,
      "grad_norm": 0.6364221572875977,
      "learning_rate": 3.376698855711406e-05,
      "loss": 0.7241,
      "step": 1155
    },
    {
      "epoch": 1.4787583006640532,
      "grad_norm": 0.6763046979904175,
      "learning_rate": 3.375569961383387e-05,
      "loss": 0.6697,
      "step": 1156
    },
    {
      "epoch": 1.4800384030722458,
      "grad_norm": 0.639137864112854,
      "learning_rate": 3.374440260170551e-05,
      "loss": 0.7435,
      "step": 1157
    },
    {
      "epoch": 1.4813185054804383,
      "grad_norm": 0.6539855599403381,
      "learning_rate": 3.3733097528482994e-05,
      "loss": 0.7751,
      "step": 1158
    },
    {
      "epoch": 1.482598607888631,
      "grad_norm": 0.6638590693473816,
      "learning_rate": 3.372178440192588e-05,
      "loss": 0.7445,
      "step": 1159
    },
    {
      "epoch": 1.4838787102968238,
      "grad_norm": 0.726880669593811,
      "learning_rate": 3.3710463229799226e-05,
      "loss": 0.6906,
      "step": 1160
    },
    {
      "epoch": 1.4851588127050164,
      "grad_norm": 0.6935184001922607,
      "learning_rate": 3.369913401987364e-05,
      "loss": 0.7329,
      "step": 1161
    },
    {
      "epoch": 1.4864389151132091,
      "grad_norm": 0.7429734468460083,
      "learning_rate": 3.368779677992524e-05,
      "loss": 0.7127,
      "step": 1162
    },
    {
      "epoch": 1.4877190175214017,
      "grad_norm": 0.6960635185241699,
      "learning_rate": 3.367645151773565e-05,
      "loss": 0.7299,
      "step": 1163
    },
    {
      "epoch": 1.4889991199295944,
      "grad_norm": 0.6126359105110168,
      "learning_rate": 3.366509824109201e-05,
      "loss": 0.7215,
      "step": 1164
    },
    {
      "epoch": 1.490279222337787,
      "grad_norm": 0.6716033816337585,
      "learning_rate": 3.365373695778694e-05,
      "loss": 0.6996,
      "step": 1165
    },
    {
      "epoch": 1.4915593247459797,
      "grad_norm": 0.6191390156745911,
      "learning_rate": 3.3642367675618576e-05,
      "loss": 0.7291,
      "step": 1166
    },
    {
      "epoch": 1.4928394271541723,
      "grad_norm": 0.6926732659339905,
      "learning_rate": 3.363099040239055e-05,
      "loss": 0.6972,
      "step": 1167
    },
    {
      "epoch": 1.494119529562365,
      "grad_norm": 0.5820816159248352,
      "learning_rate": 3.3619605145911935e-05,
      "loss": 0.7238,
      "step": 1168
    },
    {
      "epoch": 1.4953996319705576,
      "grad_norm": 0.6890411972999573,
      "learning_rate": 3.360821191399736e-05,
      "loss": 0.7052,
      "step": 1169
    },
    {
      "epoch": 1.4966797343787503,
      "grad_norm": 0.6308721899986267,
      "learning_rate": 3.359681071446685e-05,
      "loss": 0.7523,
      "step": 1170
    },
    {
      "epoch": 1.4966797343787503,
      "eval_loss": 0.011831529438495636,
      "eval_runtime": 10.8133,
      "eval_samples_per_second": 45.869,
      "eval_steps_per_second": 5.734,
      "step": 1170
    },
    {
      "epoch": 1.4979598367869429,
      "grad_norm": 0.625635027885437,
      "learning_rate": 3.358540155514594e-05,
      "loss": 0.7298,
      "step": 1171
    },
    {
      "epoch": 1.4992399391951357,
      "grad_norm": 0.6210200190544128,
      "learning_rate": 3.357398444386562e-05,
      "loss": 0.7257,
      "step": 1172
    },
    {
      "epoch": 1.5005200416033282,
      "grad_norm": 0.6220295429229736,
      "learning_rate": 3.356255938846234e-05,
      "loss": 0.7261,
      "step": 1173
    },
    {
      "epoch": 1.501800144011521,
      "grad_norm": 0.6124227046966553,
      "learning_rate": 3.3551126396778004e-05,
      "loss": 0.7632,
      "step": 1174
    },
    {
      "epoch": 1.5030802464197137,
      "grad_norm": 0.6557436585426331,
      "learning_rate": 3.3539685476659964e-05,
      "loss": 0.7005,
      "step": 1175
    },
    {
      "epoch": 1.5043603488279063,
      "grad_norm": 0.580775797367096,
      "learning_rate": 3.3528236635961004e-05,
      "loss": 0.7007,
      "step": 1176
    },
    {
      "epoch": 1.5056404512360988,
      "grad_norm": 0.5564892888069153,
      "learning_rate": 3.3516779882539355e-05,
      "loss": 0.7085,
      "step": 1177
    },
    {
      "epoch": 1.5069205536442916,
      "grad_norm": 0.607314944267273,
      "learning_rate": 3.3505315224258675e-05,
      "loss": 0.719,
      "step": 1178
    },
    {
      "epoch": 1.5082006560524843,
      "grad_norm": 0.5661142468452454,
      "learning_rate": 3.349384266898804e-05,
      "loss": 0.7111,
      "step": 1179
    },
    {
      "epoch": 1.5094807584606769,
      "grad_norm": 0.5816230773925781,
      "learning_rate": 3.348236222460196e-05,
      "loss": 0.7368,
      "step": 1180
    },
    {
      "epoch": 1.5107608608688694,
      "grad_norm": 0.6284036040306091,
      "learning_rate": 3.347087389898036e-05,
      "loss": 0.7318,
      "step": 1181
    },
    {
      "epoch": 1.5120409632770622,
      "grad_norm": 0.576706051826477,
      "learning_rate": 3.3459377700008564e-05,
      "loss": 0.7124,
      "step": 1182
    },
    {
      "epoch": 1.513321065685255,
      "grad_norm": 0.575192391872406,
      "learning_rate": 3.34478736355773e-05,
      "loss": 0.7159,
      "step": 1183
    },
    {
      "epoch": 1.5146011680934475,
      "grad_norm": 0.6095222234725952,
      "learning_rate": 3.343636171358271e-05,
      "loss": 0.7305,
      "step": 1184
    },
    {
      "epoch": 1.51588127050164,
      "grad_norm": 0.6324830055236816,
      "learning_rate": 3.3424841941926304e-05,
      "loss": 0.7361,
      "step": 1185
    },
    {
      "epoch": 1.5171613729098328,
      "grad_norm": 0.7316963076591492,
      "learning_rate": 3.3413314328515e-05,
      "loss": 0.7239,
      "step": 1186
    },
    {
      "epoch": 1.5184414753180255,
      "grad_norm": 0.6551388502120972,
      "learning_rate": 3.3401778881261104e-05,
      "loss": 0.7508,
      "step": 1187
    },
    {
      "epoch": 1.519721577726218,
      "grad_norm": 0.5726660490036011,
      "learning_rate": 3.3390235608082276e-05,
      "loss": 0.7083,
      "step": 1188
    },
    {
      "epoch": 1.5210016801344106,
      "grad_norm": 0.617388129234314,
      "learning_rate": 3.337868451690157e-05,
      "loss": 0.7125,
      "step": 1189
    },
    {
      "epoch": 1.5222817825426034,
      "grad_norm": 0.5886286497116089,
      "learning_rate": 3.336712561564738e-05,
      "loss": 0.7239,
      "step": 1190
    },
    {
      "epoch": 1.5235618849507961,
      "grad_norm": 0.6524567604064941,
      "learning_rate": 3.335555891225349e-05,
      "loss": 0.7474,
      "step": 1191
    },
    {
      "epoch": 1.5248419873589887,
      "grad_norm": 0.5842376947402954,
      "learning_rate": 3.334398441465901e-05,
      "loss": 0.7182,
      "step": 1192
    },
    {
      "epoch": 1.5261220897671812,
      "grad_norm": 0.7023752927780151,
      "learning_rate": 3.3332402130808435e-05,
      "loss": 0.7857,
      "step": 1193
    },
    {
      "epoch": 1.527402192175374,
      "grad_norm": 0.6121941804885864,
      "learning_rate": 3.332081206865158e-05,
      "loss": 0.7033,
      "step": 1194
    },
    {
      "epoch": 1.5286822945835667,
      "grad_norm": 0.6894810199737549,
      "learning_rate": 3.3309214236143585e-05,
      "loss": 0.7374,
      "step": 1195
    },
    {
      "epoch": 1.5299623969917593,
      "grad_norm": 0.6285365223884583,
      "learning_rate": 3.329760864124496e-05,
      "loss": 0.7546,
      "step": 1196
    },
    {
      "epoch": 1.531242499399952,
      "grad_norm": 0.6261930465698242,
      "learning_rate": 3.3285995291921515e-05,
      "loss": 0.7145,
      "step": 1197
    },
    {
      "epoch": 1.5325226018081448,
      "grad_norm": 0.6478322744369507,
      "learning_rate": 3.327437419614441e-05,
      "loss": 0.6797,
      "step": 1198
    },
    {
      "epoch": 1.5338027042163374,
      "grad_norm": 0.6763138175010681,
      "learning_rate": 3.326274536189008e-05,
      "loss": 0.7568,
      "step": 1199
    },
    {
      "epoch": 1.53508280662453,
      "grad_norm": 0.6303013563156128,
      "learning_rate": 3.3251108797140314e-05,
      "loss": 0.7257,
      "step": 1200
    },
    {
      "epoch": 1.53508280662453,
      "eval_loss": 0.011818733997642994,
      "eval_runtime": 10.8189,
      "eval_samples_per_second": 45.846,
      "eval_steps_per_second": 5.731,
      "step": 1200
    },
    {
      "epoch": 1.5363629090327227,
      "grad_norm": 0.586087703704834,
      "learning_rate": 3.323946450988217e-05,
      "loss": 0.7265,
      "step": 1201
    },
    {
      "epoch": 1.5376430114409154,
      "grad_norm": 0.6344309449195862,
      "learning_rate": 3.322781250810805e-05,
      "loss": 0.7154,
      "step": 1202
    },
    {
      "epoch": 1.538923113849108,
      "grad_norm": 0.690115213394165,
      "learning_rate": 3.32161527998156e-05,
      "loss": 0.7333,
      "step": 1203
    },
    {
      "epoch": 1.5402032162573005,
      "grad_norm": 0.7052741050720215,
      "learning_rate": 3.32044853930078e-05,
      "loss": 0.7678,
      "step": 1204
    },
    {
      "epoch": 1.5414833186654933,
      "grad_norm": 0.770147442817688,
      "learning_rate": 3.3192810295692885e-05,
      "loss": 0.6863,
      "step": 1205
    },
    {
      "epoch": 1.542763421073686,
      "grad_norm": 0.6130920052528381,
      "learning_rate": 3.3181127515884394e-05,
      "loss": 0.7185,
      "step": 1206
    },
    {
      "epoch": 1.5440435234818786,
      "grad_norm": 0.6697155833244324,
      "learning_rate": 3.316943706160111e-05,
      "loss": 0.7371,
      "step": 1207
    },
    {
      "epoch": 1.545323625890071,
      "grad_norm": 0.7463666200637817,
      "learning_rate": 3.3157738940867115e-05,
      "loss": 0.7144,
      "step": 1208
    },
    {
      "epoch": 1.5466037282982639,
      "grad_norm": 0.5947985053062439,
      "learning_rate": 3.314603316171171e-05,
      "loss": 0.738,
      "step": 1209
    },
    {
      "epoch": 1.5478838307064566,
      "grad_norm": 0.6855900883674622,
      "learning_rate": 3.3134319732169504e-05,
      "loss": 0.7157,
      "step": 1210
    },
    {
      "epoch": 1.5491639331146492,
      "grad_norm": 0.6108332872390747,
      "learning_rate": 3.312259866028032e-05,
      "loss": 0.7233,
      "step": 1211
    },
    {
      "epoch": 1.5504440355228417,
      "grad_norm": 0.686443030834198,
      "learning_rate": 3.3110869954089235e-05,
      "loss": 0.745,
      "step": 1212
    },
    {
      "epoch": 1.5517241379310345,
      "grad_norm": 0.564744234085083,
      "learning_rate": 3.3099133621646574e-05,
      "loss": 0.6993,
      "step": 1213
    },
    {
      "epoch": 1.5530042403392272,
      "grad_norm": 0.6294586062431335,
      "learning_rate": 3.30873896710079e-05,
      "loss": 0.7171,
      "step": 1214
    },
    {
      "epoch": 1.5542843427474198,
      "grad_norm": 0.6412985920906067,
      "learning_rate": 3.307563811023399e-05,
      "loss": 0.7504,
      "step": 1215
    },
    {
      "epoch": 1.5555644451556123,
      "grad_norm": 0.5940372943878174,
      "learning_rate": 3.306387894739085e-05,
      "loss": 0.7235,
      "step": 1216
    },
    {
      "epoch": 1.556844547563805,
      "grad_norm": 0.6947694420814514,
      "learning_rate": 3.3052112190549705e-05,
      "loss": 0.7364,
      "step": 1217
    },
    {
      "epoch": 1.5581246499719978,
      "grad_norm": 0.6133434772491455,
      "learning_rate": 3.3040337847786994e-05,
      "loss": 0.7348,
      "step": 1218
    },
    {
      "epoch": 1.5594047523801904,
      "grad_norm": 0.629682183265686,
      "learning_rate": 3.302855592718436e-05,
      "loss": 0.6679,
      "step": 1219
    },
    {
      "epoch": 1.560684854788383,
      "grad_norm": 0.6369724273681641,
      "learning_rate": 3.301676643682866e-05,
      "loss": 0.6835,
      "step": 1220
    },
    {
      "epoch": 1.5619649571965757,
      "grad_norm": 0.7039076089859009,
      "learning_rate": 3.3004969384811915e-05,
      "loss": 0.7356,
      "step": 1221
    },
    {
      "epoch": 1.5632450596047684,
      "grad_norm": 0.6234503388404846,
      "learning_rate": 3.299316477923137e-05,
      "loss": 0.7322,
      "step": 1222
    },
    {
      "epoch": 1.564525162012961,
      "grad_norm": 0.701338529586792,
      "learning_rate": 3.298135262818944e-05,
      "loss": 0.7168,
      "step": 1223
    },
    {
      "epoch": 1.5658052644211535,
      "grad_norm": 0.6774275898933411,
      "learning_rate": 3.296953293979372e-05,
      "loss": 0.7247,
      "step": 1224
    },
    {
      "epoch": 1.5670853668293463,
      "grad_norm": 0.6598162055015564,
      "learning_rate": 3.295770572215697e-05,
      "loss": 0.7224,
      "step": 1225
    },
    {
      "epoch": 1.568365469237539,
      "grad_norm": 0.6808010339736938,
      "learning_rate": 3.2945870983397144e-05,
      "loss": 0.6826,
      "step": 1226
    },
    {
      "epoch": 1.5696455716457316,
      "grad_norm": 0.6423587799072266,
      "learning_rate": 3.293402873163732e-05,
      "loss": 0.6742,
      "step": 1227
    },
    {
      "epoch": 1.5709256740539244,
      "grad_norm": 0.6605172753334045,
      "learning_rate": 3.292217897500577e-05,
      "loss": 0.7093,
      "step": 1228
    },
    {
      "epoch": 1.5722057764621171,
      "grad_norm": 0.6553050875663757,
      "learning_rate": 3.29103217216359e-05,
      "loss": 0.6686,
      "step": 1229
    },
    {
      "epoch": 1.5734858788703097,
      "grad_norm": 0.64751136302948,
      "learning_rate": 3.289845697966625e-05,
      "loss": 0.7487,
      "step": 1230
    },
    {
      "epoch": 1.5734858788703097,
      "eval_loss": 0.011799683794379234,
      "eval_runtime": 10.819,
      "eval_samples_per_second": 45.845,
      "eval_steps_per_second": 5.731,
      "step": 1230
    },
    {
      "epoch": 1.5747659812785022,
      "grad_norm": 0.6843473315238953,
      "learning_rate": 3.288658475724052e-05,
      "loss": 0.7258,
      "step": 1231
    },
    {
      "epoch": 1.576046083686695,
      "grad_norm": 0.6456937193870544,
      "learning_rate": 3.2874705062507534e-05,
      "loss": 0.6962,
      "step": 1232
    },
    {
      "epoch": 1.5773261860948877,
      "grad_norm": 0.6985472440719604,
      "learning_rate": 3.286281790362125e-05,
      "loss": 0.6923,
      "step": 1233
    },
    {
      "epoch": 1.5786062885030803,
      "grad_norm": 0.650560736656189,
      "learning_rate": 3.285092328874076e-05,
      "loss": 0.7315,
      "step": 1234
    },
    {
      "epoch": 1.5798863909112728,
      "grad_norm": 0.6405664086341858,
      "learning_rate": 3.283902122603023e-05,
      "loss": 0.7038,
      "step": 1235
    },
    {
      "epoch": 1.5811664933194656,
      "grad_norm": 0.7400343418121338,
      "learning_rate": 3.2827111723659e-05,
      "loss": 0.7304,
      "step": 1236
    },
    {
      "epoch": 1.5824465957276583,
      "grad_norm": 0.6203374862670898,
      "learning_rate": 3.2815194789801466e-05,
      "loss": 0.7427,
      "step": 1237
    },
    {
      "epoch": 1.5837266981358509,
      "grad_norm": 0.6931282877922058,
      "learning_rate": 3.2803270432637144e-05,
      "loss": 0.7121,
      "step": 1238
    },
    {
      "epoch": 1.5850068005440434,
      "grad_norm": 0.6139877438545227,
      "learning_rate": 3.279133866035065e-05,
      "loss": 0.7534,
      "step": 1239
    },
    {
      "epoch": 1.5862869029522362,
      "grad_norm": 0.6333911418914795,
      "learning_rate": 3.277939948113169e-05,
      "loss": 0.716,
      "step": 1240
    },
    {
      "epoch": 1.587567005360429,
      "grad_norm": 0.5837564468383789,
      "learning_rate": 3.276745290317504e-05,
      "loss": 0.7123,
      "step": 1241
    },
    {
      "epoch": 1.5888471077686215,
      "grad_norm": 0.6742558479309082,
      "learning_rate": 3.275549893468057e-05,
      "loss": 0.7269,
      "step": 1242
    },
    {
      "epoch": 1.590127210176814,
      "grad_norm": 0.6741624474525452,
      "learning_rate": 3.27435375838532e-05,
      "loss": 0.6949,
      "step": 1243
    },
    {
      "epoch": 1.5914073125850068,
      "grad_norm": 0.6758027672767639,
      "learning_rate": 3.273156885890295e-05,
      "loss": 0.7519,
      "step": 1244
    },
    {
      "epoch": 1.5926874149931995,
      "grad_norm": 0.6422121524810791,
      "learning_rate": 3.271959276804488e-05,
      "loss": 0.7358,
      "step": 1245
    },
    {
      "epoch": 1.593967517401392,
      "grad_norm": 0.52058345079422,
      "learning_rate": 3.27076093194991e-05,
      "loss": 0.6998,
      "step": 1246
    },
    {
      "epoch": 1.5952476198095846,
      "grad_norm": 0.7028553485870361,
      "learning_rate": 3.269561852149079e-05,
      "loss": 0.7796,
      "step": 1247
    },
    {
      "epoch": 1.5965277222177774,
      "grad_norm": 0.6633913516998291,
      "learning_rate": 3.268362038225017e-05,
      "loss": 0.7342,
      "step": 1248
    },
    {
      "epoch": 1.5978078246259702,
      "grad_norm": 0.6191238760948181,
      "learning_rate": 3.2671614910012475e-05,
      "loss": 0.7369,
      "step": 1249
    },
    {
      "epoch": 1.5990879270341627,
      "grad_norm": 0.6972527503967285,
      "learning_rate": 3.2659602113018e-05,
      "loss": 0.7614,
      "step": 1250
    },
    {
      "epoch": 1.6003680294423552,
      "grad_norm": 0.6203116178512573,
      "learning_rate": 3.264758199951206e-05,
      "loss": 0.7246,
      "step": 1251
    },
    {
      "epoch": 1.601648131850548,
      "grad_norm": 0.6659989356994629,
      "learning_rate": 3.2635554577744996e-05,
      "loss": 0.7388,
      "step": 1252
    },
    {
      "epoch": 1.6029282342587408,
      "grad_norm": 0.6051566004753113,
      "learning_rate": 3.2623519855972156e-05,
      "loss": 0.7021,
      "step": 1253
    },
    {
      "epoch": 1.6042083366669333,
      "grad_norm": 0.5739110112190247,
      "learning_rate": 3.26114778424539e-05,
      "loss": 0.7295,
      "step": 1254
    },
    {
      "epoch": 1.605488439075126,
      "grad_norm": 0.685315728187561,
      "learning_rate": 3.259942854545559e-05,
      "loss": 0.7476,
      "step": 1255
    },
    {
      "epoch": 1.6067685414833188,
      "grad_norm": 0.5766253471374512,
      "learning_rate": 3.258737197324761e-05,
      "loss": 0.7064,
      "step": 1256
    },
    {
      "epoch": 1.6080486438915114,
      "grad_norm": 0.6512503027915955,
      "learning_rate": 3.2575308134105314e-05,
      "loss": 0.7204,
      "step": 1257
    },
    {
      "epoch": 1.609328746299704,
      "grad_norm": 0.5989083051681519,
      "learning_rate": 3.256323703630904e-05,
      "loss": 0.7184,
      "step": 1258
    },
    {
      "epoch": 1.6106088487078967,
      "grad_norm": 0.5565375685691833,
      "learning_rate": 3.255115868814413e-05,
      "loss": 0.6673,
      "step": 1259
    },
    {
      "epoch": 1.6118889511160894,
      "grad_norm": 0.5859665274620056,
      "learning_rate": 3.253907309790088e-05,
      "loss": 0.7342,
      "step": 1260
    },
    {
      "epoch": 1.6118889511160894,
      "eval_loss": 0.011774944141507149,
      "eval_runtime": 10.813,
      "eval_samples_per_second": 45.871,
      "eval_steps_per_second": 5.734,
      "step": 1260
    },
    {
      "epoch": 1.613169053524282,
      "grad_norm": 0.5405060052871704,
      "learning_rate": 3.252698027387458e-05,
      "loss": 0.699,
      "step": 1261
    },
    {
      "epoch": 1.6144491559324745,
      "grad_norm": 0.5568755865097046,
      "learning_rate": 3.2514880224365456e-05,
      "loss": 0.6908,
      "step": 1262
    },
    {
      "epoch": 1.6157292583406673,
      "grad_norm": 0.6005760431289673,
      "learning_rate": 3.250277295767873e-05,
      "loss": 0.7013,
      "step": 1263
    },
    {
      "epoch": 1.61700936074886,
      "grad_norm": 0.5673319101333618,
      "learning_rate": 3.249065848212455e-05,
      "loss": 0.7145,
      "step": 1264
    },
    {
      "epoch": 1.6182894631570526,
      "grad_norm": 0.6222371459007263,
      "learning_rate": 3.247853680601802e-05,
      "loss": 0.6893,
      "step": 1265
    },
    {
      "epoch": 1.6195695655652451,
      "grad_norm": 0.5672705769538879,
      "learning_rate": 3.246640793767918e-05,
      "loss": 0.7074,
      "step": 1266
    },
    {
      "epoch": 1.6208496679734379,
      "grad_norm": 0.6348983645439148,
      "learning_rate": 3.245427188543302e-05,
      "loss": 0.7155,
      "step": 1267
    },
    {
      "epoch": 1.6221297703816306,
      "grad_norm": 0.6012302041053772,
      "learning_rate": 3.2442128657609454e-05,
      "loss": 0.721,
      "step": 1268
    },
    {
      "epoch": 1.6234098727898232,
      "grad_norm": 0.6491580605506897,
      "learning_rate": 3.2429978262543316e-05,
      "loss": 0.7705,
      "step": 1269
    },
    {
      "epoch": 1.6246899751980157,
      "grad_norm": 0.5992140173912048,
      "learning_rate": 3.241782070857436e-05,
      "loss": 0.7396,
      "step": 1270
    },
    {
      "epoch": 1.6259700776062085,
      "grad_norm": 0.5799919366836548,
      "learning_rate": 3.240565600404727e-05,
      "loss": 0.6946,
      "step": 1271
    },
    {
      "epoch": 1.6272501800144012,
      "grad_norm": 0.6085742115974426,
      "learning_rate": 3.239348415731163e-05,
      "loss": 0.7173,
      "step": 1272
    },
    {
      "epoch": 1.6285302824225938,
      "grad_norm": 0.6014842391014099,
      "learning_rate": 3.23813051767219e-05,
      "loss": 0.7014,
      "step": 1273
    },
    {
      "epoch": 1.6298103848307863,
      "grad_norm": 0.5858080983161926,
      "learning_rate": 3.236911907063747e-05,
      "loss": 0.6868,
      "step": 1274
    },
    {
      "epoch": 1.631090487238979,
      "grad_norm": 0.6003361940383911,
      "learning_rate": 3.235692584742262e-05,
      "loss": 0.7129,
      "step": 1275
    },
    {
      "epoch": 1.6323705896471719,
      "grad_norm": 0.5887556076049805,
      "learning_rate": 3.2344725515446495e-05,
      "loss": 0.7461,
      "step": 1276
    },
    {
      "epoch": 1.6336506920553644,
      "grad_norm": 0.617440402507782,
      "learning_rate": 3.233251808308312e-05,
      "loss": 0.6828,
      "step": 1277
    },
    {
      "epoch": 1.634930794463557,
      "grad_norm": 0.6037922501564026,
      "learning_rate": 3.232030355871143e-05,
      "loss": 0.7335,
      "step": 1278
    },
    {
      "epoch": 1.6362108968717497,
      "grad_norm": 0.5489148497581482,
      "learning_rate": 3.2308081950715165e-05,
      "loss": 0.6934,
      "step": 1279
    },
    {
      "epoch": 1.6374909992799425,
      "grad_norm": 0.6495527029037476,
      "learning_rate": 3.229585326748298e-05,
      "loss": 0.7198,
      "step": 1280
    },
    {
      "epoch": 1.638771101688135,
      "grad_norm": 0.5367933511734009,
      "learning_rate": 3.228361751740837e-05,
      "loss": 0.7092,
      "step": 1281
    },
    {
      "epoch": 1.6400512040963275,
      "grad_norm": 0.6641536951065063,
      "learning_rate": 3.2271374708889686e-05,
      "loss": 0.7325,
      "step": 1282
    },
    {
      "epoch": 1.6413313065045205,
      "grad_norm": 0.562475323677063,
      "learning_rate": 3.22591248503301e-05,
      "loss": 0.6978,
      "step": 1283
    },
    {
      "epoch": 1.642611408912713,
      "grad_norm": 0.6081634163856506,
      "learning_rate": 3.224686795013765e-05,
      "loss": 0.7034,
      "step": 1284
    },
    {
      "epoch": 1.6438915113209056,
      "grad_norm": 0.6050977110862732,
      "learning_rate": 3.223460401672518e-05,
      "loss": 0.7237,
      "step": 1285
    },
    {
      "epoch": 1.6451716137290984,
      "grad_norm": 0.6373354196548462,
      "learning_rate": 3.222233305851041e-05,
      "loss": 0.6897,
      "step": 1286
    },
    {
      "epoch": 1.6464517161372911,
      "grad_norm": 0.6571948528289795,
      "learning_rate": 3.221005508391583e-05,
      "loss": 0.7214,
      "step": 1287
    },
    {
      "epoch": 1.6477318185454837,
      "grad_norm": 0.6069954633712769,
      "learning_rate": 3.219777010136876e-05,
      "loss": 0.6992,
      "step": 1288
    },
    {
      "epoch": 1.6490119209536762,
      "grad_norm": 0.675748884677887,
      "learning_rate": 3.218547811930136e-05,
      "loss": 0.722,
      "step": 1289
    },
    {
      "epoch": 1.650292023361869,
      "grad_norm": 0.6166329383850098,
      "learning_rate": 3.2173179146150536e-05,
      "loss": 0.709,
      "step": 1290
    },
    {
      "epoch": 1.650292023361869,
      "eval_loss": 0.011765520088374615,
      "eval_runtime": 10.8236,
      "eval_samples_per_second": 45.826,
      "eval_steps_per_second": 5.728,
      "step": 1290
    },
    {
      "epoch": 1.6515721257700617,
      "grad_norm": 0.5665214657783508,
      "learning_rate": 3.2160873190358056e-05,
      "loss": 0.7298,
      "step": 1291
    },
    {
      "epoch": 1.6528522281782543,
      "grad_norm": 0.6093210577964783,
      "learning_rate": 3.214856026037044e-05,
      "loss": 0.6962,
      "step": 1292
    },
    {
      "epoch": 1.6541323305864468,
      "grad_norm": 0.582926332950592,
      "learning_rate": 3.213624036463901e-05,
      "loss": 0.7092,
      "step": 1293
    },
    {
      "epoch": 1.6554124329946396,
      "grad_norm": 0.5293163657188416,
      "learning_rate": 3.212391351161987e-05,
      "loss": 0.7229,
      "step": 1294
    },
    {
      "epoch": 1.6566925354028323,
      "grad_norm": 0.6089814901351929,
      "learning_rate": 3.211157970977388e-05,
      "loss": 0.7248,
      "step": 1295
    },
    {
      "epoch": 1.6579726378110249,
      "grad_norm": 0.6263851523399353,
      "learning_rate": 3.2099238967566694e-05,
      "loss": 0.7146,
      "step": 1296
    },
    {
      "epoch": 1.6592527402192174,
      "grad_norm": 0.6339204907417297,
      "learning_rate": 3.208689129346872e-05,
      "loss": 0.696,
      "step": 1297
    },
    {
      "epoch": 1.6605328426274102,
      "grad_norm": 0.641028642654419,
      "learning_rate": 3.207453669595513e-05,
      "loss": 0.7304,
      "step": 1298
    },
    {
      "epoch": 1.661812945035603,
      "grad_norm": 0.6381056904792786,
      "learning_rate": 3.206217518350584e-05,
      "loss": 0.7138,
      "step": 1299
    },
    {
      "epoch": 1.6630930474437955,
      "grad_norm": 0.5982184410095215,
      "learning_rate": 3.20498067646055e-05,
      "loss": 0.7047,
      "step": 1300
    },
    {
      "epoch": 1.664373149851988,
      "grad_norm": 0.6246302723884583,
      "learning_rate": 3.2037431447743524e-05,
      "loss": 0.7068,
      "step": 1301
    },
    {
      "epoch": 1.6656532522601808,
      "grad_norm": 0.6043433547019958,
      "learning_rate": 3.202504924141406e-05,
      "loss": 0.6991,
      "step": 1302
    },
    {
      "epoch": 1.6669333546683736,
      "grad_norm": 0.6269601583480835,
      "learning_rate": 3.2012660154115974e-05,
      "loss": 0.7147,
      "step": 1303
    },
    {
      "epoch": 1.668213457076566,
      "grad_norm": 0.6441345810890198,
      "learning_rate": 3.200026419435284e-05,
      "loss": 0.6935,
      "step": 1304
    },
    {
      "epoch": 1.6694935594847586,
      "grad_norm": 0.5528939962387085,
      "learning_rate": 3.198786137063299e-05,
      "loss": 0.6888,
      "step": 1305
    },
    {
      "epoch": 1.6707736618929514,
      "grad_norm": 0.7520976662635803,
      "learning_rate": 3.197545169146943e-05,
      "loss": 0.759,
      "step": 1306
    },
    {
      "epoch": 1.6720537643011442,
      "grad_norm": 0.6293416023254395,
      "learning_rate": 3.1963035165379884e-05,
      "loss": 0.7478,
      "step": 1307
    },
    {
      "epoch": 1.6733338667093367,
      "grad_norm": 0.634181022644043,
      "learning_rate": 3.195061180088679e-05,
      "loss": 0.6713,
      "step": 1308
    },
    {
      "epoch": 1.6746139691175292,
      "grad_norm": 0.6558919548988342,
      "learning_rate": 3.1938181606517244e-05,
      "loss": 0.7406,
      "step": 1309
    },
    {
      "epoch": 1.675894071525722,
      "grad_norm": 0.5954368710517883,
      "learning_rate": 3.1925744590803074e-05,
      "loss": 0.7172,
      "step": 1310
    },
    {
      "epoch": 1.6771741739339148,
      "grad_norm": 0.6629621386528015,
      "learning_rate": 3.191330076228076e-05,
      "loss": 0.7153,
      "step": 1311
    },
    {
      "epoch": 1.6784542763421073,
      "grad_norm": 0.6511289477348328,
      "learning_rate": 3.190085012949146e-05,
      "loss": 0.6955,
      "step": 1312
    },
    {
      "epoch": 1.6797343787503,
      "grad_norm": 0.6525564789772034,
      "learning_rate": 3.188839270098101e-05,
      "loss": 0.7245,
      "step": 1313
    },
    {
      "epoch": 1.6810144811584928,
      "grad_norm": 0.6946167349815369,
      "learning_rate": 3.1875928485299925e-05,
      "loss": 0.7295,
      "step": 1314
    },
    {
      "epoch": 1.6822945835666854,
      "grad_norm": 0.6380343437194824,
      "learning_rate": 3.186345749100335e-05,
      "loss": 0.7514,
      "step": 1315
    },
    {
      "epoch": 1.683574685974878,
      "grad_norm": 0.6703137159347534,
      "learning_rate": 3.185097972665109e-05,
      "loss": 0.7179,
      "step": 1316
    },
    {
      "epoch": 1.6848547883830707,
      "grad_norm": 0.6641091704368591,
      "learning_rate": 3.1838495200807615e-05,
      "loss": 0.7181,
      "step": 1317
    },
    {
      "epoch": 1.6861348907912634,
      "grad_norm": 0.6047271490097046,
      "learning_rate": 3.1826003922042016e-05,
      "loss": 0.7045,
      "step": 1318
    },
    {
      "epoch": 1.687414993199456,
      "grad_norm": 0.6599025130271912,
      "learning_rate": 3.1813505898928017e-05,
      "loss": 0.7344,
      "step": 1319
    },
    {
      "epoch": 1.6886950956076485,
      "grad_norm": 0.6088016629219055,
      "learning_rate": 3.1801001140044e-05,
      "loss": 0.692,
      "step": 1320
    },
    {
      "epoch": 1.6886950956076485,
      "eval_loss": 0.011746264062821865,
      "eval_runtime": 10.8084,
      "eval_samples_per_second": 45.89,
      "eval_steps_per_second": 5.736,
      "step": 1320
    },
    {
      "epoch": 1.6899751980158413,
      "grad_norm": 0.6381211280822754,
      "learning_rate": 3.178848965397294e-05,
      "loss": 0.7146,
      "step": 1321
    },
    {
      "epoch": 1.691255300424034,
      "grad_norm": 0.6384326815605164,
      "learning_rate": 3.177597144930244e-05,
      "loss": 0.7369,
      "step": 1322
    },
    {
      "epoch": 1.6925354028322266,
      "grad_norm": 0.6405301094055176,
      "learning_rate": 3.176344653462472e-05,
      "loss": 0.722,
      "step": 1323
    },
    {
      "epoch": 1.6938155052404191,
      "grad_norm": 0.6216254234313965,
      "learning_rate": 3.175091491853659e-05,
      "loss": 0.7354,
      "step": 1324
    },
    {
      "epoch": 1.695095607648612,
      "grad_norm": 0.6169127821922302,
      "learning_rate": 3.173837660963948e-05,
      "loss": 0.7308,
      "step": 1325
    },
    {
      "epoch": 1.6963757100568047,
      "grad_norm": 0.5891127586364746,
      "learning_rate": 3.172583161653939e-05,
      "loss": 0.7164,
      "step": 1326
    },
    {
      "epoch": 1.6976558124649972,
      "grad_norm": 0.6386512517929077,
      "learning_rate": 3.171327994784694e-05,
      "loss": 0.7332,
      "step": 1327
    },
    {
      "epoch": 1.6989359148731897,
      "grad_norm": 0.5602337718009949,
      "learning_rate": 3.17007216121773e-05,
      "loss": 0.7448,
      "step": 1328
    },
    {
      "epoch": 1.7002160172813825,
      "grad_norm": 0.6208088397979736,
      "learning_rate": 3.168815661815024e-05,
      "loss": 0.7512,
      "step": 1329
    },
    {
      "epoch": 1.7014961196895753,
      "grad_norm": 0.593197762966156,
      "learning_rate": 3.167558497439007e-05,
      "loss": 0.7195,
      "step": 1330
    },
    {
      "epoch": 1.7027762220977678,
      "grad_norm": 0.6581169962882996,
      "learning_rate": 3.166300668952571e-05,
      "loss": 0.7446,
      "step": 1331
    },
    {
      "epoch": 1.7040563245059603,
      "grad_norm": 0.5415921807289124,
      "learning_rate": 3.165042177219059e-05,
      "loss": 0.7026,
      "step": 1332
    },
    {
      "epoch": 1.705336426914153,
      "grad_norm": 0.6655652523040771,
      "learning_rate": 3.1637830231022727e-05,
      "loss": 0.7071,
      "step": 1333
    },
    {
      "epoch": 1.7066165293223459,
      "grad_norm": 0.626798689365387,
      "learning_rate": 3.162523207466468e-05,
      "loss": 0.6942,
      "step": 1334
    },
    {
      "epoch": 1.7078966317305384,
      "grad_norm": 0.6740243434906006,
      "learning_rate": 3.161262731176351e-05,
      "loss": 0.6963,
      "step": 1335
    },
    {
      "epoch": 1.709176734138731,
      "grad_norm": 0.6306195855140686,
      "learning_rate": 3.160001595097087e-05,
      "loss": 0.7314,
      "step": 1336
    },
    {
      "epoch": 1.7104568365469237,
      "grad_norm": 0.6224009990692139,
      "learning_rate": 3.1587398000942907e-05,
      "loss": 0.7144,
      "step": 1337
    },
    {
      "epoch": 1.7117369389551165,
      "grad_norm": 0.6219298839569092,
      "learning_rate": 3.15747734703403e-05,
      "loss": 0.7179,
      "step": 1338
    },
    {
      "epoch": 1.713017041363309,
      "grad_norm": 0.5488826632499695,
      "learning_rate": 3.156214236782825e-05,
      "loss": 0.7246,
      "step": 1339
    },
    {
      "epoch": 1.7142971437715018,
      "grad_norm": 0.635830283164978,
      "learning_rate": 3.154950470207644e-05,
      "loss": 0.7181,
      "step": 1340
    },
    {
      "epoch": 1.7155772461796945,
      "grad_norm": 0.61625736951828,
      "learning_rate": 3.1536860481759106e-05,
      "loss": 0.7714,
      "step": 1341
    },
    {
      "epoch": 1.716857348587887,
      "grad_norm": 0.5970157980918884,
      "learning_rate": 3.1524209715554945e-05,
      "loss": 0.7272,
      "step": 1342
    },
    {
      "epoch": 1.7181374509960796,
      "grad_norm": 0.6091165542602539,
      "learning_rate": 3.151155241214715e-05,
      "loss": 0.7184,
      "step": 1343
    },
    {
      "epoch": 1.7194175534042724,
      "grad_norm": 0.5880416035652161,
      "learning_rate": 3.149888858022343e-05,
      "loss": 0.7287,
      "step": 1344
    },
    {
      "epoch": 1.7206976558124651,
      "grad_norm": 0.5537950992584229,
      "learning_rate": 3.1486218228475936e-05,
      "loss": 0.7108,
      "step": 1345
    },
    {
      "epoch": 1.7219777582206577,
      "grad_norm": 0.6520135998725891,
      "learning_rate": 3.1473541365601316e-05,
      "loss": 0.7525,
      "step": 1346
    },
    {
      "epoch": 1.7232578606288502,
      "grad_norm": 0.6083378791809082,
      "learning_rate": 3.14608580003007e-05,
      "loss": 0.7385,
      "step": 1347
    },
    {
      "epoch": 1.724537963037043,
      "grad_norm": 0.5835146307945251,
      "learning_rate": 3.144816814127964e-05,
      "loss": 0.7452,
      "step": 1348
    },
    {
      "epoch": 1.7258180654452357,
      "grad_norm": 0.6272082328796387,
      "learning_rate": 3.143547179724819e-05,
      "loss": 0.7485,
      "step": 1349
    },
    {
      "epoch": 1.7270981678534283,
      "grad_norm": 0.6215714812278748,
      "learning_rate": 3.142276897692083e-05,
      "loss": 0.7566,
      "step": 1350
    },
    {
      "epoch": 1.7270981678534283,
      "eval_loss": 0.011728114448487759,
      "eval_runtime": 10.8488,
      "eval_samples_per_second": 45.719,
      "eval_steps_per_second": 5.715,
      "step": 1350
    },
    {
      "epoch": 1.7283782702616208,
      "grad_norm": 0.6123554706573486,
      "learning_rate": 3.1410059689016485e-05,
      "loss": 0.7301,
      "step": 1351
    },
    {
      "epoch": 1.7296583726698136,
      "grad_norm": 0.6339563727378845,
      "learning_rate": 3.1397343942258535e-05,
      "loss": 0.7093,
      "step": 1352
    },
    {
      "epoch": 1.7309384750780064,
      "grad_norm": 0.5856004953384399,
      "learning_rate": 3.138462174537477e-05,
      "loss": 0.7094,
      "step": 1353
    },
    {
      "epoch": 1.732218577486199,
      "grad_norm": 0.5875114798545837,
      "learning_rate": 3.1371893107097434e-05,
      "loss": 0.743,
      "step": 1354
    },
    {
      "epoch": 1.7334986798943914,
      "grad_norm": 0.6438591480255127,
      "learning_rate": 3.1359158036163165e-05,
      "loss": 0.7349,
      "step": 1355
    },
    {
      "epoch": 1.7347787823025842,
      "grad_norm": 0.5997960567474365,
      "learning_rate": 3.134641654131304e-05,
      "loss": 0.7195,
      "step": 1356
    },
    {
      "epoch": 1.736058884710777,
      "grad_norm": 0.6425419449806213,
      "learning_rate": 3.133366863129253e-05,
      "loss": 0.7038,
      "step": 1357
    },
    {
      "epoch": 1.7373389871189695,
      "grad_norm": 0.611407458782196,
      "learning_rate": 3.132091431485152e-05,
      "loss": 0.7101,
      "step": 1358
    },
    {
      "epoch": 1.738619089527162,
      "grad_norm": 0.6099973917007446,
      "learning_rate": 3.130815360074428e-05,
      "loss": 0.7212,
      "step": 1359
    },
    {
      "epoch": 1.7398991919353548,
      "grad_norm": 0.7355599403381348,
      "learning_rate": 3.1295386497729476e-05,
      "loss": 0.7393,
      "step": 1360
    },
    {
      "epoch": 1.7411792943435476,
      "grad_norm": 0.6639415621757507,
      "learning_rate": 3.1282613014570166e-05,
      "loss": 0.6859,
      "step": 1361
    },
    {
      "epoch": 1.74245939675174,
      "grad_norm": 0.6623911261558533,
      "learning_rate": 3.126983316003378e-05,
      "loss": 0.7371,
      "step": 1362
    },
    {
      "epoch": 1.7437394991599326,
      "grad_norm": 0.6310201287269592,
      "learning_rate": 3.125704694289212e-05,
      "loss": 0.7426,
      "step": 1363
    },
    {
      "epoch": 1.7450196015681254,
      "grad_norm": 0.5712582468986511,
      "learning_rate": 3.124425437192137e-05,
      "loss": 0.684,
      "step": 1364
    },
    {
      "epoch": 1.7462997039763182,
      "grad_norm": 0.6233215928077698,
      "learning_rate": 3.123145545590204e-05,
      "loss": 0.734,
      "step": 1365
    },
    {
      "epoch": 1.7475798063845107,
      "grad_norm": 0.5527714490890503,
      "learning_rate": 3.121865020361904e-05,
      "loss": 0.7468,
      "step": 1366
    },
    {
      "epoch": 1.7488599087927033,
      "grad_norm": 0.6191003322601318,
      "learning_rate": 3.12058386238616e-05,
      "loss": 0.7245,
      "step": 1367
    },
    {
      "epoch": 1.750140011200896,
      "grad_norm": 0.567948043346405,
      "learning_rate": 3.11930207254233e-05,
      "loss": 0.6857,
      "step": 1368
    },
    {
      "epoch": 1.7514201136090888,
      "grad_norm": 0.6133219599723816,
      "learning_rate": 3.118019651710207e-05,
      "loss": 0.7193,
      "step": 1369
    },
    {
      "epoch": 1.7527002160172813,
      "grad_norm": 0.6336676478385925,
      "learning_rate": 3.1167366007700135e-05,
      "loss": 0.754,
      "step": 1370
    },
    {
      "epoch": 1.753980318425474,
      "grad_norm": 0.595684826374054,
      "learning_rate": 3.1154529206024075e-05,
      "loss": 0.7211,
      "step": 1371
    },
    {
      "epoch": 1.7552604208336668,
      "grad_norm": 0.6025580167770386,
      "learning_rate": 3.1141686120884804e-05,
      "loss": 0.6876,
      "step": 1372
    },
    {
      "epoch": 1.7565405232418594,
      "grad_norm": 0.643245279788971,
      "learning_rate": 3.112883676109751e-05,
      "loss": 0.6822,
      "step": 1373
    },
    {
      "epoch": 1.757820625650052,
      "grad_norm": 0.5711572170257568,
      "learning_rate": 3.11159811354817e-05,
      "loss": 0.7839,
      "step": 1374
    },
    {
      "epoch": 1.7591007280582447,
      "grad_norm": 0.5706717371940613,
      "learning_rate": 3.110311925286119e-05,
      "loss": 0.7058,
      "step": 1375
    },
    {
      "epoch": 1.7603808304664375,
      "grad_norm": 0.6271959543228149,
      "learning_rate": 3.10902511220641e-05,
      "loss": 0.7339,
      "step": 1376
    },
    {
      "epoch": 1.76166093287463,
      "grad_norm": 0.586730420589447,
      "learning_rate": 3.1077376751922805e-05,
      "loss": 0.726,
      "step": 1377
    },
    {
      "epoch": 1.7629410352828225,
      "grad_norm": 0.664039134979248,
      "learning_rate": 3.1064496151274e-05,
      "loss": 0.7304,
      "step": 1378
    },
    {
      "epoch": 1.7642211376910153,
      "grad_norm": 0.6285074949264526,
      "learning_rate": 3.105160932895864e-05,
      "loss": 0.7559,
      "step": 1379
    },
    {
      "epoch": 1.765501240099208,
      "grad_norm": 0.5458385944366455,
      "learning_rate": 3.103871629382193e-05,
      "loss": 0.7084,
      "step": 1380
    },
    {
      "epoch": 1.765501240099208,
      "eval_loss": 0.011709917336702347,
      "eval_runtime": 10.8161,
      "eval_samples_per_second": 45.858,
      "eval_steps_per_second": 5.732,
      "step": 1380
    },
    {
      "epoch": 1.7667813425074006,
      "grad_norm": 0.6007266640663147,
      "learning_rate": 3.1025817054713385e-05,
      "loss": 0.717,
      "step": 1381
    },
    {
      "epoch": 1.7680614449155931,
      "grad_norm": 0.585239827632904,
      "learning_rate": 3.101291162048675e-05,
      "loss": 0.7251,
      "step": 1382
    },
    {
      "epoch": 1.769341547323786,
      "grad_norm": 0.5831015110015869,
      "learning_rate": 3.1e-05,
      "loss": 0.7547,
      "step": 1383
    },
    {
      "epoch": 1.7706216497319787,
      "grad_norm": 0.6495603322982788,
      "learning_rate": 3.098708220211541e-05,
      "loss": 0.7123,
      "step": 1384
    },
    {
      "epoch": 1.7719017521401712,
      "grad_norm": 0.5565499663352966,
      "learning_rate": 3.097415823569946e-05,
      "loss": 0.6802,
      "step": 1385
    },
    {
      "epoch": 1.7731818545483637,
      "grad_norm": 0.6719182729721069,
      "learning_rate": 3.0961228109622856e-05,
      "loss": 0.7222,
      "step": 1386
    },
    {
      "epoch": 1.7744619569565565,
      "grad_norm": 0.5855692028999329,
      "learning_rate": 3.094829183276056e-05,
      "loss": 0.7315,
      "step": 1387
    },
    {
      "epoch": 1.7757420593647493,
      "grad_norm": 0.6577022075653076,
      "learning_rate": 3.093534941399174e-05,
      "loss": 0.7452,
      "step": 1388
    },
    {
      "epoch": 1.7770221617729418,
      "grad_norm": 0.6627672910690308,
      "learning_rate": 3.092240086219978e-05,
      "loss": 0.7316,
      "step": 1389
    },
    {
      "epoch": 1.7783022641811344,
      "grad_norm": 0.6464568376541138,
      "learning_rate": 3.090944618627227e-05,
      "loss": 0.7092,
      "step": 1390
    },
    {
      "epoch": 1.7795823665893271,
      "grad_norm": 0.6542267799377441,
      "learning_rate": 3.0896485395101016e-05,
      "loss": 0.7473,
      "step": 1391
    },
    {
      "epoch": 1.7808624689975199,
      "grad_norm": 0.6975298523902893,
      "learning_rate": 3.088351849758201e-05,
      "loss": 0.7326,
      "step": 1392
    },
    {
      "epoch": 1.7821425714057124,
      "grad_norm": 0.6230096220970154,
      "learning_rate": 3.087054550261544e-05,
      "loss": 0.7036,
      "step": 1393
    },
    {
      "epoch": 1.783422673813905,
      "grad_norm": 0.6111627817153931,
      "learning_rate": 3.0857566419105676e-05,
      "loss": 0.7174,
      "step": 1394
    },
    {
      "epoch": 1.7847027762220977,
      "grad_norm": 0.6360964179039001,
      "learning_rate": 3.0844581255961275e-05,
      "loss": 0.7118,
      "step": 1395
    },
    {
      "epoch": 1.7859828786302905,
      "grad_norm": 0.616003155708313,
      "learning_rate": 3.083159002209497e-05,
      "loss": 0.7045,
      "step": 1396
    },
    {
      "epoch": 1.787262981038483,
      "grad_norm": 0.6242329478263855,
      "learning_rate": 3.0818592726423616e-05,
      "loss": 0.7129,
      "step": 1397
    },
    {
      "epoch": 1.7885430834466758,
      "grad_norm": 0.5756609439849854,
      "learning_rate": 3.080558937786831e-05,
      "loss": 0.6677,
      "step": 1398
    },
    {
      "epoch": 1.7898231858548685,
      "grad_norm": 0.5721912980079651,
      "learning_rate": 3.079257998535423e-05,
      "loss": 0.7129,
      "step": 1399
    },
    {
      "epoch": 1.791103288263061,
      "grad_norm": 0.618423581123352,
      "learning_rate": 3.077956455781073e-05,
      "loss": 0.7091,
      "step": 1400
    },
    {
      "epoch": 1.7923833906712536,
      "grad_norm": 0.5906749367713928,
      "learning_rate": 3.0766543104171326e-05,
      "loss": 0.7071,
      "step": 1401
    },
    {
      "epoch": 1.7936634930794464,
      "grad_norm": 0.6126925945281982,
      "learning_rate": 3.0753515633373634e-05,
      "loss": 0.7585,
      "step": 1402
    },
    {
      "epoch": 1.7949435954876392,
      "grad_norm": 0.5896530151367188,
      "learning_rate": 3.0740482154359434e-05,
      "loss": 0.7083,
      "step": 1403
    },
    {
      "epoch": 1.7962236978958317,
      "grad_norm": 0.5912639498710632,
      "learning_rate": 3.0727442676074595e-05,
      "loss": 0.6997,
      "step": 1404
    },
    {
      "epoch": 1.7975038003040242,
      "grad_norm": 0.5408932566642761,
      "learning_rate": 3.071439720746913e-05,
      "loss": 0.7064,
      "step": 1405
    },
    {
      "epoch": 1.798783902712217,
      "grad_norm": 0.6020301580429077,
      "learning_rate": 3.070134575749717e-05,
      "loss": 0.7263,
      "step": 1406
    },
    {
      "epoch": 1.8000640051204098,
      "grad_norm": 0.5722630023956299,
      "learning_rate": 3.0688288335116906e-05,
      "loss": 0.7221,
      "step": 1407
    },
    {
      "epoch": 1.8013441075286023,
      "grad_norm": 0.6015200614929199,
      "learning_rate": 3.067522494929069e-05,
      "loss": 0.6838,
      "step": 1408
    },
    {
      "epoch": 1.8026242099367948,
      "grad_norm": 0.6352903842926025,
      "learning_rate": 3.066215560898491e-05,
      "loss": 0.7169,
      "step": 1409
    },
    {
      "epoch": 1.8039043123449876,
      "grad_norm": 0.7391602993011475,
      "learning_rate": 3.064908032317009e-05,
      "loss": 0.7231,
      "step": 1410
    },
    {
      "epoch": 1.8039043123449876,
      "eval_loss": 0.011707385070621967,
      "eval_runtime": 10.8079,
      "eval_samples_per_second": 45.892,
      "eval_steps_per_second": 5.737,
      "step": 1410
    },
    {
      "epoch": 1.8051844147531804,
      "grad_norm": 0.6522735953330994,
      "learning_rate": 3.06359991008208e-05,
      "loss": 0.7032,
      "step": 1411
    },
    {
      "epoch": 1.806464517161373,
      "grad_norm": 0.5974452495574951,
      "learning_rate": 3.06229119509157e-05,
      "loss": 0.7052,
      "step": 1412
    },
    {
      "epoch": 1.8077446195695654,
      "grad_norm": 0.6504031419754028,
      "learning_rate": 3.060981888243752e-05,
      "loss": 0.6873,
      "step": 1413
    },
    {
      "epoch": 1.8090247219777582,
      "grad_norm": 0.5884857177734375,
      "learning_rate": 3.0596719904373036e-05,
      "loss": 0.7174,
      "step": 1414
    },
    {
      "epoch": 1.810304824385951,
      "grad_norm": 0.6082894206047058,
      "learning_rate": 3.058361502571312e-05,
      "loss": 0.7143,
      "step": 1415
    },
    {
      "epoch": 1.8115849267941435,
      "grad_norm": 0.609159529209137,
      "learning_rate": 3.057050425545263e-05,
      "loss": 0.7208,
      "step": 1416
    },
    {
      "epoch": 1.812865029202336,
      "grad_norm": 0.6024379134178162,
      "learning_rate": 3.0557387602590526e-05,
      "loss": 0.7147,
      "step": 1417
    },
    {
      "epoch": 1.8141451316105288,
      "grad_norm": 0.5601730942726135,
      "learning_rate": 3.054426507612977e-05,
      "loss": 0.6914,
      "step": 1418
    },
    {
      "epoch": 1.8154252340187216,
      "grad_norm": 0.5741651654243469,
      "learning_rate": 3.0531136685077397e-05,
      "loss": 0.7139,
      "step": 1419
    },
    {
      "epoch": 1.8167053364269141,
      "grad_norm": 0.5556854009628296,
      "learning_rate": 3.0518002438444406e-05,
      "loss": 0.717,
      "step": 1420
    },
    {
      "epoch": 1.8179854388351067,
      "grad_norm": 0.6205037832260132,
      "learning_rate": 3.0504862345245875e-05,
      "loss": 0.6988,
      "step": 1421
    },
    {
      "epoch": 1.8192655412432994,
      "grad_norm": 0.5890035629272461,
      "learning_rate": 3.0491716414500847e-05,
      "loss": 0.7116,
      "step": 1422
    },
    {
      "epoch": 1.8205456436514922,
      "grad_norm": 0.5740336179733276,
      "learning_rate": 3.0478564655232414e-05,
      "loss": 0.6982,
      "step": 1423
    },
    {
      "epoch": 1.8218257460596847,
      "grad_norm": 0.5904762744903564,
      "learning_rate": 3.0465407076467638e-05,
      "loss": 0.7251,
      "step": 1424
    },
    {
      "epoch": 1.8231058484678773,
      "grad_norm": 0.5810261964797974,
      "learning_rate": 3.045224368723758e-05,
      "loss": 0.7191,
      "step": 1425
    },
    {
      "epoch": 1.8243859508760703,
      "grad_norm": 0.6284822225570679,
      "learning_rate": 3.043907449657731e-05,
      "loss": 0.7284,
      "step": 1426
    },
    {
      "epoch": 1.8256660532842628,
      "grad_norm": 0.6124885678291321,
      "learning_rate": 3.042589951352585e-05,
      "loss": 0.7308,
      "step": 1427
    },
    {
      "epoch": 1.8269461556924553,
      "grad_norm": 0.6600826382637024,
      "learning_rate": 3.0412718747126218e-05,
      "loss": 0.7391,
      "step": 1428
    },
    {
      "epoch": 1.828226258100648,
      "grad_norm": 0.6918718814849854,
      "learning_rate": 3.0399532206425402e-05,
      "loss": 0.7395,
      "step": 1429
    },
    {
      "epoch": 1.8295063605088409,
      "grad_norm": 0.680570125579834,
      "learning_rate": 3.0386339900474337e-05,
      "loss": 0.7022,
      "step": 1430
    },
    {
      "epoch": 1.8307864629170334,
      "grad_norm": 0.6630198359489441,
      "learning_rate": 3.037314183832793e-05,
      "loss": 0.7269,
      "step": 1431
    },
    {
      "epoch": 1.832066565325226,
      "grad_norm": 0.6430322527885437,
      "learning_rate": 3.035993802904504e-05,
      "loss": 0.7219,
      "step": 1432
    },
    {
      "epoch": 1.8333466677334187,
      "grad_norm": 0.6303563714027405,
      "learning_rate": 3.0346728481688452e-05,
      "loss": 0.7107,
      "step": 1433
    },
    {
      "epoch": 1.8346267701416115,
      "grad_norm": 0.6679186224937439,
      "learning_rate": 3.0333513205324916e-05,
      "loss": 0.7277,
      "step": 1434
    },
    {
      "epoch": 1.835906872549804,
      "grad_norm": 0.6839330196380615,
      "learning_rate": 3.0320292209025092e-05,
      "loss": 0.6928,
      "step": 1435
    },
    {
      "epoch": 1.8371869749579965,
      "grad_norm": 0.5909537672996521,
      "learning_rate": 3.0307065501863576e-05,
      "loss": 0.6842,
      "step": 1436
    },
    {
      "epoch": 1.8384670773661893,
      "grad_norm": 0.6561462879180908,
      "learning_rate": 3.0293833092918886e-05,
      "loss": 0.6976,
      "step": 1437
    },
    {
      "epoch": 1.839747179774382,
      "grad_norm": 0.6374539732933044,
      "learning_rate": 3.0280594991273452e-05,
      "loss": 0.7088,
      "step": 1438
    },
    {
      "epoch": 1.8410272821825746,
      "grad_norm": 0.6158241033554077,
      "learning_rate": 3.0267351206013602e-05,
      "loss": 0.7154,
      "step": 1439
    },
    {
      "epoch": 1.8423073845907671,
      "grad_norm": 0.5730755925178528,
      "learning_rate": 3.025410174622958e-05,
      "loss": 0.7301,
      "step": 1440
    },
    {
      "epoch": 1.8423073845907671,
      "eval_loss": 0.011684227734804153,
      "eval_runtime": 10.8184,
      "eval_samples_per_second": 45.848,
      "eval_steps_per_second": 5.731,
      "step": 1440
    },
    {
      "epoch": 1.84358748699896,
      "grad_norm": 0.5943868160247803,
      "learning_rate": 3.0240846621015523e-05,
      "loss": 0.7165,
      "step": 1441
    },
    {
      "epoch": 1.8448675894071527,
      "grad_norm": 0.5782945156097412,
      "learning_rate": 3.022758583946943e-05,
      "loss": 0.691,
      "step": 1442
    },
    {
      "epoch": 1.8461476918153452,
      "grad_norm": 0.6446921825408936,
      "learning_rate": 3.0214319410693223e-05,
      "loss": 0.715,
      "step": 1443
    },
    {
      "epoch": 1.8474277942235378,
      "grad_norm": 0.5803105235099792,
      "learning_rate": 3.0201047343792672e-05,
      "loss": 0.7356,
      "step": 1444
    },
    {
      "epoch": 1.8487078966317305,
      "grad_norm": 0.6254404187202454,
      "learning_rate": 3.018776964787743e-05,
      "loss": 0.7254,
      "step": 1445
    },
    {
      "epoch": 1.8499879990399233,
      "grad_norm": 0.6068440675735474,
      "learning_rate": 3.017448633206101e-05,
      "loss": 0.708,
      "step": 1446
    },
    {
      "epoch": 1.8512681014481158,
      "grad_norm": 0.6016970276832581,
      "learning_rate": 3.0161197405460774e-05,
      "loss": 0.7177,
      "step": 1447
    },
    {
      "epoch": 1.8525482038563084,
      "grad_norm": 0.5711326003074646,
      "learning_rate": 3.0147902877197954e-05,
      "loss": 0.6752,
      "step": 1448
    },
    {
      "epoch": 1.8538283062645011,
      "grad_norm": 0.6343945860862732,
      "learning_rate": 3.0134602756397603e-05,
      "loss": 0.6957,
      "step": 1449
    },
    {
      "epoch": 1.8551084086726939,
      "grad_norm": 0.5972104668617249,
      "learning_rate": 3.0121297052188638e-05,
      "loss": 0.7336,
      "step": 1450
    },
    {
      "epoch": 1.8563885110808864,
      "grad_norm": 0.6045268774032593,
      "learning_rate": 3.0107985773703783e-05,
      "loss": 0.7555,
      "step": 1451
    },
    {
      "epoch": 1.857668613489079,
      "grad_norm": 0.62050461769104,
      "learning_rate": 3.0094668930079606e-05,
      "loss": 0.7215,
      "step": 1452
    },
    {
      "epoch": 1.8589487158972717,
      "grad_norm": 0.6099078059196472,
      "learning_rate": 3.0081346530456496e-05,
      "loss": 0.7602,
      "step": 1453
    },
    {
      "epoch": 1.8602288183054645,
      "grad_norm": 0.7151839733123779,
      "learning_rate": 3.006801858397864e-05,
      "loss": 0.7397,
      "step": 1454
    },
    {
      "epoch": 1.861508920713657,
      "grad_norm": 0.6388906836509705,
      "learning_rate": 3.005468509979404e-05,
      "loss": 0.7086,
      "step": 1455
    },
    {
      "epoch": 1.8627890231218498,
      "grad_norm": 0.7014924883842468,
      "learning_rate": 3.0041346087054508e-05,
      "loss": 0.7201,
      "step": 1456
    },
    {
      "epoch": 1.8640691255300426,
      "grad_norm": 0.6249627470970154,
      "learning_rate": 3.002800155491564e-05,
      "loss": 0.7459,
      "step": 1457
    },
    {
      "epoch": 1.865349227938235,
      "grad_norm": 0.6529703736305237,
      "learning_rate": 3.0014651512536826e-05,
      "loss": 0.7287,
      "step": 1458
    },
    {
      "epoch": 1.8666293303464276,
      "grad_norm": 0.7031069397926331,
      "learning_rate": 3.000129596908122e-05,
      "loss": 0.722,
      "step": 1459
    },
    {
      "epoch": 1.8679094327546204,
      "grad_norm": 0.6265938878059387,
      "learning_rate": 2.9987934933715786e-05,
      "loss": 0.7507,
      "step": 1460
    },
    {
      "epoch": 1.8691895351628132,
      "grad_norm": 0.698419451713562,
      "learning_rate": 2.997456841561123e-05,
      "loss": 0.7158,
      "step": 1461
    },
    {
      "epoch": 1.8704696375710057,
      "grad_norm": 0.5844295024871826,
      "learning_rate": 2.9961196423942026e-05,
      "loss": 0.7028,
      "step": 1462
    },
    {
      "epoch": 1.8717497399791982,
      "grad_norm": 0.623924732208252,
      "learning_rate": 2.9947818967886416e-05,
      "loss": 0.7285,
      "step": 1463
    },
    {
      "epoch": 1.873029842387391,
      "grad_norm": 0.6289079785346985,
      "learning_rate": 2.9934436056626386e-05,
      "loss": 0.7072,
      "step": 1464
    },
    {
      "epoch": 1.8743099447955838,
      "grad_norm": 0.6276214122772217,
      "learning_rate": 2.9921047699347653e-05,
      "loss": 0.7251,
      "step": 1465
    },
    {
      "epoch": 1.8755900472037763,
      "grad_norm": 0.5650027394294739,
      "learning_rate": 2.9907653905239703e-05,
      "loss": 0.692,
      "step": 1466
    },
    {
      "epoch": 1.8768701496119689,
      "grad_norm": 0.5894197225570679,
      "learning_rate": 2.9894254683495724e-05,
      "loss": 0.6986,
      "step": 1467
    },
    {
      "epoch": 1.8781502520201616,
      "grad_norm": 0.5582399368286133,
      "learning_rate": 2.988085004331264e-05,
      "loss": 0.7064,
      "step": 1468
    },
    {
      "epoch": 1.8794303544283544,
      "grad_norm": 0.6042588949203491,
      "learning_rate": 2.9867439993891104e-05,
      "loss": 0.7042,
      "step": 1469
    },
    {
      "epoch": 1.880710456836547,
      "grad_norm": 0.6372030377388,
      "learning_rate": 2.9854024544435463e-05,
      "loss": 0.7293,
      "step": 1470
    },
    {
      "epoch": 1.880710456836547,
      "eval_loss": 0.011673283763229847,
      "eval_runtime": 10.8394,
      "eval_samples_per_second": 45.759,
      "eval_steps_per_second": 5.72,
      "step": 1470
    },
    {
      "epoch": 1.8819905592447395,
      "grad_norm": 0.6210693717002869,
      "learning_rate": 2.9840603704153785e-05,
      "loss": 0.7323,
      "step": 1471
    },
    {
      "epoch": 1.8832706616529322,
      "grad_norm": 0.605437695980072,
      "learning_rate": 2.9827177482257834e-05,
      "loss": 0.6665,
      "step": 1472
    },
    {
      "epoch": 1.884550764061125,
      "grad_norm": 0.60888671875,
      "learning_rate": 2.9813745887963065e-05,
      "loss": 0.7256,
      "step": 1473
    },
    {
      "epoch": 1.8858308664693175,
      "grad_norm": 0.6271731853485107,
      "learning_rate": 2.980030893048862e-05,
      "loss": 0.7381,
      "step": 1474
    },
    {
      "epoch": 1.88711096887751,
      "grad_norm": 0.5811633467674255,
      "learning_rate": 2.9786866619057328e-05,
      "loss": 0.686,
      "step": 1475
    },
    {
      "epoch": 1.8883910712857028,
      "grad_norm": 0.6107152104377747,
      "learning_rate": 2.9773418962895695e-05,
      "loss": 0.7083,
      "step": 1476
    },
    {
      "epoch": 1.8896711736938956,
      "grad_norm": 0.6394736766815186,
      "learning_rate": 2.975996597123388e-05,
      "loss": 0.7104,
      "step": 1477
    },
    {
      "epoch": 1.8909512761020881,
      "grad_norm": 0.612646222114563,
      "learning_rate": 2.974650765330572e-05,
      "loss": 0.6754,
      "step": 1478
    },
    {
      "epoch": 1.8922313785102807,
      "grad_norm": 0.6603043675422668,
      "learning_rate": 2.9733044018348707e-05,
      "loss": 0.7313,
      "step": 1479
    },
    {
      "epoch": 1.8935114809184734,
      "grad_norm": 0.5434257984161377,
      "learning_rate": 2.971957507560397e-05,
      "loss": 0.7345,
      "step": 1480
    },
    {
      "epoch": 1.8947915833266662,
      "grad_norm": 0.6442853212356567,
      "learning_rate": 2.9706100834316282e-05,
      "loss": 0.7532,
      "step": 1481
    },
    {
      "epoch": 1.8960716857348587,
      "grad_norm": 0.6383693218231201,
      "learning_rate": 2.9692621303734073e-05,
      "loss": 0.7122,
      "step": 1482
    },
    {
      "epoch": 1.8973517881430515,
      "grad_norm": 0.6708807945251465,
      "learning_rate": 2.9679136493109392e-05,
      "loss": 0.7786,
      "step": 1483
    },
    {
      "epoch": 1.8986318905512443,
      "grad_norm": 0.6603577136993408,
      "learning_rate": 2.96656464116979e-05,
      "loss": 0.7659,
      "step": 1484
    },
    {
      "epoch": 1.8999119929594368,
      "grad_norm": 0.6571595072746277,
      "learning_rate": 2.9652151068758882e-05,
      "loss": 0.7279,
      "step": 1485
    },
    {
      "epoch": 1.9011920953676293,
      "grad_norm": 0.6065995693206787,
      "learning_rate": 2.963865047355525e-05,
      "loss": 0.7526,
      "step": 1486
    },
    {
      "epoch": 1.902472197775822,
      "grad_norm": 0.6256991624832153,
      "learning_rate": 2.9625144635353514e-05,
      "loss": 0.7241,
      "step": 1487
    },
    {
      "epoch": 1.9037523001840149,
      "grad_norm": 0.6634142398834229,
      "learning_rate": 2.9611633563423755e-05,
      "loss": 0.7721,
      "step": 1488
    },
    {
      "epoch": 1.9050324025922074,
      "grad_norm": 0.6693702936172485,
      "learning_rate": 2.959811726703969e-05,
      "loss": 0.7149,
      "step": 1489
    },
    {
      "epoch": 1.9063125050004,
      "grad_norm": 0.6248482465744019,
      "learning_rate": 2.9584595755478597e-05,
      "loss": 0.7005,
      "step": 1490
    },
    {
      "epoch": 1.9075926074085927,
      "grad_norm": 0.595628559589386,
      "learning_rate": 2.9571069038021324e-05,
      "loss": 0.7344,
      "step": 1491
    },
    {
      "epoch": 1.9088727098167855,
      "grad_norm": 0.6211913824081421,
      "learning_rate": 2.955753712395233e-05,
      "loss": 0.7195,
      "step": 1492
    },
    {
      "epoch": 1.910152812224978,
      "grad_norm": 0.5958669185638428,
      "learning_rate": 2.9544000022559598e-05,
      "loss": 0.678,
      "step": 1493
    },
    {
      "epoch": 1.9114329146331706,
      "grad_norm": 0.4980773627758026,
      "learning_rate": 2.9530457743134694e-05,
      "loss": 0.7005,
      "step": 1494
    },
    {
      "epoch": 1.9127130170413633,
      "grad_norm": 0.5755796432495117,
      "learning_rate": 2.9516910294972745e-05,
      "loss": 0.7069,
      "step": 1495
    },
    {
      "epoch": 1.913993119449556,
      "grad_norm": 0.5686249136924744,
      "learning_rate": 2.9503357687372402e-05,
      "loss": 0.7228,
      "step": 1496
    },
    {
      "epoch": 1.9152732218577486,
      "grad_norm": 0.5395983457565308,
      "learning_rate": 2.9489799929635875e-05,
      "loss": 0.6777,
      "step": 1497
    },
    {
      "epoch": 1.9165533242659412,
      "grad_norm": 0.595479428768158,
      "learning_rate": 2.947623703106891e-05,
      "loss": 0.7173,
      "step": 1498
    },
    {
      "epoch": 1.917833426674134,
      "grad_norm": 0.5761099457740784,
      "learning_rate": 2.9462669000980763e-05,
      "loss": 0.719,
      "step": 1499
    },
    {
      "epoch": 1.9191135290823267,
      "grad_norm": 0.5389598608016968,
      "learning_rate": 2.9449095848684233e-05,
      "loss": 0.7312,
      "step": 1500
    },
    {
      "epoch": 1.9191135290823267,
      "eval_loss": 0.01165055762976408,
      "eval_runtime": 10.8082,
      "eval_samples_per_second": 45.891,
      "eval_steps_per_second": 5.736,
      "step": 1500
    },
    {
      "epoch": 1.9203936314905192,
      "grad_norm": 0.5576955080032349,
      "learning_rate": 2.9435517583495633e-05,
      "loss": 0.7266,
      "step": 1501
    },
    {
      "epoch": 1.9216737338987118,
      "grad_norm": 0.5951018333435059,
      "learning_rate": 2.9421934214734768e-05,
      "loss": 0.7187,
      "step": 1502
    },
    {
      "epoch": 1.9229538363069045,
      "grad_norm": 0.524469792842865,
      "learning_rate": 2.9408345751724967e-05,
      "loss": 0.6702,
      "step": 1503
    },
    {
      "epoch": 1.9242339387150973,
      "grad_norm": 0.5927678942680359,
      "learning_rate": 2.9394752203793038e-05,
      "loss": 0.7109,
      "step": 1504
    },
    {
      "epoch": 1.9255140411232898,
      "grad_norm": 0.5567402243614197,
      "learning_rate": 2.938115358026929e-05,
      "loss": 0.7023,
      "step": 1505
    },
    {
      "epoch": 1.9267941435314824,
      "grad_norm": 0.5909476280212402,
      "learning_rate": 2.9367549890487513e-05,
      "loss": 0.698,
      "step": 1506
    },
    {
      "epoch": 1.9280742459396751,
      "grad_norm": 0.5594074726104736,
      "learning_rate": 2.9353941143784977e-05,
      "loss": 0.7081,
      "step": 1507
    },
    {
      "epoch": 1.929354348347868,
      "grad_norm": 0.6216226816177368,
      "learning_rate": 2.934032734950241e-05,
      "loss": 0.7221,
      "step": 1508
    },
    {
      "epoch": 1.9306344507560604,
      "grad_norm": 0.6358545422554016,
      "learning_rate": 2.9326708516984032e-05,
      "loss": 0.7614,
      "step": 1509
    },
    {
      "epoch": 1.931914553164253,
      "grad_norm": 0.5890107154846191,
      "learning_rate": 2.9313084655577486e-05,
      "loss": 0.6897,
      "step": 1510
    },
    {
      "epoch": 1.9331946555724457,
      "grad_norm": 0.6627918481826782,
      "learning_rate": 2.9299455774633894e-05,
      "loss": 0.7093,
      "step": 1511
    },
    {
      "epoch": 1.9344747579806385,
      "grad_norm": 0.566710352897644,
      "learning_rate": 2.928582188350781e-05,
      "loss": 0.7231,
      "step": 1512
    },
    {
      "epoch": 1.935754860388831,
      "grad_norm": 0.6416410803794861,
      "learning_rate": 2.927218299155723e-05,
      "loss": 0.7182,
      "step": 1513
    },
    {
      "epoch": 1.9370349627970238,
      "grad_norm": 0.5659756064414978,
      "learning_rate": 2.9258539108143588e-05,
      "loss": 0.6968,
      "step": 1514
    },
    {
      "epoch": 1.9383150652052166,
      "grad_norm": 0.5462925434112549,
      "learning_rate": 2.924489024263174e-05,
      "loss": 0.6949,
      "step": 1515
    },
    {
      "epoch": 1.939595167613409,
      "grad_norm": 0.5830070972442627,
      "learning_rate": 2.923123640438996e-05,
      "loss": 0.7338,
      "step": 1516
    },
    {
      "epoch": 1.9408752700216017,
      "grad_norm": 0.5738305449485779,
      "learning_rate": 2.921757760278992e-05,
      "loss": 0.7469,
      "step": 1517
    },
    {
      "epoch": 1.9421553724297944,
      "grad_norm": 0.5708524584770203,
      "learning_rate": 2.9203913847206737e-05,
      "loss": 0.7122,
      "step": 1518
    },
    {
      "epoch": 1.9434354748379872,
      "grad_norm": 0.5795454382896423,
      "learning_rate": 2.9190245147018892e-05,
      "loss": 0.7516,
      "step": 1519
    },
    {
      "epoch": 1.9447155772461797,
      "grad_norm": 0.5814696550369263,
      "learning_rate": 2.9176571511608278e-05,
      "loss": 0.7455,
      "step": 1520
    },
    {
      "epoch": 1.9459956796543723,
      "grad_norm": 0.576891303062439,
      "learning_rate": 2.9162892950360175e-05,
      "loss": 0.7284,
      "step": 1521
    },
    {
      "epoch": 1.947275782062565,
      "grad_norm": 0.5155112147331238,
      "learning_rate": 2.914920947266324e-05,
      "loss": 0.7182,
      "step": 1522
    },
    {
      "epoch": 1.9485558844707578,
      "grad_norm": 0.5739624500274658,
      "learning_rate": 2.9135521087909494e-05,
      "loss": 0.7504,
      "step": 1523
    },
    {
      "epoch": 1.9498359868789503,
      "grad_norm": 0.4973497986793518,
      "learning_rate": 2.9121827805494347e-05,
      "loss": 0.6832,
      "step": 1524
    },
    {
      "epoch": 1.9511160892871429,
      "grad_norm": 0.5796115398406982,
      "learning_rate": 2.910812963481656e-05,
      "loss": 0.7019,
      "step": 1525
    },
    {
      "epoch": 1.9523961916953356,
      "grad_norm": 0.5252601504325867,
      "learning_rate": 2.9094426585278247e-05,
      "loss": 0.6923,
      "step": 1526
    },
    {
      "epoch": 1.9536762941035284,
      "grad_norm": 0.6089382171630859,
      "learning_rate": 2.908071866628487e-05,
      "loss": 0.6918,
      "step": 1527
    },
    {
      "epoch": 1.954956396511721,
      "grad_norm": 0.587431013584137,
      "learning_rate": 2.906700588724525e-05,
      "loss": 0.7439,
      "step": 1528
    },
    {
      "epoch": 1.9562364989199135,
      "grad_norm": 0.5854923129081726,
      "learning_rate": 2.9053288257571514e-05,
      "loss": 0.7272,
      "step": 1529
    },
    {
      "epoch": 1.9575166013281062,
      "grad_norm": 0.611205518245697,
      "learning_rate": 2.9039565786679142e-05,
      "loss": 0.7163,
      "step": 1530
    },
    {
      "epoch": 1.9575166013281062,
      "eval_loss": 0.011649779044091702,
      "eval_runtime": 10.8121,
      "eval_samples_per_second": 45.875,
      "eval_steps_per_second": 5.734,
      "step": 1530
    },
    {
      "epoch": 1.958796703736299,
      "grad_norm": 0.5490525364875793,
      "learning_rate": 2.9025838483986928e-05,
      "loss": 0.7181,
      "step": 1531
    },
    {
      "epoch": 1.9600768061444915,
      "grad_norm": 0.6593901515007019,
      "learning_rate": 2.9012106358916994e-05,
      "loss": 0.7268,
      "step": 1532
    },
    {
      "epoch": 1.961356908552684,
      "grad_norm": 0.6241600513458252,
      "learning_rate": 2.8998369420894746e-05,
      "loss": 0.6878,
      "step": 1533
    },
    {
      "epoch": 1.9626370109608768,
      "grad_norm": 0.5867098569869995,
      "learning_rate": 2.8984627679348922e-05,
      "loss": 0.7205,
      "step": 1534
    },
    {
      "epoch": 1.9639171133690696,
      "grad_norm": 0.5838148593902588,
      "learning_rate": 2.8970881143711542e-05,
      "loss": 0.7081,
      "step": 1535
    },
    {
      "epoch": 1.9651972157772621,
      "grad_norm": 0.6911570429801941,
      "learning_rate": 2.8957129823417916e-05,
      "loss": 0.742,
      "step": 1536
    },
    {
      "epoch": 1.9664773181854547,
      "grad_norm": 0.5508971214294434,
      "learning_rate": 2.894337372790664e-05,
      "loss": 0.689,
      "step": 1537
    },
    {
      "epoch": 1.9677574205936474,
      "grad_norm": 0.691329300403595,
      "learning_rate": 2.8929612866619603e-05,
      "loss": 0.7386,
      "step": 1538
    },
    {
      "epoch": 1.9690375230018402,
      "grad_norm": 0.5353361368179321,
      "learning_rate": 2.8915847249001934e-05,
      "loss": 0.7202,
      "step": 1539
    },
    {
      "epoch": 1.9703176254100327,
      "grad_norm": 0.768527626991272,
      "learning_rate": 2.890207688450205e-05,
      "loss": 0.7442,
      "step": 1540
    },
    {
      "epoch": 1.9715977278182255,
      "grad_norm": 0.6241726875305176,
      "learning_rate": 2.8888301782571618e-05,
      "loss": 0.7467,
      "step": 1541
    },
    {
      "epoch": 1.9728778302264183,
      "grad_norm": 0.6419370770454407,
      "learning_rate": 2.8874521952665566e-05,
      "loss": 0.7197,
      "step": 1542
    },
    {
      "epoch": 1.9741579326346108,
      "grad_norm": 0.5724735856056213,
      "learning_rate": 2.886073740424205e-05,
      "loss": 0.716,
      "step": 1543
    },
    {
      "epoch": 1.9754380350428034,
      "grad_norm": 0.6717302799224854,
      "learning_rate": 2.8846948146762476e-05,
      "loss": 0.7077,
      "step": 1544
    },
    {
      "epoch": 1.9767181374509961,
      "grad_norm": 0.5468635559082031,
      "learning_rate": 2.8833154189691495e-05,
      "loss": 0.7065,
      "step": 1545
    },
    {
      "epoch": 1.9779982398591889,
      "grad_norm": 0.5952850580215454,
      "learning_rate": 2.8819355542496946e-05,
      "loss": 0.6833,
      "step": 1546
    },
    {
      "epoch": 1.9792783422673814,
      "grad_norm": 0.5658401846885681,
      "learning_rate": 2.880555221464993e-05,
      "loss": 0.686,
      "step": 1547
    },
    {
      "epoch": 1.980558444675574,
      "grad_norm": 0.6140429377555847,
      "learning_rate": 2.8791744215624737e-05,
      "loss": 0.71,
      "step": 1548
    },
    {
      "epoch": 1.9818385470837667,
      "grad_norm": 0.5493887662887573,
      "learning_rate": 2.8777931554898855e-05,
      "loss": 0.73,
      "step": 1549
    },
    {
      "epoch": 1.9831186494919595,
      "grad_norm": 0.608961820602417,
      "learning_rate": 2.8764114241952996e-05,
      "loss": 0.7189,
      "step": 1550
    },
    {
      "epoch": 1.984398751900152,
      "grad_norm": 0.6077911853790283,
      "learning_rate": 2.8750292286271047e-05,
      "loss": 0.717,
      "step": 1551
    },
    {
      "epoch": 1.9856788543083446,
      "grad_norm": 0.6050461530685425,
      "learning_rate": 2.87364656973401e-05,
      "loss": 0.721,
      "step": 1552
    },
    {
      "epoch": 1.9869589567165373,
      "grad_norm": 0.5760023593902588,
      "learning_rate": 2.8722634484650397e-05,
      "loss": 0.7347,
      "step": 1553
    },
    {
      "epoch": 1.98823905912473,
      "grad_norm": 0.5113067030906677,
      "learning_rate": 2.8708798657695384e-05,
      "loss": 0.684,
      "step": 1554
    },
    {
      "epoch": 1.9895191615329226,
      "grad_norm": 0.5542285442352295,
      "learning_rate": 2.8694958225971664e-05,
      "loss": 0.679,
      "step": 1555
    },
    {
      "epoch": 1.9907992639411152,
      "grad_norm": 0.5793948173522949,
      "learning_rate": 2.8681113198978984e-05,
      "loss": 0.7055,
      "step": 1556
    },
    {
      "epoch": 1.992079366349308,
      "grad_norm": 0.5892935395240784,
      "learning_rate": 2.866726358622028e-05,
      "loss": 0.7034,
      "step": 1557
    },
    {
      "epoch": 1.9933594687575007,
      "grad_norm": 0.5793341994285583,
      "learning_rate": 2.8653409397201596e-05,
      "loss": 0.699,
      "step": 1558
    },
    {
      "epoch": 1.9946395711656932,
      "grad_norm": 0.5980877876281738,
      "learning_rate": 2.8639550641432145e-05,
      "loss": 0.7303,
      "step": 1559
    },
    {
      "epoch": 1.9959196735738858,
      "grad_norm": 0.5784668326377869,
      "learning_rate": 2.8625687328424266e-05,
      "loss": 0.7412,
      "step": 1560
    },
    {
      "epoch": 1.9959196735738858,
      "eval_loss": 0.011627291329205036,
      "eval_runtime": 10.8138,
      "eval_samples_per_second": 45.867,
      "eval_steps_per_second": 5.733,
      "step": 1560
    },
    {
      "epoch": 1.9971997759820785,
      "grad_norm": 0.5964111089706421,
      "learning_rate": 2.8611819467693424e-05,
      "loss": 0.7598,
      "step": 1561
    },
    {
      "epoch": 1.9984798783902713,
      "grad_norm": 0.5467501282691956,
      "learning_rate": 2.8597947068758206e-05,
      "loss": 0.7211,
      "step": 1562
    },
    {
      "epoch": 1.9997599807984638,
      "grad_norm": 0.5621283650398254,
      "learning_rate": 2.858407014114032e-05,
      "loss": 0.7303,
      "step": 1563
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.8166683316230774,
      "learning_rate": 2.8570188694364573e-05,
      "loss": 0.5976,
      "step": 1564
    },
    {
      "epoch": 2.0012801024081925,
      "grad_norm": 0.6948720812797546,
      "learning_rate": 2.855630273795888e-05,
      "loss": 0.6735,
      "step": 1565
    },
    {
      "epoch": 2.0025602048163855,
      "grad_norm": 0.6384479403495789,
      "learning_rate": 2.8542412281454245e-05,
      "loss": 0.7041,
      "step": 1566
    },
    {
      "epoch": 2.003840307224578,
      "grad_norm": 0.5531007051467896,
      "learning_rate": 2.8528517334384775e-05,
      "loss": 0.655,
      "step": 1567
    },
    {
      "epoch": 2.0051204096327706,
      "grad_norm": 0.6195947527885437,
      "learning_rate": 2.8514617906287645e-05,
      "loss": 0.6428,
      "step": 1568
    },
    {
      "epoch": 2.006400512040963,
      "grad_norm": 0.6231208443641663,
      "learning_rate": 2.85007140067031e-05,
      "loss": 0.6981,
      "step": 1569
    },
    {
      "epoch": 2.007680614449156,
      "grad_norm": 0.5673173666000366,
      "learning_rate": 2.8486805645174486e-05,
      "loss": 0.693,
      "step": 1570
    },
    {
      "epoch": 2.0089607168573487,
      "grad_norm": 0.5763584971427917,
      "learning_rate": 2.847289283124817e-05,
      "loss": 0.6891,
      "step": 1571
    },
    {
      "epoch": 2.010240819265541,
      "grad_norm": 0.5962164998054504,
      "learning_rate": 2.8458975574473596e-05,
      "loss": 0.6773,
      "step": 1572
    },
    {
      "epoch": 2.0115209216737338,
      "grad_norm": 0.5427728295326233,
      "learning_rate": 2.8445053884403276e-05,
      "loss": 0.6789,
      "step": 1573
    },
    {
      "epoch": 2.0128010240819267,
      "grad_norm": 0.5714277625083923,
      "learning_rate": 2.8431127770592723e-05,
      "loss": 0.7139,
      "step": 1574
    },
    {
      "epoch": 2.0140811264901193,
      "grad_norm": 0.5537965893745422,
      "learning_rate": 2.841719724260052e-05,
      "loss": 0.6683,
      "step": 1575
    },
    {
      "epoch": 2.015361228898312,
      "grad_norm": 0.6056164503097534,
      "learning_rate": 2.840326230998827e-05,
      "loss": 0.6768,
      "step": 1576
    },
    {
      "epoch": 2.0166413313065044,
      "grad_norm": 0.6324855089187622,
      "learning_rate": 2.8389322982320592e-05,
      "loss": 0.7167,
      "step": 1577
    },
    {
      "epoch": 2.0179214337146973,
      "grad_norm": 0.6200981736183167,
      "learning_rate": 2.8375379269165128e-05,
      "loss": 0.686,
      "step": 1578
    },
    {
      "epoch": 2.01920153612289,
      "grad_norm": 0.5917149782180786,
      "learning_rate": 2.8361431180092534e-05,
      "loss": 0.6538,
      "step": 1579
    },
    {
      "epoch": 2.0204816385310824,
      "grad_norm": 0.6273303627967834,
      "learning_rate": 2.8347478724676462e-05,
      "loss": 0.6721,
      "step": 1580
    },
    {
      "epoch": 2.021761740939275,
      "grad_norm": 0.5992140769958496,
      "learning_rate": 2.8333521912493557e-05,
      "loss": 0.6772,
      "step": 1581
    },
    {
      "epoch": 2.023041843347468,
      "grad_norm": 0.6496381759643555,
      "learning_rate": 2.831956075312348e-05,
      "loss": 0.6576,
      "step": 1582
    },
    {
      "epoch": 2.0243219457556605,
      "grad_norm": 0.5352444052696228,
      "learning_rate": 2.830559525614884e-05,
      "loss": 0.6692,
      "step": 1583
    },
    {
      "epoch": 2.025602048163853,
      "grad_norm": 0.6139085292816162,
      "learning_rate": 2.8291625431155244e-05,
      "loss": 0.6756,
      "step": 1584
    },
    {
      "epoch": 2.0268821505720456,
      "grad_norm": 0.5375662446022034,
      "learning_rate": 2.8277651287731273e-05,
      "loss": 0.6947,
      "step": 1585
    },
    {
      "epoch": 2.0281622529802386,
      "grad_norm": 0.6130796670913696,
      "learning_rate": 2.8263672835468456e-05,
      "loss": 0.692,
      "step": 1586
    },
    {
      "epoch": 2.029442355388431,
      "grad_norm": 0.5520009398460388,
      "learning_rate": 2.82496900839613e-05,
      "loss": 0.6796,
      "step": 1587
    },
    {
      "epoch": 2.0307224577966236,
      "grad_norm": 0.6222859621047974,
      "learning_rate": 2.8235703042807246e-05,
      "loss": 0.6699,
      "step": 1588
    },
    {
      "epoch": 2.032002560204816,
      "grad_norm": 0.5752348899841309,
      "learning_rate": 2.822171172160667e-05,
      "loss": 0.6819,
      "step": 1589
    },
    {
      "epoch": 2.033282662613009,
      "grad_norm": 0.6140403747558594,
      "learning_rate": 2.8207716129962926e-05,
      "loss": 0.6978,
      "step": 1590
    },
    {
      "epoch": 2.033282662613009,
      "eval_loss": 0.011672087013721466,
      "eval_runtime": 10.8084,
      "eval_samples_per_second": 45.89,
      "eval_steps_per_second": 5.736,
      "step": 1590
    },
    {
      "epoch": 2.0345627650212017,
      "grad_norm": 0.6465091109275818,
      "learning_rate": 2.819371627748226e-05,
      "loss": 0.691,
      "step": 1591
    },
    {
      "epoch": 2.0358428674293942,
      "grad_norm": 0.5918066501617432,
      "learning_rate": 2.817971217377386e-05,
      "loss": 0.7131,
      "step": 1592
    },
    {
      "epoch": 2.0371229698375872,
      "grad_norm": 0.6345959901809692,
      "learning_rate": 2.8165703828449826e-05,
      "loss": 0.649,
      "step": 1593
    },
    {
      "epoch": 2.0384030722457798,
      "grad_norm": 0.6075990200042725,
      "learning_rate": 2.815169125112517e-05,
      "loss": 0.7051,
      "step": 1594
    },
    {
      "epoch": 2.0396831746539723,
      "grad_norm": 0.5795350670814514,
      "learning_rate": 2.813767445141781e-05,
      "loss": 0.7098,
      "step": 1595
    },
    {
      "epoch": 2.040963277062165,
      "grad_norm": 0.610453724861145,
      "learning_rate": 2.812365343894857e-05,
      "loss": 0.6697,
      "step": 1596
    },
    {
      "epoch": 2.042243379470358,
      "grad_norm": 0.5452365279197693,
      "learning_rate": 2.810962822334115e-05,
      "loss": 0.6554,
      "step": 1597
    },
    {
      "epoch": 2.0435234818785504,
      "grad_norm": 0.5816706418991089,
      "learning_rate": 2.8095598814222147e-05,
      "loss": 0.6841,
      "step": 1598
    },
    {
      "epoch": 2.044803584286743,
      "grad_norm": 0.5663377642631531,
      "learning_rate": 2.808156522122103e-05,
      "loss": 0.6726,
      "step": 1599
    },
    {
      "epoch": 2.0460836866949355,
      "grad_norm": 0.5797086358070374,
      "learning_rate": 2.806752745397015e-05,
      "loss": 0.6763,
      "step": 1600
    },
    {
      "epoch": 2.0473637891031284,
      "grad_norm": 0.5999887585639954,
      "learning_rate": 2.8053485522104706e-05,
      "loss": 0.7372,
      "step": 1601
    },
    {
      "epoch": 2.048643891511321,
      "grad_norm": 0.5794069170951843,
      "learning_rate": 2.803943943526276e-05,
      "loss": 0.6784,
      "step": 1602
    },
    {
      "epoch": 2.0499239939195135,
      "grad_norm": 0.6007727384567261,
      "learning_rate": 2.802538920308524e-05,
      "loss": 0.6691,
      "step": 1603
    },
    {
      "epoch": 2.051204096327706,
      "grad_norm": 0.632462739944458,
      "learning_rate": 2.801133483521591e-05,
      "loss": 0.7218,
      "step": 1604
    },
    {
      "epoch": 2.052484198735899,
      "grad_norm": 0.6296536922454834,
      "learning_rate": 2.799727634130137e-05,
      "loss": 0.6584,
      "step": 1605
    },
    {
      "epoch": 2.0537643011440916,
      "grad_norm": 0.5758532881736755,
      "learning_rate": 2.7983213730991055e-05,
      "loss": 0.6119,
      "step": 1606
    },
    {
      "epoch": 2.055044403552284,
      "grad_norm": 0.6606136560440063,
      "learning_rate": 2.7969147013937227e-05,
      "loss": 0.6784,
      "step": 1607
    },
    {
      "epoch": 2.0563245059604767,
      "grad_norm": 0.5617915391921997,
      "learning_rate": 2.7955076199794954e-05,
      "loss": 0.6509,
      "step": 1608
    },
    {
      "epoch": 2.0576046083686697,
      "grad_norm": 0.6208764910697937,
      "learning_rate": 2.794100129822214e-05,
      "loss": 0.6751,
      "step": 1609
    },
    {
      "epoch": 2.058884710776862,
      "grad_norm": 0.5615843534469604,
      "learning_rate": 2.7926922318879474e-05,
      "loss": 0.709,
      "step": 1610
    },
    {
      "epoch": 2.0601648131850547,
      "grad_norm": 0.6210365295410156,
      "learning_rate": 2.791283927143045e-05,
      "loss": 0.6928,
      "step": 1611
    },
    {
      "epoch": 2.0614449155932473,
      "grad_norm": 0.59096759557724,
      "learning_rate": 2.7898752165541365e-05,
      "loss": 0.6579,
      "step": 1612
    },
    {
      "epoch": 2.0627250180014403,
      "grad_norm": 0.6035891175270081,
      "learning_rate": 2.7884661010881284e-05,
      "loss": 0.6522,
      "step": 1613
    },
    {
      "epoch": 2.064005120409633,
      "grad_norm": 0.6002820730209351,
      "learning_rate": 2.7870565817122054e-05,
      "loss": 0.6636,
      "step": 1614
    },
    {
      "epoch": 2.0652852228178253,
      "grad_norm": 0.648118257522583,
      "learning_rate": 2.785646659393831e-05,
      "loss": 0.6769,
      "step": 1615
    },
    {
      "epoch": 2.066565325226018,
      "grad_norm": 0.6546078324317932,
      "learning_rate": 2.7842363351007443e-05,
      "loss": 0.679,
      "step": 1616
    },
    {
      "epoch": 2.067845427634211,
      "grad_norm": 0.63655024766922,
      "learning_rate": 2.7828256098009595e-05,
      "loss": 0.7468,
      "step": 1617
    },
    {
      "epoch": 2.0691255300424034,
      "grad_norm": 0.6780474781990051,
      "learning_rate": 2.7814144844627664e-05,
      "loss": 0.6864,
      "step": 1618
    },
    {
      "epoch": 2.070405632450596,
      "grad_norm": 0.5667698383331299,
      "learning_rate": 2.7800029600547312e-05,
      "loss": 0.7232,
      "step": 1619
    },
    {
      "epoch": 2.0716857348587885,
      "grad_norm": 0.6375884413719177,
      "learning_rate": 2.7785910375456914e-05,
      "loss": 0.7079,
      "step": 1620
    },
    {
      "epoch": 2.0716857348587885,
      "eval_loss": 0.011663525365293026,
      "eval_runtime": 10.8358,
      "eval_samples_per_second": 45.774,
      "eval_steps_per_second": 5.722,
      "step": 1620
    },
    {
      "epoch": 2.0729658372669815,
      "grad_norm": 0.5780451893806458,
      "learning_rate": 2.7771787179047588e-05,
      "loss": 0.6597,
      "step": 1621
    },
    {
      "epoch": 2.074245939675174,
      "grad_norm": 0.5665174126625061,
      "learning_rate": 2.7757660021013184e-05,
      "loss": 0.6903,
      "step": 1622
    },
    {
      "epoch": 2.0755260420833666,
      "grad_norm": 0.5880652666091919,
      "learning_rate": 2.7743528911050267e-05,
      "loss": 0.6782,
      "step": 1623
    },
    {
      "epoch": 2.0768061444915595,
      "grad_norm": 0.5618604421615601,
      "learning_rate": 2.7729393858858107e-05,
      "loss": 0.6943,
      "step": 1624
    },
    {
      "epoch": 2.078086246899752,
      "grad_norm": 0.5834721326828003,
      "learning_rate": 2.7715254874138696e-05,
      "loss": 0.6962,
      "step": 1625
    },
    {
      "epoch": 2.0793663493079446,
      "grad_norm": 0.5367355942726135,
      "learning_rate": 2.7701111966596704e-05,
      "loss": 0.6948,
      "step": 1626
    },
    {
      "epoch": 2.080646451716137,
      "grad_norm": 0.5970818996429443,
      "learning_rate": 2.768696514593952e-05,
      "loss": 0.6602,
      "step": 1627
    },
    {
      "epoch": 2.08192655412433,
      "grad_norm": 0.5148850679397583,
      "learning_rate": 2.7672814421877182e-05,
      "loss": 0.6558,
      "step": 1628
    },
    {
      "epoch": 2.0832066565325227,
      "grad_norm": 0.5768963694572449,
      "learning_rate": 2.7658659804122452e-05,
      "loss": 0.6896,
      "step": 1629
    },
    {
      "epoch": 2.0844867589407152,
      "grad_norm": 0.5947193503379822,
      "learning_rate": 2.7644501302390727e-05,
      "loss": 0.6897,
      "step": 1630
    },
    {
      "epoch": 2.0857668613489078,
      "grad_norm": 0.6065978407859802,
      "learning_rate": 2.7630338926400093e-05,
      "loss": 0.6903,
      "step": 1631
    },
    {
      "epoch": 2.0870469637571007,
      "grad_norm": 0.596855878829956,
      "learning_rate": 2.761617268587128e-05,
      "loss": 0.6834,
      "step": 1632
    },
    {
      "epoch": 2.0883270661652933,
      "grad_norm": 0.625764787197113,
      "learning_rate": 2.7602002590527684e-05,
      "loss": 0.6586,
      "step": 1633
    },
    {
      "epoch": 2.089607168573486,
      "grad_norm": 0.6117658019065857,
      "learning_rate": 2.758782865009533e-05,
      "loss": 0.6924,
      "step": 1634
    },
    {
      "epoch": 2.0908872709816784,
      "grad_norm": 0.5718591809272766,
      "learning_rate": 2.7573650874302905e-05,
      "loss": 0.6327,
      "step": 1635
    },
    {
      "epoch": 2.0921673733898714,
      "grad_norm": 0.6709867715835571,
      "learning_rate": 2.7559469272881708e-05,
      "loss": 0.676,
      "step": 1636
    },
    {
      "epoch": 2.093447475798064,
      "grad_norm": 0.6376213431358337,
      "learning_rate": 2.754528385556567e-05,
      "loss": 0.6964,
      "step": 1637
    },
    {
      "epoch": 2.0947275782062564,
      "grad_norm": 0.6515515446662903,
      "learning_rate": 2.7531094632091352e-05,
      "loss": 0.698,
      "step": 1638
    },
    {
      "epoch": 2.096007680614449,
      "grad_norm": 0.57508385181427,
      "learning_rate": 2.751690161219791e-05,
      "loss": 0.6692,
      "step": 1639
    },
    {
      "epoch": 2.097287783022642,
      "grad_norm": 0.6378449201583862,
      "learning_rate": 2.750270480562712e-05,
      "loss": 0.6721,
      "step": 1640
    },
    {
      "epoch": 2.0985678854308345,
      "grad_norm": 0.608546793460846,
      "learning_rate": 2.7488504222123342e-05,
      "loss": 0.684,
      "step": 1641
    },
    {
      "epoch": 2.099847987839027,
      "grad_norm": 0.644956648349762,
      "learning_rate": 2.7474299871433543e-05,
      "loss": 0.6897,
      "step": 1642
    },
    {
      "epoch": 2.1011280902472196,
      "grad_norm": 0.6527913808822632,
      "learning_rate": 2.7460091763307275e-05,
      "loss": 0.6999,
      "step": 1643
    },
    {
      "epoch": 2.1024081926554126,
      "grad_norm": 0.5840986371040344,
      "learning_rate": 2.7445879907496656e-05,
      "loss": 0.6827,
      "step": 1644
    },
    {
      "epoch": 2.103688295063605,
      "grad_norm": 0.6439254879951477,
      "learning_rate": 2.7431664313756394e-05,
      "loss": 0.6795,
      "step": 1645
    },
    {
      "epoch": 2.1049683974717976,
      "grad_norm": 0.6222280859947205,
      "learning_rate": 2.7417444991843746e-05,
      "loss": 0.6957,
      "step": 1646
    },
    {
      "epoch": 2.10624849987999,
      "grad_norm": 0.573788583278656,
      "learning_rate": 2.7403221951518533e-05,
      "loss": 0.6791,
      "step": 1647
    },
    {
      "epoch": 2.107528602288183,
      "grad_norm": 0.6004598140716553,
      "learning_rate": 2.738899520254314e-05,
      "loss": 0.6659,
      "step": 1648
    },
    {
      "epoch": 2.1088087046963757,
      "grad_norm": 0.5855998396873474,
      "learning_rate": 2.7374764754682482e-05,
      "loss": 0.6604,
      "step": 1649
    },
    {
      "epoch": 2.1100888071045683,
      "grad_norm": 0.6659582257270813,
      "learning_rate": 2.736053061770401e-05,
      "loss": 0.6638,
      "step": 1650
    },
    {
      "epoch": 2.1100888071045683,
      "eval_loss": 0.011645011603832245,
      "eval_runtime": 10.8273,
      "eval_samples_per_second": 45.81,
      "eval_steps_per_second": 5.726,
      "step": 1650
    },
    {
      "epoch": 2.111368909512761,
      "grad_norm": 0.6403827667236328,
      "learning_rate": 2.7346292801377737e-05,
      "loss": 0.6668,
      "step": 1651
    },
    {
      "epoch": 2.112649011920954,
      "grad_norm": 0.5976318717002869,
      "learning_rate": 2.7332051315476166e-05,
      "loss": 0.7139,
      "step": 1652
    },
    {
      "epoch": 2.1139291143291463,
      "grad_norm": 0.59450763463974,
      "learning_rate": 2.7317806169774325e-05,
      "loss": 0.643,
      "step": 1653
    },
    {
      "epoch": 2.115209216737339,
      "grad_norm": 0.6102283000946045,
      "learning_rate": 2.730355737404978e-05,
      "loss": 0.6299,
      "step": 1654
    },
    {
      "epoch": 2.116489319145532,
      "grad_norm": 0.6156714558601379,
      "learning_rate": 2.7289304938082583e-05,
      "loss": 0.6439,
      "step": 1655
    },
    {
      "epoch": 2.1177694215537244,
      "grad_norm": 0.660211980342865,
      "learning_rate": 2.727504887165527e-05,
      "loss": 0.6739,
      "step": 1656
    },
    {
      "epoch": 2.119049523961917,
      "grad_norm": 0.6008319854736328,
      "learning_rate": 2.7260789184552902e-05,
      "loss": 0.6474,
      "step": 1657
    },
    {
      "epoch": 2.1203296263701095,
      "grad_norm": 0.6325721740722656,
      "learning_rate": 2.7246525886562995e-05,
      "loss": 0.657,
      "step": 1658
    },
    {
      "epoch": 2.1216097287783025,
      "grad_norm": 0.6137465834617615,
      "learning_rate": 2.723225898747556e-05,
      "loss": 0.6779,
      "step": 1659
    },
    {
      "epoch": 2.122889831186495,
      "grad_norm": 0.582832932472229,
      "learning_rate": 2.721798849708308e-05,
      "loss": 0.6648,
      "step": 1660
    },
    {
      "epoch": 2.1241699335946875,
      "grad_norm": 0.5986790060997009,
      "learning_rate": 2.7203714425180497e-05,
      "loss": 0.6925,
      "step": 1661
    },
    {
      "epoch": 2.12545003600288,
      "grad_norm": 0.593242883682251,
      "learning_rate": 2.7189436781565218e-05,
      "loss": 0.6388,
      "step": 1662
    },
    {
      "epoch": 2.126730138411073,
      "grad_norm": 0.6194828748703003,
      "learning_rate": 2.717515557603709e-05,
      "loss": 0.71,
      "step": 1663
    },
    {
      "epoch": 2.1280102408192656,
      "grad_norm": 0.6205288171768188,
      "learning_rate": 2.716087081839841e-05,
      "loss": 0.6607,
      "step": 1664
    },
    {
      "epoch": 2.129290343227458,
      "grad_norm": 0.6458435654640198,
      "learning_rate": 2.7146582518453918e-05,
      "loss": 0.7057,
      "step": 1665
    },
    {
      "epoch": 2.1305704456356507,
      "grad_norm": 0.5268846750259399,
      "learning_rate": 2.7132290686010795e-05,
      "loss": 0.6712,
      "step": 1666
    },
    {
      "epoch": 2.1318505480438437,
      "grad_norm": 0.6873646378517151,
      "learning_rate": 2.7117995330878604e-05,
      "loss": 0.6888,
      "step": 1667
    },
    {
      "epoch": 2.133130650452036,
      "grad_norm": 0.5608901977539062,
      "learning_rate": 2.7103696462869386e-05,
      "loss": 0.6615,
      "step": 1668
    },
    {
      "epoch": 2.1344107528602287,
      "grad_norm": 0.6234862804412842,
      "learning_rate": 2.7089394091797547e-05,
      "loss": 0.7017,
      "step": 1669
    },
    {
      "epoch": 2.1356908552684213,
      "grad_norm": 0.5756520628929138,
      "learning_rate": 2.7075088227479912e-05,
      "loss": 0.6798,
      "step": 1670
    },
    {
      "epoch": 2.1369709576766143,
      "grad_norm": 0.5852974057197571,
      "learning_rate": 2.7060778879735714e-05,
      "loss": 0.6588,
      "step": 1671
    },
    {
      "epoch": 2.138251060084807,
      "grad_norm": 0.5879574418067932,
      "learning_rate": 2.7046466058386567e-05,
      "loss": 0.6645,
      "step": 1672
    },
    {
      "epoch": 2.1395311624929993,
      "grad_norm": 0.6288869976997375,
      "learning_rate": 2.7032149773256454e-05,
      "loss": 0.6732,
      "step": 1673
    },
    {
      "epoch": 2.140811264901192,
      "grad_norm": 0.6149235963821411,
      "learning_rate": 2.7017830034171772e-05,
      "loss": 0.7066,
      "step": 1674
    },
    {
      "epoch": 2.142091367309385,
      "grad_norm": 0.5771135091781616,
      "learning_rate": 2.7003506850961262e-05,
      "loss": 0.6752,
      "step": 1675
    },
    {
      "epoch": 2.1433714697175774,
      "grad_norm": 0.591477632522583,
      "learning_rate": 2.6989180233456034e-05,
      "loss": 0.6893,
      "step": 1676
    },
    {
      "epoch": 2.14465157212577,
      "grad_norm": 0.5606040954589844,
      "learning_rate": 2.697485019148955e-05,
      "loss": 0.6712,
      "step": 1677
    },
    {
      "epoch": 2.145931674533963,
      "grad_norm": 0.6028363704681396,
      "learning_rate": 2.6960516734897638e-05,
      "loss": 0.659,
      "step": 1678
    },
    {
      "epoch": 2.1472117769421555,
      "grad_norm": 0.5561127662658691,
      "learning_rate": 2.6946179873518453e-05,
      "loss": 0.6857,
      "step": 1679
    },
    {
      "epoch": 2.148491879350348,
      "grad_norm": 0.5461053848266602,
      "learning_rate": 2.6931839617192507e-05,
      "loss": 0.6967,
      "step": 1680
    },
    {
      "epoch": 2.148491879350348,
      "eval_loss": 0.01164963562041521,
      "eval_runtime": 10.8268,
      "eval_samples_per_second": 45.812,
      "eval_steps_per_second": 5.727,
      "step": 1680
    },
    {
      "epoch": 2.1497719817585406,
      "grad_norm": 0.5463610291481018,
      "learning_rate": 2.6917495975762623e-05,
      "loss": 0.6513,
      "step": 1681
    },
    {
      "epoch": 2.1510520841667335,
      "grad_norm": 0.6252394914627075,
      "learning_rate": 2.6903148959073947e-05,
      "loss": 0.6859,
      "step": 1682
    },
    {
      "epoch": 2.152332186574926,
      "grad_norm": 0.5732648968696594,
      "learning_rate": 2.6888798576973955e-05,
      "loss": 0.7048,
      "step": 1683
    },
    {
      "epoch": 2.1536122889831186,
      "grad_norm": 0.565711259841919,
      "learning_rate": 2.6874444839312427e-05,
      "loss": 0.6963,
      "step": 1684
    },
    {
      "epoch": 2.154892391391311,
      "grad_norm": 0.539977490901947,
      "learning_rate": 2.6860087755941454e-05,
      "loss": 0.6532,
      "step": 1685
    },
    {
      "epoch": 2.156172493799504,
      "grad_norm": 0.5586331486701965,
      "learning_rate": 2.6845727336715393e-05,
      "loss": 0.7118,
      "step": 1686
    },
    {
      "epoch": 2.1574525962076967,
      "grad_norm": 0.5553241968154907,
      "learning_rate": 2.6831363591490945e-05,
      "loss": 0.6503,
      "step": 1687
    },
    {
      "epoch": 2.1587326986158892,
      "grad_norm": 0.5359736680984497,
      "learning_rate": 2.681699653012704e-05,
      "loss": 0.7141,
      "step": 1688
    },
    {
      "epoch": 2.1600128010240818,
      "grad_norm": 0.5906563997268677,
      "learning_rate": 2.68026261624849e-05,
      "loss": 0.6937,
      "step": 1689
    },
    {
      "epoch": 2.1612929034322748,
      "grad_norm": 0.5959696173667908,
      "learning_rate": 2.6788252498428048e-05,
      "loss": 0.6787,
      "step": 1690
    },
    {
      "epoch": 2.1625730058404673,
      "grad_norm": 0.5390402674674988,
      "learning_rate": 2.677387554782222e-05,
      "loss": 0.6971,
      "step": 1691
    },
    {
      "epoch": 2.16385310824866,
      "grad_norm": 0.5669422149658203,
      "learning_rate": 2.6759495320535448e-05,
      "loss": 0.6937,
      "step": 1692
    },
    {
      "epoch": 2.1651332106568524,
      "grad_norm": 0.5678061842918396,
      "learning_rate": 2.6745111826437987e-05,
      "loss": 0.668,
      "step": 1693
    },
    {
      "epoch": 2.1664133130650454,
      "grad_norm": 0.5993732810020447,
      "learning_rate": 2.6730725075402358e-05,
      "loss": 0.6931,
      "step": 1694
    },
    {
      "epoch": 2.167693415473238,
      "grad_norm": 0.6012265682220459,
      "learning_rate": 2.671633507730329e-05,
      "loss": 0.6892,
      "step": 1695
    },
    {
      "epoch": 2.1689735178814304,
      "grad_norm": 0.6299840211868286,
      "learning_rate": 2.6701941842017755e-05,
      "loss": 0.6923,
      "step": 1696
    },
    {
      "epoch": 2.170253620289623,
      "grad_norm": 0.5930740833282471,
      "learning_rate": 2.6687545379424954e-05,
      "loss": 0.6837,
      "step": 1697
    },
    {
      "epoch": 2.171533722697816,
      "grad_norm": 0.6240190863609314,
      "learning_rate": 2.667314569940629e-05,
      "loss": 0.715,
      "step": 1698
    },
    {
      "epoch": 2.1728138251060085,
      "grad_norm": 0.595062792301178,
      "learning_rate": 2.6658742811845377e-05,
      "loss": 0.65,
      "step": 1699
    },
    {
      "epoch": 2.174093927514201,
      "grad_norm": 0.587990939617157,
      "learning_rate": 2.664433672662804e-05,
      "loss": 0.6612,
      "step": 1700
    },
    {
      "epoch": 2.1753740299223936,
      "grad_norm": 0.6491554379463196,
      "learning_rate": 2.6629927453642293e-05,
      "loss": 0.6947,
      "step": 1701
    },
    {
      "epoch": 2.1766541323305866,
      "grad_norm": 0.5642626881599426,
      "learning_rate": 2.661551500277833e-05,
      "loss": 0.6916,
      "step": 1702
    },
    {
      "epoch": 2.177934234738779,
      "grad_norm": 0.6086171269416809,
      "learning_rate": 2.6601099383928545e-05,
      "loss": 0.6939,
      "step": 1703
    },
    {
      "epoch": 2.1792143371469717,
      "grad_norm": 0.5727648735046387,
      "learning_rate": 2.6586680606987485e-05,
      "loss": 0.682,
      "step": 1704
    },
    {
      "epoch": 2.180494439555164,
      "grad_norm": 0.5541107654571533,
      "learning_rate": 2.6572258681851878e-05,
      "loss": 0.666,
      "step": 1705
    },
    {
      "epoch": 2.181774541963357,
      "grad_norm": 0.6434026956558228,
      "learning_rate": 2.6557833618420614e-05,
      "loss": 0.6979,
      "step": 1706
    },
    {
      "epoch": 2.1830546443715497,
      "grad_norm": 0.5662131309509277,
      "learning_rate": 2.6543405426594725e-05,
      "loss": 0.6467,
      "step": 1707
    },
    {
      "epoch": 2.1843347467797423,
      "grad_norm": 0.5646231770515442,
      "learning_rate": 2.65289741162774e-05,
      "loss": 0.6804,
      "step": 1708
    },
    {
      "epoch": 2.1856148491879352,
      "grad_norm": 0.5460836887359619,
      "learning_rate": 2.6514539697373964e-05,
      "loss": 0.658,
      "step": 1709
    },
    {
      "epoch": 2.186894951596128,
      "grad_norm": 0.687023937702179,
      "learning_rate": 2.6500102179791887e-05,
      "loss": 0.6993,
      "step": 1710
    },
    {
      "epoch": 2.186894951596128,
      "eval_loss": 0.011640787124633789,
      "eval_runtime": 10.8231,
      "eval_samples_per_second": 45.828,
      "eval_steps_per_second": 5.729,
      "step": 1710
    },
    {
      "epoch": 2.1881750540043203,
      "grad_norm": 0.5741842985153198,
      "learning_rate": 2.6485661573440746e-05,
      "loss": 0.6554,
      "step": 1711
    },
    {
      "epoch": 2.189455156412513,
      "grad_norm": 0.5959102511405945,
      "learning_rate": 2.6471217888232255e-05,
      "loss": 0.6763,
      "step": 1712
    },
    {
      "epoch": 2.190735258820706,
      "grad_norm": 0.6111892461776733,
      "learning_rate": 2.645677113408023e-05,
      "loss": 0.7032,
      "step": 1713
    },
    {
      "epoch": 2.1920153612288984,
      "grad_norm": 0.621751070022583,
      "learning_rate": 2.644232132090061e-05,
      "loss": 0.6911,
      "step": 1714
    },
    {
      "epoch": 2.193295463637091,
      "grad_norm": 0.5443041324615479,
      "learning_rate": 2.6427868458611404e-05,
      "loss": 0.6594,
      "step": 1715
    },
    {
      "epoch": 2.1945755660452835,
      "grad_norm": 0.6254043579101562,
      "learning_rate": 2.6413412557132753e-05,
      "loss": 0.6904,
      "step": 1716
    },
    {
      "epoch": 2.1958556684534765,
      "grad_norm": 0.5650925636291504,
      "learning_rate": 2.6398953626386852e-05,
      "loss": 0.6876,
      "step": 1717
    },
    {
      "epoch": 2.197135770861669,
      "grad_norm": 0.5712993741035461,
      "learning_rate": 2.638449167629798e-05,
      "loss": 0.6473,
      "step": 1718
    },
    {
      "epoch": 2.1984158732698615,
      "grad_norm": 0.6259288787841797,
      "learning_rate": 2.6370026716792505e-05,
      "loss": 0.6534,
      "step": 1719
    },
    {
      "epoch": 2.199695975678054,
      "grad_norm": 0.5659099817276001,
      "learning_rate": 2.6355558757798843e-05,
      "loss": 0.6968,
      "step": 1720
    },
    {
      "epoch": 2.200976078086247,
      "grad_norm": 0.6160751581192017,
      "learning_rate": 2.634108780924749e-05,
      "loss": 0.6777,
      "step": 1721
    },
    {
      "epoch": 2.2022561804944396,
      "grad_norm": 0.5935443639755249,
      "learning_rate": 2.6326613881070958e-05,
      "loss": 0.6792,
      "step": 1722
    },
    {
      "epoch": 2.203536282902632,
      "grad_norm": 0.6939257383346558,
      "learning_rate": 2.6312136983203848e-05,
      "loss": 0.6756,
      "step": 1723
    },
    {
      "epoch": 2.2048163853108247,
      "grad_norm": 0.6069145202636719,
      "learning_rate": 2.6297657125582765e-05,
      "loss": 0.7108,
      "step": 1724
    },
    {
      "epoch": 2.2060964877190177,
      "grad_norm": 0.581441342830658,
      "learning_rate": 2.6283174318146352e-05,
      "loss": 0.6849,
      "step": 1725
    },
    {
      "epoch": 2.20737659012721,
      "grad_norm": 0.5974293351173401,
      "learning_rate": 2.6268688570835302e-05,
      "loss": 0.6587,
      "step": 1726
    },
    {
      "epoch": 2.2086566925354028,
      "grad_norm": 0.5970581769943237,
      "learning_rate": 2.6254199893592293e-05,
      "loss": 0.6656,
      "step": 1727
    },
    {
      "epoch": 2.2099367949435953,
      "grad_norm": 0.5888835787773132,
      "learning_rate": 2.623970829636203e-05,
      "loss": 0.6743,
      "step": 1728
    },
    {
      "epoch": 2.2112168973517883,
      "grad_norm": 0.5594384670257568,
      "learning_rate": 2.6225213789091218e-05,
      "loss": 0.6939,
      "step": 1729
    },
    {
      "epoch": 2.212496999759981,
      "grad_norm": 0.5629306435585022,
      "learning_rate": 2.621071638172857e-05,
      "loss": 0.6842,
      "step": 1730
    },
    {
      "epoch": 2.2137771021681734,
      "grad_norm": 0.5158959627151489,
      "learning_rate": 2.619621608422477e-05,
      "loss": 0.6321,
      "step": 1731
    },
    {
      "epoch": 2.215057204576366,
      "grad_norm": 0.5734906196594238,
      "learning_rate": 2.6181712906532502e-05,
      "loss": 0.6727,
      "step": 1732
    },
    {
      "epoch": 2.216337306984559,
      "grad_norm": 0.55968177318573,
      "learning_rate": 2.616720685860642e-05,
      "loss": 0.6652,
      "step": 1733
    },
    {
      "epoch": 2.2176174093927514,
      "grad_norm": 0.559408962726593,
      "learning_rate": 2.615269795040315e-05,
      "loss": 0.637,
      "step": 1734
    },
    {
      "epoch": 2.218897511800944,
      "grad_norm": 0.5392422080039978,
      "learning_rate": 2.613818619188129e-05,
      "loss": 0.6718,
      "step": 1735
    },
    {
      "epoch": 2.2201776142091365,
      "grad_norm": 0.5550699234008789,
      "learning_rate": 2.612367159300137e-05,
      "loss": 0.6419,
      "step": 1736
    },
    {
      "epoch": 2.2214577166173295,
      "grad_norm": 0.5608376860618591,
      "learning_rate": 2.610915416372588e-05,
      "loss": 0.7034,
      "step": 1737
    },
    {
      "epoch": 2.222737819025522,
      "grad_norm": 0.5588756203651428,
      "learning_rate": 2.6094633914019277e-05,
      "loss": 0.6844,
      "step": 1738
    },
    {
      "epoch": 2.2240179214337146,
      "grad_norm": 0.5820834040641785,
      "learning_rate": 2.6080110853847922e-05,
      "loss": 0.6203,
      "step": 1739
    },
    {
      "epoch": 2.2252980238419076,
      "grad_norm": 0.5820999145507812,
      "learning_rate": 2.6065584993180123e-05,
      "loss": 0.7025,
      "step": 1740
    },
    {
      "epoch": 2.2252980238419076,
      "eval_loss": 0.01162695325911045,
      "eval_runtime": 10.8218,
      "eval_samples_per_second": 45.833,
      "eval_steps_per_second": 5.729,
      "step": 1740
    },
    {
      "epoch": 2.2265781262501,
      "grad_norm": 0.5654394030570984,
      "learning_rate": 2.60510563419861e-05,
      "loss": 0.6474,
      "step": 1741
    },
    {
      "epoch": 2.2278582286582926,
      "grad_norm": 0.5664428472518921,
      "learning_rate": 2.6036524910237993e-05,
      "loss": 0.6808,
      "step": 1742
    },
    {
      "epoch": 2.229138331066485,
      "grad_norm": 0.580121636390686,
      "learning_rate": 2.6021990707909858e-05,
      "loss": 0.663,
      "step": 1743
    },
    {
      "epoch": 2.230418433474678,
      "grad_norm": 0.5712918043136597,
      "learning_rate": 2.600745374497764e-05,
      "loss": 0.6333,
      "step": 1744
    },
    {
      "epoch": 2.2316985358828707,
      "grad_norm": 0.556452214717865,
      "learning_rate": 2.5992914031419174e-05,
      "loss": 0.6769,
      "step": 1745
    },
    {
      "epoch": 2.2329786382910632,
      "grad_norm": 0.577888548374176,
      "learning_rate": 2.5978371577214213e-05,
      "loss": 0.6828,
      "step": 1746
    },
    {
      "epoch": 2.234258740699256,
      "grad_norm": 0.5410756468772888,
      "learning_rate": 2.596382639234435e-05,
      "loss": 0.6648,
      "step": 1747
    },
    {
      "epoch": 2.2355388431074488,
      "grad_norm": 0.5517366528511047,
      "learning_rate": 2.5949278486793086e-05,
      "loss": 0.6921,
      "step": 1748
    },
    {
      "epoch": 2.2368189455156413,
      "grad_norm": 0.6139863729476929,
      "learning_rate": 2.593472787054578e-05,
      "loss": 0.643,
      "step": 1749
    },
    {
      "epoch": 2.238099047923834,
      "grad_norm": 0.6104927659034729,
      "learning_rate": 2.5920174553589633e-05,
      "loss": 0.7687,
      "step": 1750
    },
    {
      "epoch": 2.2393791503320264,
      "grad_norm": 0.5959731936454773,
      "learning_rate": 2.5905618545913728e-05,
      "loss": 0.6851,
      "step": 1751
    },
    {
      "epoch": 2.2406592527402194,
      "grad_norm": 0.5839214324951172,
      "learning_rate": 2.5891059857508974e-05,
      "loss": 0.6752,
      "step": 1752
    },
    {
      "epoch": 2.241939355148412,
      "grad_norm": 0.5770272016525269,
      "learning_rate": 2.5876498498368135e-05,
      "loss": 0.657,
      "step": 1753
    },
    {
      "epoch": 2.2432194575566045,
      "grad_norm": 0.575888454914093,
      "learning_rate": 2.586193447848579e-05,
      "loss": 0.6853,
      "step": 1754
    },
    {
      "epoch": 2.244499559964797,
      "grad_norm": 0.5530557632446289,
      "learning_rate": 2.584736780785837e-05,
      "loss": 0.6586,
      "step": 1755
    },
    {
      "epoch": 2.24577966237299,
      "grad_norm": 0.6000199317932129,
      "learning_rate": 2.5832798496484102e-05,
      "loss": 0.7374,
      "step": 1756
    },
    {
      "epoch": 2.2470597647811825,
      "grad_norm": 0.5491329431533813,
      "learning_rate": 2.5818226554363027e-05,
      "loss": 0.6658,
      "step": 1757
    },
    {
      "epoch": 2.248339867189375,
      "grad_norm": 0.5558183789253235,
      "learning_rate": 2.580365199149701e-05,
      "loss": 0.6819,
      "step": 1758
    },
    {
      "epoch": 2.2496199695975676,
      "grad_norm": 0.630332887172699,
      "learning_rate": 2.5789074817889704e-05,
      "loss": 0.6603,
      "step": 1759
    },
    {
      "epoch": 2.2509000720057606,
      "grad_norm": 0.6096137166023254,
      "learning_rate": 2.5774495043546548e-05,
      "loss": 0.7123,
      "step": 1760
    },
    {
      "epoch": 2.252180174413953,
      "grad_norm": 0.6396390199661255,
      "learning_rate": 2.575991267847477e-05,
      "loss": 0.6985,
      "step": 1761
    },
    {
      "epoch": 2.2534602768221457,
      "grad_norm": 0.5843227505683899,
      "learning_rate": 2.5745327732683384e-05,
      "loss": 0.6679,
      "step": 1762
    },
    {
      "epoch": 2.2547403792303387,
      "grad_norm": 0.6904280781745911,
      "learning_rate": 2.5730740216183168e-05,
      "loss": 0.6861,
      "step": 1763
    },
    {
      "epoch": 2.256020481638531,
      "grad_norm": 0.6529262065887451,
      "learning_rate": 2.5716150138986654e-05,
      "loss": 0.673,
      "step": 1764
    },
    {
      "epoch": 2.2573005840467237,
      "grad_norm": 0.5620155930519104,
      "learning_rate": 2.570155751110816e-05,
      "loss": 0.6612,
      "step": 1765
    },
    {
      "epoch": 2.2585806864549163,
      "grad_norm": 0.655309796333313,
      "learning_rate": 2.568696234256373e-05,
      "loss": 0.668,
      "step": 1766
    },
    {
      "epoch": 2.259860788863109,
      "grad_norm": 0.5984154939651489,
      "learning_rate": 2.567236464337117e-05,
      "loss": 0.7186,
      "step": 1767
    },
    {
      "epoch": 2.261140891271302,
      "grad_norm": 0.5751389861106873,
      "learning_rate": 2.5657764423549994e-05,
      "loss": 0.6656,
      "step": 1768
    },
    {
      "epoch": 2.2624209936794943,
      "grad_norm": 0.5889517664909363,
      "learning_rate": 2.5643161693121476e-05,
      "loss": 0.6598,
      "step": 1769
    },
    {
      "epoch": 2.263701096087687,
      "grad_norm": 0.5623475909233093,
      "learning_rate": 2.56285564621086e-05,
      "loss": 0.6939,
      "step": 1770
    },
    {
      "epoch": 2.263701096087687,
      "eval_loss": 0.011622007936239243,
      "eval_runtime": 10.8354,
      "eval_samples_per_second": 45.776,
      "eval_steps_per_second": 5.722,
      "step": 1770
    },
    {
      "epoch": 2.26498119849588,
      "grad_norm": 0.5731052160263062,
      "learning_rate": 2.561394874053607e-05,
      "loss": 0.66,
      "step": 1771
    },
    {
      "epoch": 2.2662613009040724,
      "grad_norm": 0.5799999237060547,
      "learning_rate": 2.5599338538430304e-05,
      "loss": 0.698,
      "step": 1772
    },
    {
      "epoch": 2.267541403312265,
      "grad_norm": 0.6327428221702576,
      "learning_rate": 2.55847258658194e-05,
      "loss": 0.6876,
      "step": 1773
    },
    {
      "epoch": 2.2688215057204575,
      "grad_norm": 0.54714035987854,
      "learning_rate": 2.5570110732733185e-05,
      "loss": 0.6542,
      "step": 1774
    },
    {
      "epoch": 2.2701016081286505,
      "grad_norm": 0.6353325843811035,
      "learning_rate": 2.555549314920315e-05,
      "loss": 0.6615,
      "step": 1775
    },
    {
      "epoch": 2.271381710536843,
      "grad_norm": 0.5877797603607178,
      "learning_rate": 2.554087312526248e-05,
      "loss": 0.6869,
      "step": 1776
    },
    {
      "epoch": 2.2726618129450356,
      "grad_norm": 0.5969734787940979,
      "learning_rate": 2.5526250670946026e-05,
      "loss": 0.6694,
      "step": 1777
    },
    {
      "epoch": 2.273941915353228,
      "grad_norm": 0.5800706744194031,
      "learning_rate": 2.5511625796290314e-05,
      "loss": 0.653,
      "step": 1778
    },
    {
      "epoch": 2.275222017761421,
      "grad_norm": 0.6266164183616638,
      "learning_rate": 2.5496998511333527e-05,
      "loss": 0.7097,
      "step": 1779
    },
    {
      "epoch": 2.2765021201696136,
      "grad_norm": 0.5548282861709595,
      "learning_rate": 2.5482368826115514e-05,
      "loss": 0.6457,
      "step": 1780
    },
    {
      "epoch": 2.277782222577806,
      "grad_norm": 0.6023533940315247,
      "learning_rate": 2.546773675067775e-05,
      "loss": 0.6873,
      "step": 1781
    },
    {
      "epoch": 2.2790623249859987,
      "grad_norm": 0.5827093720436096,
      "learning_rate": 2.545310229506338e-05,
      "loss": 0.6837,
      "step": 1782
    },
    {
      "epoch": 2.2803424273941917,
      "grad_norm": 0.6479482650756836,
      "learning_rate": 2.5438465469317142e-05,
      "loss": 0.6992,
      "step": 1783
    },
    {
      "epoch": 2.2816225298023842,
      "grad_norm": 0.6267648339271545,
      "learning_rate": 2.542382628348544e-05,
      "loss": 0.7157,
      "step": 1784
    },
    {
      "epoch": 2.2829026322105768,
      "grad_norm": 0.6263638734817505,
      "learning_rate": 2.5409184747616284e-05,
      "loss": 0.6808,
      "step": 1785
    },
    {
      "epoch": 2.2841827346187698,
      "grad_norm": 0.6246517300605774,
      "learning_rate": 2.5394540871759294e-05,
      "loss": 0.6782,
      "step": 1786
    },
    {
      "epoch": 2.2854628370269623,
      "grad_norm": 0.5735856294631958,
      "learning_rate": 2.5379894665965686e-05,
      "loss": 0.6775,
      "step": 1787
    },
    {
      "epoch": 2.286742939435155,
      "grad_norm": 0.6397892236709595,
      "learning_rate": 2.5365246140288302e-05,
      "loss": 0.656,
      "step": 1788
    },
    {
      "epoch": 2.2880230418433474,
      "grad_norm": 0.6607687473297119,
      "learning_rate": 2.5350595304781557e-05,
      "loss": 0.6905,
      "step": 1789
    },
    {
      "epoch": 2.28930314425154,
      "grad_norm": 0.6366332173347473,
      "learning_rate": 2.533594216950144e-05,
      "loss": 0.7036,
      "step": 1790
    },
    {
      "epoch": 2.290583246659733,
      "grad_norm": 0.591999888420105,
      "learning_rate": 2.5321286744505558e-05,
      "loss": 0.6698,
      "step": 1791
    },
    {
      "epoch": 2.2918633490679254,
      "grad_norm": 0.6242043375968933,
      "learning_rate": 2.530662903985305e-05,
      "loss": 0.6874,
      "step": 1792
    },
    {
      "epoch": 2.293143451476118,
      "grad_norm": 0.6108874082565308,
      "learning_rate": 2.5291969065604636e-05,
      "loss": 0.6775,
      "step": 1793
    },
    {
      "epoch": 2.294423553884311,
      "grad_norm": 0.5468871593475342,
      "learning_rate": 2.52773068318226e-05,
      "loss": 0.6388,
      "step": 1794
    },
    {
      "epoch": 2.2957036562925035,
      "grad_norm": 0.5810433030128479,
      "learning_rate": 2.526264234857077e-05,
      "loss": 0.6874,
      "step": 1795
    },
    {
      "epoch": 2.296983758700696,
      "grad_norm": 0.6179283857345581,
      "learning_rate": 2.524797562591451e-05,
      "loss": 0.6789,
      "step": 1796
    },
    {
      "epoch": 2.2982638611088886,
      "grad_norm": 0.5512781739234924,
      "learning_rate": 2.5233306673920735e-05,
      "loss": 0.636,
      "step": 1797
    },
    {
      "epoch": 2.299543963517081,
      "grad_norm": 0.5917335152626038,
      "learning_rate": 2.521863550265788e-05,
      "loss": 0.677,
      "step": 1798
    },
    {
      "epoch": 2.300824065925274,
      "grad_norm": 0.5568514466285706,
      "learning_rate": 2.5203962122195915e-05,
      "loss": 0.6804,
      "step": 1799
    },
    {
      "epoch": 2.3021041683334666,
      "grad_norm": 0.5666114687919617,
      "learning_rate": 2.5189286542606302e-05,
      "loss": 0.6994,
      "step": 1800
    },
    {
      "epoch": 2.3021041683334666,
      "eval_loss": 0.011595751158893108,
      "eval_runtime": 10.8252,
      "eval_samples_per_second": 45.819,
      "eval_steps_per_second": 5.727,
      "step": 1800
    },
    {
      "epoch": 2.303384270741659,
      "grad_norm": 0.5425809025764465,
      "learning_rate": 2.517460877396205e-05,
      "loss": 0.6891,
      "step": 1801
    },
    {
      "epoch": 2.304664373149852,
      "grad_norm": 0.5229185819625854,
      "learning_rate": 2.5159928826337638e-05,
      "loss": 0.657,
      "step": 1802
    },
    {
      "epoch": 2.3059444755580447,
      "grad_norm": 0.6061989068984985,
      "learning_rate": 2.514524670980905e-05,
      "loss": 0.6868,
      "step": 1803
    },
    {
      "epoch": 2.3072245779662373,
      "grad_norm": 0.5069493651390076,
      "learning_rate": 2.5130562434453775e-05,
      "loss": 0.648,
      "step": 1804
    },
    {
      "epoch": 2.30850468037443,
      "grad_norm": 0.6272199153900146,
      "learning_rate": 2.5115876010350762e-05,
      "loss": 0.6879,
      "step": 1805
    },
    {
      "epoch": 2.309784782782623,
      "grad_norm": 0.5859187245368958,
      "learning_rate": 2.5101187447580433e-05,
      "loss": 0.6661,
      "step": 1806
    },
    {
      "epoch": 2.3110648851908153,
      "grad_norm": 0.5726884007453918,
      "learning_rate": 2.50864967562247e-05,
      "loss": 0.6557,
      "step": 1807
    },
    {
      "epoch": 2.312344987599008,
      "grad_norm": 0.557464599609375,
      "learning_rate": 2.507180394636692e-05,
      "loss": 0.7158,
      "step": 1808
    },
    {
      "epoch": 2.3136250900072004,
      "grad_norm": 0.6684802770614624,
      "learning_rate": 2.50571090280919e-05,
      "loss": 0.6756,
      "step": 1809
    },
    {
      "epoch": 2.3149051924153934,
      "grad_norm": 0.635189950466156,
      "learning_rate": 2.5042412011485918e-05,
      "loss": 0.7325,
      "step": 1810
    },
    {
      "epoch": 2.316185294823586,
      "grad_norm": 0.6043376326560974,
      "learning_rate": 2.5027712906636664e-05,
      "loss": 0.696,
      "step": 1811
    },
    {
      "epoch": 2.3174653972317785,
      "grad_norm": 0.6636263728141785,
      "learning_rate": 2.501301172363327e-05,
      "loss": 0.7139,
      "step": 1812
    },
    {
      "epoch": 2.318745499639971,
      "grad_norm": 0.631271481513977,
      "learning_rate": 2.4998308472566315e-05,
      "loss": 0.7266,
      "step": 1813
    },
    {
      "epoch": 2.320025602048164,
      "grad_norm": 0.5650892853736877,
      "learning_rate": 2.4983603163527765e-05,
      "loss": 0.6902,
      "step": 1814
    },
    {
      "epoch": 2.3213057044563565,
      "grad_norm": 0.5594301223754883,
      "learning_rate": 2.4968895806611014e-05,
      "loss": 0.6911,
      "step": 1815
    },
    {
      "epoch": 2.322585806864549,
      "grad_norm": 0.5877586007118225,
      "learning_rate": 2.4954186411910876e-05,
      "loss": 0.6776,
      "step": 1816
    },
    {
      "epoch": 2.323865909272742,
      "grad_norm": 0.5810129642486572,
      "learning_rate": 2.493947498952353e-05,
      "loss": 0.6915,
      "step": 1817
    },
    {
      "epoch": 2.3251460116809346,
      "grad_norm": 0.5952797532081604,
      "learning_rate": 2.4924761549546576e-05,
      "loss": 0.6715,
      "step": 1818
    },
    {
      "epoch": 2.326426114089127,
      "grad_norm": 0.5925062894821167,
      "learning_rate": 2.4910046102078993e-05,
      "loss": 0.6518,
      "step": 1819
    },
    {
      "epoch": 2.3277062164973197,
      "grad_norm": 0.5474934577941895,
      "learning_rate": 2.4895328657221116e-05,
      "loss": 0.666,
      "step": 1820
    },
    {
      "epoch": 2.328986318905512,
      "grad_norm": 0.6176634430885315,
      "learning_rate": 2.488060922507469e-05,
      "loss": 0.7111,
      "step": 1821
    },
    {
      "epoch": 2.330266421313705,
      "grad_norm": 0.6067846417427063,
      "learning_rate": 2.4865887815742794e-05,
      "loss": 0.6588,
      "step": 1822
    },
    {
      "epoch": 2.3315465237218977,
      "grad_norm": 0.5685158371925354,
      "learning_rate": 2.4851164439329872e-05,
      "loss": 0.6391,
      "step": 1823
    },
    {
      "epoch": 2.3328266261300903,
      "grad_norm": 0.5903010368347168,
      "learning_rate": 2.4836439105941716e-05,
      "loss": 0.6659,
      "step": 1824
    },
    {
      "epoch": 2.3341067285382833,
      "grad_norm": 0.6228030920028687,
      "learning_rate": 2.4821711825685467e-05,
      "loss": 0.6544,
      "step": 1825
    },
    {
      "epoch": 2.335386830946476,
      "grad_norm": 0.5518401265144348,
      "learning_rate": 2.48069826086696e-05,
      "loss": 0.6822,
      "step": 1826
    },
    {
      "epoch": 2.3366669333546684,
      "grad_norm": 0.6300826668739319,
      "learning_rate": 2.479225146500392e-05,
      "loss": 0.7065,
      "step": 1827
    },
    {
      "epoch": 2.337947035762861,
      "grad_norm": 0.6295920610427856,
      "learning_rate": 2.4777518404799553e-05,
      "loss": 0.6738,
      "step": 1828
    },
    {
      "epoch": 2.339227138171054,
      "grad_norm": 0.6495180130004883,
      "learning_rate": 2.476278343816893e-05,
      "loss": 0.701,
      "step": 1829
    },
    {
      "epoch": 2.3405072405792464,
      "grad_norm": 0.5673889517784119,
      "learning_rate": 2.4748046575225817e-05,
      "loss": 0.6687,
      "step": 1830
    },
    {
      "epoch": 2.3405072405792464,
      "eval_loss": 0.011586235836148262,
      "eval_runtime": 10.8232,
      "eval_samples_per_second": 45.827,
      "eval_steps_per_second": 5.728,
      "step": 1830
    },
    {
      "epoch": 2.341787342987439,
      "grad_norm": 0.5827108025550842,
      "learning_rate": 2.4733307826085265e-05,
      "loss": 0.6957,
      "step": 1831
    },
    {
      "epoch": 2.3430674453956315,
      "grad_norm": 0.5745197534561157,
      "learning_rate": 2.4718567200863605e-05,
      "loss": 0.727,
      "step": 1832
    },
    {
      "epoch": 2.3443475478038245,
      "grad_norm": 0.5660216212272644,
      "learning_rate": 2.470382470967848e-05,
      "loss": 0.7051,
      "step": 1833
    },
    {
      "epoch": 2.345627650212017,
      "grad_norm": 0.554172694683075,
      "learning_rate": 2.468908036264881e-05,
      "loss": 0.6727,
      "step": 1834
    },
    {
      "epoch": 2.3469077526202096,
      "grad_norm": 0.5728502869606018,
      "learning_rate": 2.467433416989477e-05,
      "loss": 0.6635,
      "step": 1835
    },
    {
      "epoch": 2.348187855028402,
      "grad_norm": 0.5659500956535339,
      "learning_rate": 2.465958614153783e-05,
      "loss": 0.6571,
      "step": 1836
    },
    {
      "epoch": 2.349467957436595,
      "grad_norm": 0.6034412384033203,
      "learning_rate": 2.4644836287700704e-05,
      "loss": 0.7051,
      "step": 1837
    },
    {
      "epoch": 2.3507480598447876,
      "grad_norm": 0.5409714579582214,
      "learning_rate": 2.463008461850735e-05,
      "loss": 0.6398,
      "step": 1838
    },
    {
      "epoch": 2.35202816225298,
      "grad_norm": 0.6442564725875854,
      "learning_rate": 2.4615331144082984e-05,
      "loss": 0.6884,
      "step": 1839
    },
    {
      "epoch": 2.3533082646611727,
      "grad_norm": 0.593873918056488,
      "learning_rate": 2.4600575874554072e-05,
      "loss": 0.7073,
      "step": 1840
    },
    {
      "epoch": 2.3545883670693657,
      "grad_norm": 0.5618162751197815,
      "learning_rate": 2.458581882004829e-05,
      "loss": 0.6912,
      "step": 1841
    },
    {
      "epoch": 2.3558684694775582,
      "grad_norm": 0.6092703342437744,
      "learning_rate": 2.4571059990694547e-05,
      "loss": 0.6489,
      "step": 1842
    },
    {
      "epoch": 2.3571485718857508,
      "grad_norm": 0.5499894022941589,
      "learning_rate": 2.4556299396622976e-05,
      "loss": 0.6824,
      "step": 1843
    },
    {
      "epoch": 2.3584286742939433,
      "grad_norm": 0.5832574963569641,
      "learning_rate": 2.4541537047964918e-05,
      "loss": 0.6906,
      "step": 1844
    },
    {
      "epoch": 2.3597087767021363,
      "grad_norm": 0.5843065977096558,
      "learning_rate": 2.452677295485291e-05,
      "loss": 0.6877,
      "step": 1845
    },
    {
      "epoch": 2.360988879110329,
      "grad_norm": 0.5764889717102051,
      "learning_rate": 2.4512007127420705e-05,
      "loss": 0.6987,
      "step": 1846
    },
    {
      "epoch": 2.3622689815185214,
      "grad_norm": 0.5799248814582825,
      "learning_rate": 2.4497239575803233e-05,
      "loss": 0.6681,
      "step": 1847
    },
    {
      "epoch": 2.3635490839267144,
      "grad_norm": 0.6558672189712524,
      "learning_rate": 2.4482470310136597e-05,
      "loss": 0.6965,
      "step": 1848
    },
    {
      "epoch": 2.364829186334907,
      "grad_norm": 0.5738571286201477,
      "learning_rate": 2.4467699340558108e-05,
      "loss": 0.6943,
      "step": 1849
    },
    {
      "epoch": 2.3661092887430994,
      "grad_norm": 0.5818512439727783,
      "learning_rate": 2.4452926677206215e-05,
      "loss": 0.6429,
      "step": 1850
    },
    {
      "epoch": 2.367389391151292,
      "grad_norm": 0.594440758228302,
      "learning_rate": 2.4438152330220546e-05,
      "loss": 0.6617,
      "step": 1851
    },
    {
      "epoch": 2.3686694935594845,
      "grad_norm": 0.6094481945037842,
      "learning_rate": 2.4423376309741878e-05,
      "loss": 0.6616,
      "step": 1852
    },
    {
      "epoch": 2.3699495959676775,
      "grad_norm": 0.59006667137146,
      "learning_rate": 2.4408598625912154e-05,
      "loss": 0.6822,
      "step": 1853
    },
    {
      "epoch": 2.37122969837587,
      "grad_norm": 0.557752251625061,
      "learning_rate": 2.4393819288874422e-05,
      "loss": 0.685,
      "step": 1854
    },
    {
      "epoch": 2.3725098007840626,
      "grad_norm": 0.5917968153953552,
      "learning_rate": 2.4379038308772897e-05,
      "loss": 0.7104,
      "step": 1855
    },
    {
      "epoch": 2.3737899031922556,
      "grad_norm": 0.6051133275032043,
      "learning_rate": 2.4364255695752917e-05,
      "loss": 0.6762,
      "step": 1856
    },
    {
      "epoch": 2.375070005600448,
      "grad_norm": 0.49217554926872253,
      "learning_rate": 2.4349471459960935e-05,
      "loss": 0.6687,
      "step": 1857
    },
    {
      "epoch": 2.3763501080086407,
      "grad_norm": 0.582585334777832,
      "learning_rate": 2.4334685611544505e-05,
      "loss": 0.6544,
      "step": 1858
    },
    {
      "epoch": 2.377630210416833,
      "grad_norm": 0.5836172103881836,
      "learning_rate": 2.4319898160652313e-05,
      "loss": 0.6558,
      "step": 1859
    },
    {
      "epoch": 2.378910312825026,
      "grad_norm": 0.5319933295249939,
      "learning_rate": 2.430510911743414e-05,
      "loss": 0.7057,
      "step": 1860
    },
    {
      "epoch": 2.378910312825026,
      "eval_loss": 0.011584697291254997,
      "eval_runtime": 10.8209,
      "eval_samples_per_second": 45.837,
      "eval_steps_per_second": 5.73,
      "step": 1860
    },
    {
      "epoch": 2.3801904152332187,
      "grad_norm": 0.6212523579597473,
      "learning_rate": 2.4290318492040835e-05,
      "loss": 0.6861,
      "step": 1861
    },
    {
      "epoch": 2.3814705176414113,
      "grad_norm": 0.5772797465324402,
      "learning_rate": 2.427552629462436e-05,
      "loss": 0.6624,
      "step": 1862
    },
    {
      "epoch": 2.382750620049604,
      "grad_norm": 0.6138158440589905,
      "learning_rate": 2.426073253533775e-05,
      "loss": 0.6935,
      "step": 1863
    },
    {
      "epoch": 2.384030722457797,
      "grad_norm": 0.5808906555175781,
      "learning_rate": 2.4245937224335103e-05,
      "loss": 0.6628,
      "step": 1864
    },
    {
      "epoch": 2.3853108248659893,
      "grad_norm": 0.578218400478363,
      "learning_rate": 2.4231140371771592e-05,
      "loss": 0.6716,
      "step": 1865
    },
    {
      "epoch": 2.386590927274182,
      "grad_norm": 0.577086329460144,
      "learning_rate": 2.421634198780345e-05,
      "loss": 0.6768,
      "step": 1866
    },
    {
      "epoch": 2.3878710296823744,
      "grad_norm": 0.5849325656890869,
      "learning_rate": 2.4201542082587942e-05,
      "loss": 0.6599,
      "step": 1867
    },
    {
      "epoch": 2.3891511320905674,
      "grad_norm": 0.5561035871505737,
      "learning_rate": 2.4186740666283397e-05,
      "loss": 0.6413,
      "step": 1868
    },
    {
      "epoch": 2.39043123449876,
      "grad_norm": 0.5579114556312561,
      "learning_rate": 2.4171937749049175e-05,
      "loss": 0.6621,
      "step": 1869
    },
    {
      "epoch": 2.3917113369069525,
      "grad_norm": 0.5725136399269104,
      "learning_rate": 2.415713334104567e-05,
      "loss": 0.6854,
      "step": 1870
    },
    {
      "epoch": 2.392991439315145,
      "grad_norm": 0.6551234722137451,
      "learning_rate": 2.4142327452434292e-05,
      "loss": 0.7158,
      "step": 1871
    },
    {
      "epoch": 2.394271541723338,
      "grad_norm": 0.5876832604408264,
      "learning_rate": 2.4127520093377465e-05,
      "loss": 0.6816,
      "step": 1872
    },
    {
      "epoch": 2.3955516441315305,
      "grad_norm": 0.5829140543937683,
      "learning_rate": 2.411271127403864e-05,
      "loss": 0.6899,
      "step": 1873
    },
    {
      "epoch": 2.396831746539723,
      "grad_norm": 0.582199215888977,
      "learning_rate": 2.4097901004582233e-05,
      "loss": 0.6775,
      "step": 1874
    },
    {
      "epoch": 2.3981118489479156,
      "grad_norm": 0.6164038181304932,
      "learning_rate": 2.4083089295173716e-05,
      "loss": 0.6959,
      "step": 1875
    },
    {
      "epoch": 2.3993919513561086,
      "grad_norm": 0.6018531322479248,
      "learning_rate": 2.4068276155979486e-05,
      "loss": 0.6923,
      "step": 1876
    },
    {
      "epoch": 2.400672053764301,
      "grad_norm": 0.591814398765564,
      "learning_rate": 2.405346159716696e-05,
      "loss": 0.7009,
      "step": 1877
    },
    {
      "epoch": 2.4019521561724937,
      "grad_norm": 0.6468870043754578,
      "learning_rate": 2.4038645628904515e-05,
      "loss": 0.6948,
      "step": 1878
    },
    {
      "epoch": 2.4032322585806867,
      "grad_norm": 0.5983352065086365,
      "learning_rate": 2.40238282613615e-05,
      "loss": 0.6815,
      "step": 1879
    },
    {
      "epoch": 2.404512360988879,
      "grad_norm": 0.5535333752632141,
      "learning_rate": 2.4009009504708232e-05,
      "loss": 0.6901,
      "step": 1880
    },
    {
      "epoch": 2.4057924633970718,
      "grad_norm": 0.6194718480110168,
      "learning_rate": 2.3994189369115953e-05,
      "loss": 0.6525,
      "step": 1881
    },
    {
      "epoch": 2.4070725658052643,
      "grad_norm": 0.5655761957168579,
      "learning_rate": 2.3979367864756887e-05,
      "loss": 0.6654,
      "step": 1882
    },
    {
      "epoch": 2.408352668213457,
      "grad_norm": 0.5740327835083008,
      "learning_rate": 2.396454500180418e-05,
      "loss": 0.7022,
      "step": 1883
    },
    {
      "epoch": 2.40963277062165,
      "grad_norm": 0.581233561038971,
      "learning_rate": 2.3949720790431905e-05,
      "loss": 0.6662,
      "step": 1884
    },
    {
      "epoch": 2.4109128730298424,
      "grad_norm": 0.5732260942459106,
      "learning_rate": 2.3934895240815075e-05,
      "loss": 0.6903,
      "step": 1885
    },
    {
      "epoch": 2.412192975438035,
      "grad_norm": 0.5634108781814575,
      "learning_rate": 2.3920068363129612e-05,
      "loss": 0.6402,
      "step": 1886
    },
    {
      "epoch": 2.413473077846228,
      "grad_norm": 0.6228793859481812,
      "learning_rate": 2.390524016755235e-05,
      "loss": 0.7372,
      "step": 1887
    },
    {
      "epoch": 2.4147531802544204,
      "grad_norm": 0.6371725797653198,
      "learning_rate": 2.3890410664261033e-05,
      "loss": 0.6828,
      "step": 1888
    },
    {
      "epoch": 2.416033282662613,
      "grad_norm": 0.5931169986724854,
      "learning_rate": 2.3875579863434295e-05,
      "loss": 0.7019,
      "step": 1889
    },
    {
      "epoch": 2.4173133850708055,
      "grad_norm": 0.5912045836448669,
      "learning_rate": 2.386074777525166e-05,
      "loss": 0.6805,
      "step": 1890
    },
    {
      "epoch": 2.4173133850708055,
      "eval_loss": 0.011568007990717888,
      "eval_runtime": 10.8067,
      "eval_samples_per_second": 45.898,
      "eval_steps_per_second": 5.737,
      "step": 1890
    },
    {
      "epoch": 2.4185934874789985,
      "grad_norm": 0.5691397190093994,
      "learning_rate": 2.3845914409893554e-05,
      "loss": 0.6393,
      "step": 1891
    },
    {
      "epoch": 2.419873589887191,
      "grad_norm": 0.5187730193138123,
      "learning_rate": 2.383107977754126e-05,
      "loss": 0.6547,
      "step": 1892
    },
    {
      "epoch": 2.4211536922953836,
      "grad_norm": 0.5614442825317383,
      "learning_rate": 2.3816243888376925e-05,
      "loss": 0.6563,
      "step": 1893
    },
    {
      "epoch": 2.422433794703576,
      "grad_norm": 0.606225311756134,
      "learning_rate": 2.380140675258359e-05,
      "loss": 0.6619,
      "step": 1894
    },
    {
      "epoch": 2.423713897111769,
      "grad_norm": 0.54556804895401,
      "learning_rate": 2.378656838034511e-05,
      "loss": 0.6669,
      "step": 1895
    },
    {
      "epoch": 2.4249939995199616,
      "grad_norm": 0.6027140617370605,
      "learning_rate": 2.3771728781846228e-05,
      "loss": 0.6791,
      "step": 1896
    },
    {
      "epoch": 2.426274101928154,
      "grad_norm": 0.5797204971313477,
      "learning_rate": 2.3756887967272514e-05,
      "loss": 0.6849,
      "step": 1897
    },
    {
      "epoch": 2.4275542043363467,
      "grad_norm": 0.4976787269115448,
      "learning_rate": 2.3742045946810346e-05,
      "loss": 0.6826,
      "step": 1898
    },
    {
      "epoch": 2.4288343067445397,
      "grad_norm": 0.6234809160232544,
      "learning_rate": 2.3727202730646976e-05,
      "loss": 0.6837,
      "step": 1899
    },
    {
      "epoch": 2.4301144091527322,
      "grad_norm": 0.5645691752433777,
      "learning_rate": 2.3712358328970448e-05,
      "loss": 0.6713,
      "step": 1900
    },
    {
      "epoch": 2.431394511560925,
      "grad_norm": 0.5872960090637207,
      "learning_rate": 2.3697512751969617e-05,
      "loss": 0.6783,
      "step": 1901
    },
    {
      "epoch": 2.4326746139691178,
      "grad_norm": 0.7054588198661804,
      "learning_rate": 2.3682666009834164e-05,
      "loss": 0.7079,
      "step": 1902
    },
    {
      "epoch": 2.4339547163773103,
      "grad_norm": 0.5787381529808044,
      "learning_rate": 2.3667818112754563e-05,
      "loss": 0.7025,
      "step": 1903
    },
    {
      "epoch": 2.435234818785503,
      "grad_norm": 0.6480146646499634,
      "learning_rate": 2.3652969070922062e-05,
      "loss": 0.7056,
      "step": 1904
    },
    {
      "epoch": 2.4365149211936954,
      "grad_norm": 0.6624314785003662,
      "learning_rate": 2.363811889452872e-05,
      "loss": 0.6367,
      "step": 1905
    },
    {
      "epoch": 2.437795023601888,
      "grad_norm": 0.5549313426017761,
      "learning_rate": 2.3623267593767363e-05,
      "loss": 0.6897,
      "step": 1906
    },
    {
      "epoch": 2.439075126010081,
      "grad_norm": 0.6084743142127991,
      "learning_rate": 2.360841517883159e-05,
      "loss": 0.6841,
      "step": 1907
    },
    {
      "epoch": 2.4403552284182735,
      "grad_norm": 0.5819149017333984,
      "learning_rate": 2.3593561659915767e-05,
      "loss": 0.6744,
      "step": 1908
    },
    {
      "epoch": 2.441635330826466,
      "grad_norm": 0.5500880479812622,
      "learning_rate": 2.3578707047215013e-05,
      "loss": 0.687,
      "step": 1909
    },
    {
      "epoch": 2.442915433234659,
      "grad_norm": 0.5665502548217773,
      "learning_rate": 2.35638513509252e-05,
      "loss": 0.6566,
      "step": 1910
    },
    {
      "epoch": 2.4441955356428515,
      "grad_norm": 0.5229345560073853,
      "learning_rate": 2.3548994581242948e-05,
      "loss": 0.6461,
      "step": 1911
    },
    {
      "epoch": 2.445475638051044,
      "grad_norm": 0.6144683957099915,
      "learning_rate": 2.3534136748365604e-05,
      "loss": 0.6902,
      "step": 1912
    },
    {
      "epoch": 2.4467557404592366,
      "grad_norm": 0.5567544102668762,
      "learning_rate": 2.3519277862491255e-05,
      "loss": 0.6814,
      "step": 1913
    },
    {
      "epoch": 2.448035842867429,
      "grad_norm": 0.5664990544319153,
      "learning_rate": 2.3504417933818708e-05,
      "loss": 0.6549,
      "step": 1914
    },
    {
      "epoch": 2.449315945275622,
      "grad_norm": 0.5817349553108215,
      "learning_rate": 2.348955697254748e-05,
      "loss": 0.6857,
      "step": 1915
    },
    {
      "epoch": 2.4505960476838147,
      "grad_norm": 0.5776554346084595,
      "learning_rate": 2.34746949888778e-05,
      "loss": 0.6809,
      "step": 1916
    },
    {
      "epoch": 2.451876150092007,
      "grad_norm": 0.547253429889679,
      "learning_rate": 2.3459831993010605e-05,
      "loss": 0.6898,
      "step": 1917
    },
    {
      "epoch": 2.4531562525002,
      "grad_norm": 0.5491963028907776,
      "learning_rate": 2.3444967995147524e-05,
      "loss": 0.6784,
      "step": 1918
    },
    {
      "epoch": 2.4544363549083927,
      "grad_norm": 0.5832270979881287,
      "learning_rate": 2.3430103005490863e-05,
      "loss": 0.7021,
      "step": 1919
    },
    {
      "epoch": 2.4557164573165853,
      "grad_norm": 0.5577331185340881,
      "learning_rate": 2.3415237034243618e-05,
      "loss": 0.6807,
      "step": 1920
    },
    {
      "epoch": 2.4557164573165853,
      "eval_loss": 0.011559930630028248,
      "eval_runtime": 10.8257,
      "eval_samples_per_second": 45.817,
      "eval_steps_per_second": 5.727,
      "step": 1920
    },
    {
      "epoch": 2.456996559724778,
      "grad_norm": 0.5973002314567566,
      "learning_rate": 2.3400370091609464e-05,
      "loss": 0.6664,
      "step": 1921
    },
    {
      "epoch": 2.458276662132971,
      "grad_norm": 0.5707387924194336,
      "learning_rate": 2.338550218779273e-05,
      "loss": 0.6976,
      "step": 1922
    },
    {
      "epoch": 2.4595567645411633,
      "grad_norm": 0.5324465036392212,
      "learning_rate": 2.3370633332998413e-05,
      "loss": 0.6534,
      "step": 1923
    },
    {
      "epoch": 2.460836866949356,
      "grad_norm": 0.5851748585700989,
      "learning_rate": 2.3355763537432173e-05,
      "loss": 0.6646,
      "step": 1924
    },
    {
      "epoch": 2.4621169693575484,
      "grad_norm": 0.562751293182373,
      "learning_rate": 2.3340892811300283e-05,
      "loss": 0.6806,
      "step": 1925
    },
    {
      "epoch": 2.4633970717657414,
      "grad_norm": 0.5656257271766663,
      "learning_rate": 2.3326021164809688e-05,
      "loss": 0.655,
      "step": 1926
    },
    {
      "epoch": 2.464677174173934,
      "grad_norm": 0.5876888036727905,
      "learning_rate": 2.3311148608167944e-05,
      "loss": 0.7203,
      "step": 1927
    },
    {
      "epoch": 2.4659572765821265,
      "grad_norm": 0.5621166229248047,
      "learning_rate": 2.329627515158325e-05,
      "loss": 0.6935,
      "step": 1928
    },
    {
      "epoch": 2.467237378990319,
      "grad_norm": 0.5966109037399292,
      "learning_rate": 2.3281400805264408e-05,
      "loss": 0.6902,
      "step": 1929
    },
    {
      "epoch": 2.468517481398512,
      "grad_norm": 0.5752228498458862,
      "learning_rate": 2.3266525579420833e-05,
      "loss": 0.6805,
      "step": 1930
    },
    {
      "epoch": 2.4697975838067046,
      "grad_norm": 0.5376203656196594,
      "learning_rate": 2.3251649484262546e-05,
      "loss": 0.6711,
      "step": 1931
    },
    {
      "epoch": 2.471077686214897,
      "grad_norm": 0.5711091160774231,
      "learning_rate": 2.323677253000017e-05,
      "loss": 0.7196,
      "step": 1932
    },
    {
      "epoch": 2.47235778862309,
      "grad_norm": 0.6120408177375793,
      "learning_rate": 2.322189472684491e-05,
      "loss": 0.7178,
      "step": 1933
    },
    {
      "epoch": 2.4736378910312826,
      "grad_norm": 0.544459342956543,
      "learning_rate": 2.320701608500855e-05,
      "loss": 0.6542,
      "step": 1934
    },
    {
      "epoch": 2.474917993439475,
      "grad_norm": 0.5429418087005615,
      "learning_rate": 2.319213661470346e-05,
      "loss": 0.6516,
      "step": 1935
    },
    {
      "epoch": 2.4761980958476677,
      "grad_norm": 0.6022473573684692,
      "learning_rate": 2.3177256326142578e-05,
      "loss": 0.692,
      "step": 1936
    },
    {
      "epoch": 2.4774781982558602,
      "grad_norm": 0.5954020023345947,
      "learning_rate": 2.3162375229539395e-05,
      "loss": 0.6684,
      "step": 1937
    },
    {
      "epoch": 2.4787583006640532,
      "grad_norm": 0.5663207769393921,
      "learning_rate": 2.3147493335107964e-05,
      "loss": 0.6894,
      "step": 1938
    },
    {
      "epoch": 2.4800384030722458,
      "grad_norm": 0.6509982943534851,
      "learning_rate": 2.3132610653062873e-05,
      "loss": 0.6916,
      "step": 1939
    },
    {
      "epoch": 2.4813185054804383,
      "grad_norm": 0.5313803553581238,
      "learning_rate": 2.3117727193619278e-05,
      "loss": 0.685,
      "step": 1940
    },
    {
      "epoch": 2.4825986078886313,
      "grad_norm": 0.5990623831748962,
      "learning_rate": 2.3102842966992847e-05,
      "loss": 0.6927,
      "step": 1941
    },
    {
      "epoch": 2.483878710296824,
      "grad_norm": 0.6042869091033936,
      "learning_rate": 2.3087957983399762e-05,
      "loss": 0.6509,
      "step": 1942
    },
    {
      "epoch": 2.4851588127050164,
      "grad_norm": 0.5589537024497986,
      "learning_rate": 2.307307225305676e-05,
      "loss": 0.6858,
      "step": 1943
    },
    {
      "epoch": 2.486438915113209,
      "grad_norm": 0.579803466796875,
      "learning_rate": 2.3058185786181068e-05,
      "loss": 0.6742,
      "step": 1944
    },
    {
      "epoch": 2.487719017521402,
      "grad_norm": 0.5842740535736084,
      "learning_rate": 2.304329859299042e-05,
      "loss": 0.6594,
      "step": 1945
    },
    {
      "epoch": 2.4889991199295944,
      "grad_norm": 0.5804514288902283,
      "learning_rate": 2.3028410683703045e-05,
      "loss": 0.6705,
      "step": 1946
    },
    {
      "epoch": 2.490279222337787,
      "grad_norm": 0.5964694619178772,
      "learning_rate": 2.301352206853768e-05,
      "loss": 0.7093,
      "step": 1947
    },
    {
      "epoch": 2.4915593247459795,
      "grad_norm": 0.5650797486305237,
      "learning_rate": 2.2998632757713538e-05,
      "loss": 0.6565,
      "step": 1948
    },
    {
      "epoch": 2.4928394271541725,
      "grad_norm": 0.6101258993148804,
      "learning_rate": 2.2983742761450286e-05,
      "loss": 0.6905,
      "step": 1949
    },
    {
      "epoch": 2.494119529562365,
      "grad_norm": 0.5399543642997742,
      "learning_rate": 2.296885208996811e-05,
      "loss": 0.6694,
      "step": 1950
    },
    {
      "epoch": 2.494119529562365,
      "eval_loss": 0.01154709979891777,
      "eval_runtime": 10.8182,
      "eval_samples_per_second": 45.849,
      "eval_steps_per_second": 5.731,
      "step": 1950
    },
    {
      "epoch": 2.4953996319705576,
      "grad_norm": 0.5777952671051025,
      "learning_rate": 2.2953960753487623e-05,
      "loss": 0.6888,
      "step": 1951
    },
    {
      "epoch": 2.49667973437875,
      "grad_norm": 0.6142259836196899,
      "learning_rate": 2.2939068762229894e-05,
      "loss": 0.6948,
      "step": 1952
    },
    {
      "epoch": 2.497959836786943,
      "grad_norm": 0.5753744840621948,
      "learning_rate": 2.2924176126416463e-05,
      "loss": 0.6701,
      "step": 1953
    },
    {
      "epoch": 2.4992399391951357,
      "grad_norm": 0.6336653828620911,
      "learning_rate": 2.2909282856269298e-05,
      "loss": 0.6747,
      "step": 1954
    },
    {
      "epoch": 2.500520041603328,
      "grad_norm": 0.5536977648735046,
      "learning_rate": 2.2894388962010798e-05,
      "loss": 0.6711,
      "step": 1955
    },
    {
      "epoch": 2.501800144011521,
      "grad_norm": 0.6057724356651306,
      "learning_rate": 2.287949445386381e-05,
      "loss": 0.662,
      "step": 1956
    },
    {
      "epoch": 2.5030802464197137,
      "grad_norm": 0.6078692078590393,
      "learning_rate": 2.2864599342051584e-05,
      "loss": 0.6596,
      "step": 1957
    },
    {
      "epoch": 2.5043603488279063,
      "grad_norm": 0.5415582656860352,
      "learning_rate": 2.2849703636797788e-05,
      "loss": 0.6788,
      "step": 1958
    },
    {
      "epoch": 2.505640451236099,
      "grad_norm": 0.5932589173316956,
      "learning_rate": 2.2834807348326497e-05,
      "loss": 0.6573,
      "step": 1959
    },
    {
      "epoch": 2.5069205536442913,
      "grad_norm": 0.5789892077445984,
      "learning_rate": 2.2819910486862194e-05,
      "loss": 0.6725,
      "step": 1960
    },
    {
      "epoch": 2.5082006560524843,
      "grad_norm": 0.5233830809593201,
      "learning_rate": 2.280501306262975e-05,
      "loss": 0.6391,
      "step": 1961
    },
    {
      "epoch": 2.509480758460677,
      "grad_norm": 0.5784941911697388,
      "learning_rate": 2.279011508585442e-05,
      "loss": 0.6877,
      "step": 1962
    },
    {
      "epoch": 2.5107608608688694,
      "grad_norm": 0.576754629611969,
      "learning_rate": 2.2775216566761843e-05,
      "loss": 0.6868,
      "step": 1963
    },
    {
      "epoch": 2.5120409632770624,
      "grad_norm": 0.5483225584030151,
      "learning_rate": 2.2760317515578024e-05,
      "loss": 0.6789,
      "step": 1964
    },
    {
      "epoch": 2.513321065685255,
      "grad_norm": 0.5463926792144775,
      "learning_rate": 2.2745417942529343e-05,
      "loss": 0.6548,
      "step": 1965
    },
    {
      "epoch": 2.5146011680934475,
      "grad_norm": 0.5364739894866943,
      "learning_rate": 2.2730517857842526e-05,
      "loss": 0.6882,
      "step": 1966
    },
    {
      "epoch": 2.51588127050164,
      "grad_norm": 0.5220996737480164,
      "learning_rate": 2.271561727174467e-05,
      "loss": 0.6954,
      "step": 1967
    },
    {
      "epoch": 2.5171613729098326,
      "grad_norm": 0.5596213340759277,
      "learning_rate": 2.2700716194463177e-05,
      "loss": 0.6907,
      "step": 1968
    },
    {
      "epoch": 2.5184414753180255,
      "grad_norm": 0.5397947430610657,
      "learning_rate": 2.2685814636225837e-05,
      "loss": 0.6783,
      "step": 1969
    },
    {
      "epoch": 2.519721577726218,
      "grad_norm": 0.5562774538993835,
      "learning_rate": 2.2670912607260732e-05,
      "loss": 0.6575,
      "step": 1970
    },
    {
      "epoch": 2.5210016801344106,
      "grad_norm": 0.5784458518028259,
      "learning_rate": 2.2656010117796278e-05,
      "loss": 0.7069,
      "step": 1971
    },
    {
      "epoch": 2.5222817825426036,
      "grad_norm": 0.5621782541275024,
      "learning_rate": 2.2641107178061225e-05,
      "loss": 0.6993,
      "step": 1972
    },
    {
      "epoch": 2.523561884950796,
      "grad_norm": 0.5110465884208679,
      "learning_rate": 2.26262037982846e-05,
      "loss": 0.6715,
      "step": 1973
    },
    {
      "epoch": 2.5248419873589887,
      "grad_norm": 0.5507041215896606,
      "learning_rate": 2.2611299988695744e-05,
      "loss": 0.6913,
      "step": 1974
    },
    {
      "epoch": 2.5261220897671812,
      "grad_norm": 0.563943088054657,
      "learning_rate": 2.2596395759524316e-05,
      "loss": 0.6893,
      "step": 1975
    },
    {
      "epoch": 2.5274021921753738,
      "grad_norm": 0.5315508246421814,
      "learning_rate": 2.258149112100023e-05,
      "loss": 0.6292,
      "step": 1976
    },
    {
      "epoch": 2.5286822945835667,
      "grad_norm": 0.5950513482093811,
      "learning_rate": 2.25665860833537e-05,
      "loss": 0.712,
      "step": 1977
    },
    {
      "epoch": 2.5299623969917593,
      "grad_norm": 0.5365733504295349,
      "learning_rate": 2.2551680656815217e-05,
      "loss": 0.705,
      "step": 1978
    },
    {
      "epoch": 2.5312424993999523,
      "grad_norm": 0.6335738301277161,
      "learning_rate": 2.2536774851615507e-05,
      "loss": 0.7038,
      "step": 1979
    },
    {
      "epoch": 2.532522601808145,
      "grad_norm": 0.6017379760742188,
      "learning_rate": 2.252186867798561e-05,
      "loss": 0.7075,
      "step": 1980
    },
    {
      "epoch": 2.532522601808145,
      "eval_loss": 0.011548144742846489,
      "eval_runtime": 10.8069,
      "eval_samples_per_second": 45.897,
      "eval_steps_per_second": 5.737,
      "step": 1980
    },
    {
      "epoch": 2.5338027042163374,
      "grad_norm": 0.5744271278381348,
      "learning_rate": 2.2506962146156764e-05,
      "loss": 0.6894,
      "step": 1981
    },
    {
      "epoch": 2.53508280662453,
      "grad_norm": 0.6365877389907837,
      "learning_rate": 2.249205526636049e-05,
      "loss": 0.7013,
      "step": 1982
    },
    {
      "epoch": 2.5363629090327224,
      "grad_norm": 0.5367043018341064,
      "learning_rate": 2.2477148048828536e-05,
      "loss": 0.6719,
      "step": 1983
    },
    {
      "epoch": 2.5376430114409154,
      "grad_norm": 0.6105454564094543,
      "learning_rate": 2.2462240503792882e-05,
      "loss": 0.6941,
      "step": 1984
    },
    {
      "epoch": 2.538923113849108,
      "grad_norm": 0.5651337504386902,
      "learning_rate": 2.2447332641485717e-05,
      "loss": 0.6705,
      "step": 1985
    },
    {
      "epoch": 2.5402032162573005,
      "grad_norm": 0.5650824904441833,
      "learning_rate": 2.2432424472139483e-05,
      "loss": 0.67,
      "step": 1986
    },
    {
      "epoch": 2.5414833186654935,
      "grad_norm": 0.5914794206619263,
      "learning_rate": 2.2417516005986806e-05,
      "loss": 0.6654,
      "step": 1987
    },
    {
      "epoch": 2.542763421073686,
      "grad_norm": 0.5018816590309143,
      "learning_rate": 2.2402607253260514e-05,
      "loss": 0.6412,
      "step": 1988
    },
    {
      "epoch": 2.5440435234818786,
      "grad_norm": 0.657543420791626,
      "learning_rate": 2.2387698224193653e-05,
      "loss": 0.6738,
      "step": 1989
    },
    {
      "epoch": 2.545323625890071,
      "grad_norm": 0.528235673904419,
      "learning_rate": 2.2372788929019432e-05,
      "loss": 0.6523,
      "step": 1990
    },
    {
      "epoch": 2.5466037282982636,
      "grad_norm": 0.6170380115509033,
      "learning_rate": 2.2357879377971265e-05,
      "loss": 0.6623,
      "step": 1991
    },
    {
      "epoch": 2.5478838307064566,
      "grad_norm": 0.6315262317657471,
      "learning_rate": 2.2342969581282734e-05,
      "loss": 0.6882,
      "step": 1992
    },
    {
      "epoch": 2.549163933114649,
      "grad_norm": 0.5583043098449707,
      "learning_rate": 2.2328059549187582e-05,
      "loss": 0.6935,
      "step": 1993
    },
    {
      "epoch": 2.5504440355228417,
      "grad_norm": 0.636369526386261,
      "learning_rate": 2.2313149291919726e-05,
      "loss": 0.6455,
      "step": 1994
    },
    {
      "epoch": 2.5517241379310347,
      "grad_norm": 0.5487728714942932,
      "learning_rate": 2.2298238819713234e-05,
      "loss": 0.6777,
      "step": 1995
    },
    {
      "epoch": 2.5530042403392272,
      "grad_norm": 0.6297091841697693,
      "learning_rate": 2.228332814280231e-05,
      "loss": 0.7039,
      "step": 1996
    },
    {
      "epoch": 2.55428434274742,
      "grad_norm": 0.6231347322463989,
      "learning_rate": 2.2268417271421315e-05,
      "loss": 0.6975,
      "step": 1997
    },
    {
      "epoch": 2.5555644451556123,
      "grad_norm": 0.538118302822113,
      "learning_rate": 2.2253506215804733e-05,
      "loss": 0.6506,
      "step": 1998
    },
    {
      "epoch": 2.556844547563805,
      "grad_norm": 0.5905261635780334,
      "learning_rate": 2.2238594986187177e-05,
      "loss": 0.7189,
      "step": 1999
    },
    {
      "epoch": 2.558124649971998,
      "grad_norm": 0.5073800086975098,
      "learning_rate": 2.2223683592803383e-05,
      "loss": 0.6605,
      "step": 2000
    },
    {
      "epoch": 2.5594047523801904,
      "grad_norm": 0.5784037113189697,
      "learning_rate": 2.2208772045888193e-05,
      "loss": 0.7136,
      "step": 2001
    },
    {
      "epoch": 2.560684854788383,
      "grad_norm": 0.5811653733253479,
      "learning_rate": 2.2193860355676563e-05,
      "loss": 0.6847,
      "step": 2002
    },
    {
      "epoch": 2.561964957196576,
      "grad_norm": 0.5476462841033936,
      "learning_rate": 2.2178948532403538e-05,
      "loss": 0.6695,
      "step": 2003
    },
    {
      "epoch": 2.5632450596047684,
      "grad_norm": 0.5981435179710388,
      "learning_rate": 2.2164036586304254e-05,
      "loss": 0.6816,
      "step": 2004
    },
    {
      "epoch": 2.564525162012961,
      "grad_norm": 0.5810021162033081,
      "learning_rate": 2.2149124527613953e-05,
      "loss": 0.6992,
      "step": 2005
    },
    {
      "epoch": 2.5658052644211535,
      "grad_norm": 0.5890992879867554,
      "learning_rate": 2.213421236656792e-05,
      "loss": 0.6655,
      "step": 2006
    },
    {
      "epoch": 2.567085366829346,
      "grad_norm": 0.546022891998291,
      "learning_rate": 2.2119300113401527e-05,
      "loss": 0.7123,
      "step": 2007
    },
    {
      "epoch": 2.568365469237539,
      "grad_norm": 0.5739374160766602,
      "learning_rate": 2.2104387778350226e-05,
      "loss": 0.708,
      "step": 2008
    },
    {
      "epoch": 2.5696455716457316,
      "grad_norm": 0.5746329426765442,
      "learning_rate": 2.2089475371649493e-05,
      "loss": 0.6881,
      "step": 2009
    },
    {
      "epoch": 2.5709256740539246,
      "grad_norm": 0.516336977481842,
      "learning_rate": 2.2074562903534883e-05,
      "loss": 0.6309,
      "step": 2010
    },
    {
      "epoch": 2.5709256740539246,
      "eval_loss": 0.011530693620443344,
      "eval_runtime": 10.8221,
      "eval_samples_per_second": 45.832,
      "eval_steps_per_second": 5.729,
      "step": 2010
    },
    {
      "epoch": 2.572205776462117,
      "grad_norm": 0.6161022186279297,
      "learning_rate": 2.2059650384241956e-05,
      "loss": 0.6715,
      "step": 2011
    },
    {
      "epoch": 2.5734858788703097,
      "grad_norm": 0.5700750350952148,
      "learning_rate": 2.2044737824006354e-05,
      "loss": 0.6906,
      "step": 2012
    },
    {
      "epoch": 2.574765981278502,
      "grad_norm": 0.5499996542930603,
      "learning_rate": 2.2029825233063716e-05,
      "loss": 0.6954,
      "step": 2013
    },
    {
      "epoch": 2.5760460836866947,
      "grad_norm": 0.5665808320045471,
      "learning_rate": 2.2014912621649702e-05,
      "loss": 0.7002,
      "step": 2014
    },
    {
      "epoch": 2.5773261860948877,
      "grad_norm": 0.583620548248291,
      "learning_rate": 2.2000000000000003e-05,
      "loss": 0.6931,
      "step": 2015
    },
    {
      "epoch": 2.5786062885030803,
      "grad_norm": 0.5770348310470581,
      "learning_rate": 2.19850873783503e-05,
      "loss": 0.7037,
      "step": 2016
    },
    {
      "epoch": 2.579886390911273,
      "grad_norm": 0.5996818542480469,
      "learning_rate": 2.19701747669363e-05,
      "loss": 0.6934,
      "step": 2017
    },
    {
      "epoch": 2.581166493319466,
      "grad_norm": 0.5798315405845642,
      "learning_rate": 2.195526217599365e-05,
      "loss": 0.6618,
      "step": 2018
    },
    {
      "epoch": 2.5824465957276583,
      "grad_norm": 0.5487486720085144,
      "learning_rate": 2.1940349615758046e-05,
      "loss": 0.6581,
      "step": 2019
    },
    {
      "epoch": 2.583726698135851,
      "grad_norm": 0.5613217353820801,
      "learning_rate": 2.1925437096465133e-05,
      "loss": 0.6696,
      "step": 2020
    },
    {
      "epoch": 2.5850068005440434,
      "grad_norm": 0.5560593008995056,
      "learning_rate": 2.1910524628350513e-05,
      "loss": 0.692,
      "step": 2021
    },
    {
      "epoch": 2.586286902952236,
      "grad_norm": 0.5987349152565002,
      "learning_rate": 2.189561222164978e-05,
      "loss": 0.7086,
      "step": 2022
    },
    {
      "epoch": 2.587567005360429,
      "grad_norm": 0.5258426070213318,
      "learning_rate": 2.188069988659848e-05,
      "loss": 0.6233,
      "step": 2023
    },
    {
      "epoch": 2.5888471077686215,
      "grad_norm": 0.559288501739502,
      "learning_rate": 2.1865787633432086e-05,
      "loss": 0.6872,
      "step": 2024
    },
    {
      "epoch": 2.590127210176814,
      "grad_norm": 0.5850808024406433,
      "learning_rate": 2.1850875472386056e-05,
      "loss": 0.6954,
      "step": 2025
    },
    {
      "epoch": 2.591407312585007,
      "grad_norm": 0.5960245132446289,
      "learning_rate": 2.183596341369575e-05,
      "loss": 0.707,
      "step": 2026
    },
    {
      "epoch": 2.5926874149931995,
      "grad_norm": 0.569518506526947,
      "learning_rate": 2.1821051467596467e-05,
      "loss": 0.6821,
      "step": 2027
    },
    {
      "epoch": 2.593967517401392,
      "grad_norm": 0.6339453458786011,
      "learning_rate": 2.180613964432344e-05,
      "loss": 0.7125,
      "step": 2028
    },
    {
      "epoch": 2.5952476198095846,
      "grad_norm": 0.5328766703605652,
      "learning_rate": 2.1791227954111816e-05,
      "loss": 0.6793,
      "step": 2029
    },
    {
      "epoch": 2.596527722217777,
      "grad_norm": 0.5256121754646301,
      "learning_rate": 2.1776316407196626e-05,
      "loss": 0.6838,
      "step": 2030
    },
    {
      "epoch": 2.59780782462597,
      "grad_norm": 0.5012363791465759,
      "learning_rate": 2.1761405013812836e-05,
      "loss": 0.662,
      "step": 2031
    },
    {
      "epoch": 2.5990879270341627,
      "grad_norm": 0.5150095820426941,
      "learning_rate": 2.174649378419528e-05,
      "loss": 0.6852,
      "step": 2032
    },
    {
      "epoch": 2.6003680294423552,
      "grad_norm": 0.5986401438713074,
      "learning_rate": 2.1731582728578694e-05,
      "loss": 0.699,
      "step": 2033
    },
    {
      "epoch": 2.601648131850548,
      "grad_norm": 0.5160393118858337,
      "learning_rate": 2.17166718571977e-05,
      "loss": 0.6918,
      "step": 2034
    },
    {
      "epoch": 2.6029282342587408,
      "grad_norm": 0.5472270250320435,
      "learning_rate": 2.1701761180286775e-05,
      "loss": 0.6787,
      "step": 2035
    },
    {
      "epoch": 2.6042083366669333,
      "grad_norm": 0.5348479747772217,
      "learning_rate": 2.1686850708080276e-05,
      "loss": 0.6631,
      "step": 2036
    },
    {
      "epoch": 2.605488439075126,
      "grad_norm": 0.5178807377815247,
      "learning_rate": 2.1671940450812427e-05,
      "loss": 0.6643,
      "step": 2037
    },
    {
      "epoch": 2.606768541483319,
      "grad_norm": 0.5322620868682861,
      "learning_rate": 2.1657030418717275e-05,
      "loss": 0.6771,
      "step": 2038
    },
    {
      "epoch": 2.6080486438915114,
      "grad_norm": 0.5004392862319946,
      "learning_rate": 2.1642120622028737e-05,
      "loss": 0.6639,
      "step": 2039
    },
    {
      "epoch": 2.609328746299704,
      "grad_norm": 0.5184484124183655,
      "learning_rate": 2.162721107098058e-05,
      "loss": 0.6604,
      "step": 2040
    },
    {
      "epoch": 2.609328746299704,
      "eval_loss": 0.011514472775161266,
      "eval_runtime": 10.8316,
      "eval_samples_per_second": 45.792,
      "eval_steps_per_second": 5.724,
      "step": 2040
    },
    {
      "epoch": 2.610608848707897,
      "grad_norm": 0.5928523540496826,
      "learning_rate": 2.161230177580636e-05,
      "loss": 0.6964,
      "step": 2041
    },
    {
      "epoch": 2.6118889511160894,
      "grad_norm": 0.5536078810691833,
      "learning_rate": 2.1597392746739492e-05,
      "loss": 0.6719,
      "step": 2042
    },
    {
      "epoch": 2.613169053524282,
      "grad_norm": 0.5903907418251038,
      "learning_rate": 2.1582483994013207e-05,
      "loss": 0.6985,
      "step": 2043
    },
    {
      "epoch": 2.6144491559324745,
      "grad_norm": 0.5517407059669495,
      "learning_rate": 2.156757552786052e-05,
      "loss": 0.6997,
      "step": 2044
    },
    {
      "epoch": 2.615729258340667,
      "grad_norm": 0.5654640197753906,
      "learning_rate": 2.1552667358514285e-05,
      "loss": 0.6816,
      "step": 2045
    },
    {
      "epoch": 2.61700936074886,
      "grad_norm": 0.5533310770988464,
      "learning_rate": 2.153775949620713e-05,
      "loss": 0.6586,
      "step": 2046
    },
    {
      "epoch": 2.6182894631570526,
      "grad_norm": 0.5827234387397766,
      "learning_rate": 2.152285195117147e-05,
      "loss": 0.6718,
      "step": 2047
    },
    {
      "epoch": 2.619569565565245,
      "grad_norm": 0.5745716691017151,
      "learning_rate": 2.1507944733639513e-05,
      "loss": 0.702,
      "step": 2048
    },
    {
      "epoch": 2.620849667973438,
      "grad_norm": 0.5552913546562195,
      "learning_rate": 2.149303785384324e-05,
      "loss": 0.7033,
      "step": 2049
    },
    {
      "epoch": 2.6221297703816306,
      "grad_norm": 0.5924190282821655,
      "learning_rate": 2.14781313220144e-05,
      "loss": 0.6385,
      "step": 2050
    },
    {
      "epoch": 2.623409872789823,
      "grad_norm": 0.516633927822113,
      "learning_rate": 2.1463225148384495e-05,
      "loss": 0.6736,
      "step": 2051
    },
    {
      "epoch": 2.6246899751980157,
      "grad_norm": 0.6046395301818848,
      "learning_rate": 2.1448319343184792e-05,
      "loss": 0.6833,
      "step": 2052
    },
    {
      "epoch": 2.6259700776062083,
      "grad_norm": 0.547670304775238,
      "learning_rate": 2.1433413916646305e-05,
      "loss": 0.703,
      "step": 2053
    },
    {
      "epoch": 2.6272501800144012,
      "grad_norm": 0.5369134545326233,
      "learning_rate": 2.1418508878999774e-05,
      "loss": 0.6661,
      "step": 2054
    },
    {
      "epoch": 2.628530282422594,
      "grad_norm": 0.5275795459747314,
      "learning_rate": 2.140360424047569e-05,
      "loss": 0.6534,
      "step": 2055
    },
    {
      "epoch": 2.6298103848307863,
      "grad_norm": 0.5670477747917175,
      "learning_rate": 2.138870001130426e-05,
      "loss": 0.6576,
      "step": 2056
    },
    {
      "epoch": 2.6310904872389793,
      "grad_norm": 0.5423495173454285,
      "learning_rate": 2.1373796201715415e-05,
      "loss": 0.6454,
      "step": 2057
    },
    {
      "epoch": 2.632370589647172,
      "grad_norm": 0.5598854422569275,
      "learning_rate": 2.1358892821938784e-05,
      "loss": 0.6621,
      "step": 2058
    },
    {
      "epoch": 2.6336506920553644,
      "grad_norm": 0.5291751623153687,
      "learning_rate": 2.1343989882203728e-05,
      "loss": 0.6952,
      "step": 2059
    },
    {
      "epoch": 2.634930794463557,
      "grad_norm": 0.5671200156211853,
      "learning_rate": 2.1329087392739277e-05,
      "loss": 0.6884,
      "step": 2060
    },
    {
      "epoch": 2.6362108968717495,
      "grad_norm": 0.5808565020561218,
      "learning_rate": 2.1314185363774168e-05,
      "loss": 0.6867,
      "step": 2061
    },
    {
      "epoch": 2.6374909992799425,
      "grad_norm": 0.5669856667518616,
      "learning_rate": 2.1299283805536825e-05,
      "loss": 0.7044,
      "step": 2062
    },
    {
      "epoch": 2.638771101688135,
      "grad_norm": 0.5911791324615479,
      "learning_rate": 2.1284382728255343e-05,
      "loss": 0.6779,
      "step": 2063
    },
    {
      "epoch": 2.6400512040963275,
      "grad_norm": 0.5535954236984253,
      "learning_rate": 2.1269482142157476e-05,
      "loss": 0.6743,
      "step": 2064
    },
    {
      "epoch": 2.6413313065045205,
      "grad_norm": 0.5271023511886597,
      "learning_rate": 2.125458205747066e-05,
      "loss": 0.6701,
      "step": 2065
    },
    {
      "epoch": 2.642611408912713,
      "grad_norm": 0.5959221124649048,
      "learning_rate": 2.123968248442198e-05,
      "loss": 0.6852,
      "step": 2066
    },
    {
      "epoch": 2.6438915113209056,
      "grad_norm": 0.5608366131782532,
      "learning_rate": 2.1224783433238163e-05,
      "loss": 0.6901,
      "step": 2067
    },
    {
      "epoch": 2.645171613729098,
      "grad_norm": 0.545486569404602,
      "learning_rate": 2.1209884914145588e-05,
      "loss": 0.7153,
      "step": 2068
    },
    {
      "epoch": 2.646451716137291,
      "grad_norm": 0.5992303490638733,
      "learning_rate": 2.119498693737026e-05,
      "loss": 0.6782,
      "step": 2069
    },
    {
      "epoch": 2.6477318185454837,
      "grad_norm": 0.5763704180717468,
      "learning_rate": 2.1180089513137815e-05,
      "loss": 0.6815,
      "step": 2070
    },
    {
      "epoch": 2.6477318185454837,
      "eval_loss": 0.011513450182974339,
      "eval_runtime": 10.8052,
      "eval_samples_per_second": 45.904,
      "eval_steps_per_second": 5.738,
      "step": 2070
    },
    {
      "epoch": 2.649011920953676,
      "grad_norm": 0.5506516695022583,
      "learning_rate": 2.116519265167351e-05,
      "loss": 0.6864,
      "step": 2071
    },
    {
      "epoch": 2.650292023361869,
      "grad_norm": 0.6050803661346436,
      "learning_rate": 2.1150296363202224e-05,
      "loss": 0.687,
      "step": 2072
    },
    {
      "epoch": 2.6515721257700617,
      "grad_norm": 0.5962046980857849,
      "learning_rate": 2.113540065794842e-05,
      "loss": 0.6871,
      "step": 2073
    },
    {
      "epoch": 2.6528522281782543,
      "grad_norm": 0.5660064816474915,
      "learning_rate": 2.112050554613619e-05,
      "loss": 0.6847,
      "step": 2074
    },
    {
      "epoch": 2.654132330586447,
      "grad_norm": 0.5870113968849182,
      "learning_rate": 2.1105611037989208e-05,
      "loss": 0.6992,
      "step": 2075
    },
    {
      "epoch": 2.6554124329946394,
      "grad_norm": 0.5855358242988586,
      "learning_rate": 2.1090717143730708e-05,
      "loss": 0.6828,
      "step": 2076
    },
    {
      "epoch": 2.6566925354028323,
      "grad_norm": 0.6082198023796082,
      "learning_rate": 2.107582387358354e-05,
      "loss": 0.6838,
      "step": 2077
    },
    {
      "epoch": 2.657972637811025,
      "grad_norm": 0.5873337388038635,
      "learning_rate": 2.106093123777011e-05,
      "loss": 0.664,
      "step": 2078
    },
    {
      "epoch": 2.6592527402192174,
      "grad_norm": 0.589705228805542,
      "learning_rate": 2.1046039246512383e-05,
      "loss": 0.6912,
      "step": 2079
    },
    {
      "epoch": 2.6605328426274104,
      "grad_norm": 0.5821221470832825,
      "learning_rate": 2.103114791003189e-05,
      "loss": 0.7278,
      "step": 2080
    },
    {
      "epoch": 2.661812945035603,
      "grad_norm": 0.5650155544281006,
      "learning_rate": 2.1016257238549716e-05,
      "loss": 0.7029,
      "step": 2081
    },
    {
      "epoch": 2.6630930474437955,
      "grad_norm": 0.5573188662528992,
      "learning_rate": 2.100136724228647e-05,
      "loss": 0.7012,
      "step": 2082
    },
    {
      "epoch": 2.664373149851988,
      "grad_norm": 0.5489970445632935,
      "learning_rate": 2.098647793146233e-05,
      "loss": 0.6912,
      "step": 2083
    },
    {
      "epoch": 2.6656532522601806,
      "grad_norm": 0.5229962468147278,
      "learning_rate": 2.097158931629696e-05,
      "loss": 0.6467,
      "step": 2084
    },
    {
      "epoch": 2.6669333546683736,
      "grad_norm": 0.5697285532951355,
      "learning_rate": 2.095670140700959e-05,
      "loss": 0.6912,
      "step": 2085
    },
    {
      "epoch": 2.668213457076566,
      "grad_norm": 0.523398220539093,
      "learning_rate": 2.0941814213818944e-05,
      "loss": 0.7016,
      "step": 2086
    },
    {
      "epoch": 2.6694935594847586,
      "grad_norm": 0.5511962175369263,
      "learning_rate": 2.0926927746943243e-05,
      "loss": 0.6761,
      "step": 2087
    },
    {
      "epoch": 2.6707736618929516,
      "grad_norm": 0.5299444198608398,
      "learning_rate": 2.0912042016600237e-05,
      "loss": 0.6643,
      "step": 2088
    },
    {
      "epoch": 2.672053764301144,
      "grad_norm": 0.5698947906494141,
      "learning_rate": 2.0897157033007166e-05,
      "loss": 0.6952,
      "step": 2089
    },
    {
      "epoch": 2.6733338667093367,
      "grad_norm": 0.534382700920105,
      "learning_rate": 2.0882272806380728e-05,
      "loss": 0.6813,
      "step": 2090
    },
    {
      "epoch": 2.6746139691175292,
      "grad_norm": 0.5265630483627319,
      "learning_rate": 2.086738934693713e-05,
      "loss": 0.6847,
      "step": 2091
    },
    {
      "epoch": 2.675894071525722,
      "grad_norm": 0.5158559083938599,
      "learning_rate": 2.0852506664892048e-05,
      "loss": 0.6887,
      "step": 2092
    },
    {
      "epoch": 2.6771741739339148,
      "grad_norm": 0.5293145775794983,
      "learning_rate": 2.083762477046061e-05,
      "loss": 0.6698,
      "step": 2093
    },
    {
      "epoch": 2.6784542763421073,
      "grad_norm": 0.5267837047576904,
      "learning_rate": 2.0822743673857424e-05,
      "loss": 0.6506,
      "step": 2094
    },
    {
      "epoch": 2.6797343787503003,
      "grad_norm": 0.515688955783844,
      "learning_rate": 2.0807863385296544e-05,
      "loss": 0.6693,
      "step": 2095
    },
    {
      "epoch": 2.681014481158493,
      "grad_norm": 0.5605611205101013,
      "learning_rate": 2.0792983914991453e-05,
      "loss": 0.6627,
      "step": 2096
    },
    {
      "epoch": 2.6822945835666854,
      "grad_norm": 0.5887176990509033,
      "learning_rate": 2.0778105273155095e-05,
      "loss": 0.6672,
      "step": 2097
    },
    {
      "epoch": 2.683574685974878,
      "grad_norm": 0.5191556811332703,
      "learning_rate": 2.0763227469999834e-05,
      "loss": 0.6873,
      "step": 2098
    },
    {
      "epoch": 2.6848547883830705,
      "grad_norm": 0.6014405488967896,
      "learning_rate": 2.0748350515737456e-05,
      "loss": 0.6465,
      "step": 2099
    },
    {
      "epoch": 2.6861348907912634,
      "grad_norm": 0.5366295576095581,
      "learning_rate": 2.0733474420579173e-05,
      "loss": 0.7184,
      "step": 2100
    },
    {
      "epoch": 2.6861348907912634,
      "eval_loss": 0.011498779058456421,
      "eval_runtime": 10.8231,
      "eval_samples_per_second": 45.828,
      "eval_steps_per_second": 5.728,
      "step": 2100
    },
    {
      "epoch": 2.687414993199456,
      "grad_norm": 0.5625792741775513,
      "learning_rate": 2.07185991947356e-05,
      "loss": 0.6684,
      "step": 2101
    },
    {
      "epoch": 2.6886950956076485,
      "grad_norm": 0.5656289458274841,
      "learning_rate": 2.0703724848416757e-05,
      "loss": 0.6989,
      "step": 2102
    },
    {
      "epoch": 2.6899751980158415,
      "grad_norm": 0.5664001107215881,
      "learning_rate": 2.0688851391832058e-05,
      "loss": 0.7038,
      "step": 2103
    },
    {
      "epoch": 2.691255300424034,
      "grad_norm": 0.6160801649093628,
      "learning_rate": 2.0673978835190324e-05,
      "loss": 0.7173,
      "step": 2104
    },
    {
      "epoch": 2.6925354028322266,
      "grad_norm": 0.5975323915481567,
      "learning_rate": 2.0659107188699722e-05,
      "loss": 0.6824,
      "step": 2105
    },
    {
      "epoch": 2.693815505240419,
      "grad_norm": 0.5666054487228394,
      "learning_rate": 2.0644236462567836e-05,
      "loss": 0.6513,
      "step": 2106
    },
    {
      "epoch": 2.6950956076486117,
      "grad_norm": 0.5504162311553955,
      "learning_rate": 2.062936666700159e-05,
      "loss": 0.6821,
      "step": 2107
    },
    {
      "epoch": 2.6963757100568047,
      "grad_norm": 0.4987417757511139,
      "learning_rate": 2.0614497812207275e-05,
      "loss": 0.6465,
      "step": 2108
    },
    {
      "epoch": 2.697655812464997,
      "grad_norm": 0.6021193861961365,
      "learning_rate": 2.0599629908390545e-05,
      "loss": 0.6774,
      "step": 2109
    },
    {
      "epoch": 2.6989359148731897,
      "grad_norm": 0.581078290939331,
      "learning_rate": 2.058476296575639e-05,
      "loss": 0.71,
      "step": 2110
    },
    {
      "epoch": 2.7002160172813827,
      "grad_norm": 0.5573481917381287,
      "learning_rate": 2.0569896994509146e-05,
      "loss": 0.7002,
      "step": 2111
    },
    {
      "epoch": 2.7014961196895753,
      "grad_norm": 0.5607424378395081,
      "learning_rate": 2.055503200485249e-05,
      "loss": 0.695,
      "step": 2112
    },
    {
      "epoch": 2.702776222097768,
      "grad_norm": 0.6108943819999695,
      "learning_rate": 2.05401680069894e-05,
      "loss": 0.6971,
      "step": 2113
    },
    {
      "epoch": 2.7040563245059603,
      "grad_norm": 0.6459068655967712,
      "learning_rate": 2.0525305011122202e-05,
      "loss": 0.6911,
      "step": 2114
    },
    {
      "epoch": 2.705336426914153,
      "grad_norm": 0.5489556193351746,
      "learning_rate": 2.051044302745253e-05,
      "loss": 0.658,
      "step": 2115
    },
    {
      "epoch": 2.706616529322346,
      "grad_norm": 0.5386958718299866,
      "learning_rate": 2.04955820661813e-05,
      "loss": 0.6594,
      "step": 2116
    },
    {
      "epoch": 2.7078966317305384,
      "grad_norm": 0.5512974262237549,
      "learning_rate": 2.0480722137508754e-05,
      "loss": 0.6526,
      "step": 2117
    },
    {
      "epoch": 2.709176734138731,
      "grad_norm": 0.5657424330711365,
      "learning_rate": 2.046586325163441e-05,
      "loss": 0.7312,
      "step": 2118
    },
    {
      "epoch": 2.710456836546924,
      "grad_norm": 0.5624411702156067,
      "learning_rate": 2.0451005418757058e-05,
      "loss": 0.6464,
      "step": 2119
    },
    {
      "epoch": 2.7117369389551165,
      "grad_norm": 0.5503666996955872,
      "learning_rate": 2.043614864907481e-05,
      "loss": 0.6633,
      "step": 2120
    },
    {
      "epoch": 2.713017041363309,
      "grad_norm": 0.5300235748291016,
      "learning_rate": 2.0421292952784996e-05,
      "loss": 0.6531,
      "step": 2121
    },
    {
      "epoch": 2.7142971437715016,
      "grad_norm": 0.5729533433914185,
      "learning_rate": 2.040643834008424e-05,
      "loss": 0.6629,
      "step": 2122
    },
    {
      "epoch": 2.7155772461796945,
      "grad_norm": 0.5653989315032959,
      "learning_rate": 2.0391584821168413e-05,
      "loss": 0.7041,
      "step": 2123
    },
    {
      "epoch": 2.716857348587887,
      "grad_norm": 0.5225719809532166,
      "learning_rate": 2.037673240623264e-05,
      "loss": 0.6805,
      "step": 2124
    },
    {
      "epoch": 2.7181374509960796,
      "grad_norm": 0.5666506290435791,
      "learning_rate": 2.0361881105471286e-05,
      "loss": 0.7004,
      "step": 2125
    },
    {
      "epoch": 2.7194175534042726,
      "grad_norm": 0.5524999499320984,
      "learning_rate": 2.0347030929077943e-05,
      "loss": 0.6828,
      "step": 2126
    },
    {
      "epoch": 2.720697655812465,
      "grad_norm": 0.547503650188446,
      "learning_rate": 2.0332181887245443e-05,
      "loss": 0.6429,
      "step": 2127
    },
    {
      "epoch": 2.7219777582206577,
      "grad_norm": 0.5649715065956116,
      "learning_rate": 2.0317333990165838e-05,
      "loss": 0.6497,
      "step": 2128
    },
    {
      "epoch": 2.7232578606288502,
      "grad_norm": 0.5688164830207825,
      "learning_rate": 2.0302487248030385e-05,
      "loss": 0.7109,
      "step": 2129
    },
    {
      "epoch": 2.7245379630370428,
      "grad_norm": 0.5821229219436646,
      "learning_rate": 2.0287641671029564e-05,
      "loss": 0.6687,
      "step": 2130
    },
    {
      "epoch": 2.7245379630370428,
      "eval_loss": 0.011482841335237026,
      "eval_runtime": 10.808,
      "eval_samples_per_second": 45.892,
      "eval_steps_per_second": 5.736,
      "step": 2130
    },
    {
      "epoch": 2.7258180654452357,
      "grad_norm": 0.5121495723724365,
      "learning_rate": 2.0272797269353033e-05,
      "loss": 0.6682,
      "step": 2131
    },
    {
      "epoch": 2.7270981678534283,
      "grad_norm": 0.5901171565055847,
      "learning_rate": 2.0257954053189657e-05,
      "loss": 0.6803,
      "step": 2132
    },
    {
      "epoch": 2.728378270261621,
      "grad_norm": 0.5494524836540222,
      "learning_rate": 2.02431120327275e-05,
      "loss": 0.6676,
      "step": 2133
    },
    {
      "epoch": 2.729658372669814,
      "grad_norm": 0.5600875020027161,
      "learning_rate": 2.0228271218153774e-05,
      "loss": 0.7007,
      "step": 2134
    },
    {
      "epoch": 2.7309384750780064,
      "grad_norm": 0.5640694499015808,
      "learning_rate": 2.0213431619654894e-05,
      "loss": 0.6878,
      "step": 2135
    },
    {
      "epoch": 2.732218577486199,
      "grad_norm": 0.5384992361068726,
      "learning_rate": 2.019859324741642e-05,
      "loss": 0.6405,
      "step": 2136
    },
    {
      "epoch": 2.7334986798943914,
      "grad_norm": 0.5485089421272278,
      "learning_rate": 2.0183756111623084e-05,
      "loss": 0.6673,
      "step": 2137
    },
    {
      "epoch": 2.734778782302584,
      "grad_norm": 0.5543821454048157,
      "learning_rate": 2.0168920222458754e-05,
      "loss": 0.6415,
      "step": 2138
    },
    {
      "epoch": 2.736058884710777,
      "grad_norm": 0.6076808571815491,
      "learning_rate": 2.0154085590106455e-05,
      "loss": 0.6935,
      "step": 2139
    },
    {
      "epoch": 2.7373389871189695,
      "grad_norm": 0.593870222568512,
      "learning_rate": 2.013925222474835e-05,
      "loss": 0.6667,
      "step": 2140
    },
    {
      "epoch": 2.738619089527162,
      "grad_norm": 0.6134325265884399,
      "learning_rate": 2.0124420136565717e-05,
      "loss": 0.6887,
      "step": 2141
    },
    {
      "epoch": 2.739899191935355,
      "grad_norm": 0.6129072308540344,
      "learning_rate": 2.010958933573898e-05,
      "loss": 0.6376,
      "step": 2142
    },
    {
      "epoch": 2.7411792943435476,
      "grad_norm": 0.604393720626831,
      "learning_rate": 2.009475983244766e-05,
      "loss": 0.7126,
      "step": 2143
    },
    {
      "epoch": 2.74245939675174,
      "grad_norm": 0.566626787185669,
      "learning_rate": 2.00799316368704e-05,
      "loss": 0.6841,
      "step": 2144
    },
    {
      "epoch": 2.7437394991599326,
      "grad_norm": 0.5514710545539856,
      "learning_rate": 2.0065104759184934e-05,
      "loss": 0.6639,
      "step": 2145
    },
    {
      "epoch": 2.745019601568125,
      "grad_norm": 0.5645005106925964,
      "learning_rate": 2.0050279209568097e-05,
      "loss": 0.6392,
      "step": 2146
    },
    {
      "epoch": 2.746299703976318,
      "grad_norm": 0.5400156378746033,
      "learning_rate": 2.0035454998195832e-05,
      "loss": 0.6235,
      "step": 2147
    },
    {
      "epoch": 2.7475798063845107,
      "grad_norm": 0.5787555575370789,
      "learning_rate": 2.002063213524312e-05,
      "loss": 0.6649,
      "step": 2148
    },
    {
      "epoch": 2.7488599087927033,
      "grad_norm": 0.5392046570777893,
      "learning_rate": 2.000581063088405e-05,
      "loss": 0.7122,
      "step": 2149
    },
    {
      "epoch": 2.7501400112008962,
      "grad_norm": 0.5718721747398376,
      "learning_rate": 1.999099049529178e-05,
      "loss": 0.656,
      "step": 2150
    },
    {
      "epoch": 2.751420113609089,
      "grad_norm": 0.5379940867424011,
      "learning_rate": 1.9976171738638504e-05,
      "loss": 0.6869,
      "step": 2151
    },
    {
      "epoch": 2.7527002160172813,
      "grad_norm": 0.5417284369468689,
      "learning_rate": 1.996135437109549e-05,
      "loss": 0.6845,
      "step": 2152
    },
    {
      "epoch": 2.753980318425474,
      "grad_norm": 0.5735756754875183,
      "learning_rate": 1.9946538402833047e-05,
      "loss": 0.6431,
      "step": 2153
    },
    {
      "epoch": 2.755260420833667,
      "grad_norm": 0.5329331755638123,
      "learning_rate": 1.993172384402052e-05,
      "loss": 0.6568,
      "step": 2154
    },
    {
      "epoch": 2.7565405232418594,
      "grad_norm": 0.5652323961257935,
      "learning_rate": 1.991691070482629e-05,
      "loss": 0.6733,
      "step": 2155
    },
    {
      "epoch": 2.757820625650052,
      "grad_norm": 0.5698742866516113,
      "learning_rate": 1.990209899541777e-05,
      "loss": 0.6936,
      "step": 2156
    },
    {
      "epoch": 2.759100728058245,
      "grad_norm": 0.5140464305877686,
      "learning_rate": 1.9887288725961374e-05,
      "loss": 0.6926,
      "step": 2157
    },
    {
      "epoch": 2.7603808304664375,
      "grad_norm": 0.5658466815948486,
      "learning_rate": 1.9872479906622538e-05,
      "loss": 0.6972,
      "step": 2158
    },
    {
      "epoch": 2.76166093287463,
      "grad_norm": 0.542084276676178,
      "learning_rate": 1.985767254756572e-05,
      "loss": 0.6793,
      "step": 2159
    },
    {
      "epoch": 2.7629410352828225,
      "grad_norm": 0.514428436756134,
      "learning_rate": 1.9842866658954336e-05,
      "loss": 0.6798,
      "step": 2160
    },
    {
      "epoch": 2.7629410352828225,
      "eval_loss": 0.011475411243736744,
      "eval_runtime": 10.8142,
      "eval_samples_per_second": 45.866,
      "eval_steps_per_second": 5.733,
      "step": 2160
    },
    {
      "epoch": 2.764221137691015,
      "grad_norm": 0.5466222167015076,
      "learning_rate": 1.982806225095083e-05,
      "loss": 0.6618,
      "step": 2161
    },
    {
      "epoch": 2.765501240099208,
      "grad_norm": 0.5842472314834595,
      "learning_rate": 1.981325933371661e-05,
      "loss": 0.6795,
      "step": 2162
    },
    {
      "epoch": 2.7667813425074006,
      "grad_norm": 0.5357396602630615,
      "learning_rate": 1.979845791741206e-05,
      "loss": 0.6719,
      "step": 2163
    },
    {
      "epoch": 2.768061444915593,
      "grad_norm": 0.5639327168464661,
      "learning_rate": 1.9783658012196563e-05,
      "loss": 0.7064,
      "step": 2164
    },
    {
      "epoch": 2.769341547323786,
      "grad_norm": 0.5845109820365906,
      "learning_rate": 1.9768859628228414e-05,
      "loss": 0.707,
      "step": 2165
    },
    {
      "epoch": 2.7706216497319787,
      "grad_norm": 0.5103625059127808,
      "learning_rate": 1.97540627756649e-05,
      "loss": 0.6566,
      "step": 2166
    },
    {
      "epoch": 2.771901752140171,
      "grad_norm": 0.5660591125488281,
      "learning_rate": 1.973926746466226e-05,
      "loss": 0.6662,
      "step": 2167
    },
    {
      "epoch": 2.7731818545483637,
      "grad_norm": 0.5204382538795471,
      "learning_rate": 1.9724473705375645e-05,
      "loss": 0.6914,
      "step": 2168
    },
    {
      "epoch": 2.7744619569565563,
      "grad_norm": 0.5099734663963318,
      "learning_rate": 1.9709681507959174e-05,
      "loss": 0.6996,
      "step": 2169
    },
    {
      "epoch": 2.7757420593647493,
      "grad_norm": 0.5796478986740112,
      "learning_rate": 1.9694890882565874e-05,
      "loss": 0.6934,
      "step": 2170
    },
    {
      "epoch": 2.777022161772942,
      "grad_norm": 0.5376036167144775,
      "learning_rate": 1.968010183934769e-05,
      "loss": 0.677,
      "step": 2171
    },
    {
      "epoch": 2.7783022641811344,
      "grad_norm": 0.5756266713142395,
      "learning_rate": 1.9665314388455498e-05,
      "loss": 0.6809,
      "step": 2172
    },
    {
      "epoch": 2.7795823665893273,
      "grad_norm": 0.5075539946556091,
      "learning_rate": 1.9650528540039077e-05,
      "loss": 0.6694,
      "step": 2173
    },
    {
      "epoch": 2.78086246899752,
      "grad_norm": 0.5039442181587219,
      "learning_rate": 1.963574430424709e-05,
      "loss": 0.6893,
      "step": 2174
    },
    {
      "epoch": 2.7821425714057124,
      "grad_norm": 0.5452905297279358,
      "learning_rate": 1.9620961691227106e-05,
      "loss": 0.6965,
      "step": 2175
    },
    {
      "epoch": 2.783422673813905,
      "grad_norm": 0.5262690782546997,
      "learning_rate": 1.960618071112559e-05,
      "loss": 0.6551,
      "step": 2176
    },
    {
      "epoch": 2.7847027762220975,
      "grad_norm": 0.49697020649909973,
      "learning_rate": 1.9591401374087855e-05,
      "loss": 0.6638,
      "step": 2177
    },
    {
      "epoch": 2.7859828786302905,
      "grad_norm": 0.5466814637184143,
      "learning_rate": 1.9576623690258117e-05,
      "loss": 0.711,
      "step": 2178
    },
    {
      "epoch": 2.787262981038483,
      "grad_norm": 0.5326719880104065,
      "learning_rate": 1.956184766977946e-05,
      "loss": 0.6363,
      "step": 2179
    },
    {
      "epoch": 2.788543083446676,
      "grad_norm": 0.52897047996521,
      "learning_rate": 1.9547073322793788e-05,
      "loss": 0.6677,
      "step": 2180
    },
    {
      "epoch": 2.7898231858548685,
      "grad_norm": 0.5449843406677246,
      "learning_rate": 1.9532300659441895e-05,
      "loss": 0.6682,
      "step": 2181
    },
    {
      "epoch": 2.791103288263061,
      "grad_norm": 0.5270798206329346,
      "learning_rate": 1.9517529689863406e-05,
      "loss": 0.6736,
      "step": 2182
    },
    {
      "epoch": 2.7923833906712536,
      "grad_norm": 0.5694547295570374,
      "learning_rate": 1.9502760424196773e-05,
      "loss": 0.6788,
      "step": 2183
    },
    {
      "epoch": 2.793663493079446,
      "grad_norm": 0.5537572503089905,
      "learning_rate": 1.9487992872579297e-05,
      "loss": 0.6846,
      "step": 2184
    },
    {
      "epoch": 2.794943595487639,
      "grad_norm": 0.46792593598365784,
      "learning_rate": 1.9473227045147096e-05,
      "loss": 0.6322,
      "step": 2185
    },
    {
      "epoch": 2.7962236978958317,
      "grad_norm": 0.520228385925293,
      "learning_rate": 1.9458462952035088e-05,
      "loss": 0.6903,
      "step": 2186
    },
    {
      "epoch": 2.7975038003040242,
      "grad_norm": 0.5050407648086548,
      "learning_rate": 1.9443700603377037e-05,
      "loss": 0.6958,
      "step": 2187
    },
    {
      "epoch": 2.798783902712217,
      "grad_norm": 0.4765312671661377,
      "learning_rate": 1.9428940009305462e-05,
      "loss": 0.6526,
      "step": 2188
    },
    {
      "epoch": 2.8000640051204098,
      "grad_norm": 0.521946907043457,
      "learning_rate": 1.9414181179951715e-05,
      "loss": 0.6242,
      "step": 2189
    },
    {
      "epoch": 2.8013441075286023,
      "grad_norm": 0.5431380867958069,
      "learning_rate": 1.939942412544594e-05,
      "loss": 0.6895,
      "step": 2190
    },
    {
      "epoch": 2.8013441075286023,
      "eval_loss": 0.01147318072617054,
      "eval_runtime": 10.841,
      "eval_samples_per_second": 45.752,
      "eval_steps_per_second": 5.719,
      "step": 2190
    },
    {
      "epoch": 2.802624209936795,
      "grad_norm": 0.5367740988731384,
      "learning_rate": 1.938466885591702e-05,
      "loss": 0.685,
      "step": 2191
    },
    {
      "epoch": 2.8039043123449874,
      "grad_norm": 0.5043015480041504,
      "learning_rate": 1.9369915381492655e-05,
      "loss": 0.6499,
      "step": 2192
    },
    {
      "epoch": 2.8051844147531804,
      "grad_norm": 0.5388321280479431,
      "learning_rate": 1.935516371229931e-05,
      "loss": 0.6536,
      "step": 2193
    },
    {
      "epoch": 2.806464517161373,
      "grad_norm": 0.5287108421325684,
      "learning_rate": 1.9340413858462174e-05,
      "loss": 0.6999,
      "step": 2194
    },
    {
      "epoch": 2.8077446195695654,
      "grad_norm": 0.5044527649879456,
      "learning_rate": 1.9325665830105232e-05,
      "loss": 0.6906,
      "step": 2195
    },
    {
      "epoch": 2.8090247219777584,
      "grad_norm": 0.5276654958724976,
      "learning_rate": 1.9310919637351202e-05,
      "loss": 0.6857,
      "step": 2196
    },
    {
      "epoch": 2.810304824385951,
      "grad_norm": 0.5158083438873291,
      "learning_rate": 1.9296175290321527e-05,
      "loss": 0.681,
      "step": 2197
    },
    {
      "epoch": 2.8115849267941435,
      "grad_norm": 0.5858495235443115,
      "learning_rate": 1.9281432799136398e-05,
      "loss": 0.7142,
      "step": 2198
    },
    {
      "epoch": 2.812865029202336,
      "grad_norm": 0.547903835773468,
      "learning_rate": 1.9266692173914747e-05,
      "loss": 0.6866,
      "step": 2199
    },
    {
      "epoch": 2.8141451316105286,
      "grad_norm": 0.5883592963218689,
      "learning_rate": 1.9251953424774185e-05,
      "loss": 0.7297,
      "step": 2200
    },
    {
      "epoch": 2.8154252340187216,
      "grad_norm": 0.5361905097961426,
      "learning_rate": 1.923721656183107e-05,
      "loss": 0.686,
      "step": 2201
    },
    {
      "epoch": 2.816705336426914,
      "grad_norm": 0.5558068156242371,
      "learning_rate": 1.9222481595200456e-05,
      "loss": 0.62,
      "step": 2202
    },
    {
      "epoch": 2.8179854388351067,
      "grad_norm": 0.5331078171730042,
      "learning_rate": 1.9207748534996086e-05,
      "loss": 0.6919,
      "step": 2203
    },
    {
      "epoch": 2.8192655412432996,
      "grad_norm": 0.5322447419166565,
      "learning_rate": 1.91930173913304e-05,
      "loss": 0.6701,
      "step": 2204
    },
    {
      "epoch": 2.820545643651492,
      "grad_norm": 0.5190942883491516,
      "learning_rate": 1.9178288174314535e-05,
      "loss": 0.6958,
      "step": 2205
    },
    {
      "epoch": 2.8218257460596847,
      "grad_norm": 0.5811094641685486,
      "learning_rate": 1.916356089405829e-05,
      "loss": 0.6871,
      "step": 2206
    },
    {
      "epoch": 2.8231058484678773,
      "grad_norm": 0.4891628324985504,
      "learning_rate": 1.914883556067013e-05,
      "loss": 0.6635,
      "step": 2207
    },
    {
      "epoch": 2.8243859508760703,
      "grad_norm": 0.5712688565254211,
      "learning_rate": 1.9134112184257208e-05,
      "loss": 0.657,
      "step": 2208
    },
    {
      "epoch": 2.825666053284263,
      "grad_norm": 0.5606522560119629,
      "learning_rate": 1.9119390774925315e-05,
      "loss": 0.6685,
      "step": 2209
    },
    {
      "epoch": 2.8269461556924553,
      "grad_norm": 0.5698298215866089,
      "learning_rate": 1.9104671342778883e-05,
      "loss": 0.6723,
      "step": 2210
    },
    {
      "epoch": 2.8282262581006483,
      "grad_norm": 0.6309710144996643,
      "learning_rate": 1.908995389792102e-05,
      "loss": 0.7099,
      "step": 2211
    },
    {
      "epoch": 2.829506360508841,
      "grad_norm": 0.5285724997520447,
      "learning_rate": 1.907523845045343e-05,
      "loss": 0.6682,
      "step": 2212
    },
    {
      "epoch": 2.8307864629170334,
      "grad_norm": 0.5792762041091919,
      "learning_rate": 1.906052501047648e-05,
      "loss": 0.6989,
      "step": 2213
    },
    {
      "epoch": 2.832066565325226,
      "grad_norm": 0.5514551997184753,
      "learning_rate": 1.9045813588089133e-05,
      "loss": 0.6749,
      "step": 2214
    },
    {
      "epoch": 2.8333466677334185,
      "grad_norm": 0.48829367756843567,
      "learning_rate": 1.903110419338899e-05,
      "loss": 0.6416,
      "step": 2215
    },
    {
      "epoch": 2.8346267701416115,
      "grad_norm": 0.5097596049308777,
      "learning_rate": 1.9016396836472248e-05,
      "loss": 0.6798,
      "step": 2216
    },
    {
      "epoch": 2.835906872549804,
      "grad_norm": 0.5160478949546814,
      "learning_rate": 1.900169152743369e-05,
      "loss": 0.6621,
      "step": 2217
    },
    {
      "epoch": 2.8371869749579965,
      "grad_norm": 0.5284377336502075,
      "learning_rate": 1.8986988276366732e-05,
      "loss": 0.6917,
      "step": 2218
    },
    {
      "epoch": 2.8384670773661895,
      "grad_norm": 0.5650047063827515,
      "learning_rate": 1.8972287093363345e-05,
      "loss": 0.6576,
      "step": 2219
    },
    {
      "epoch": 2.839747179774382,
      "grad_norm": 0.4914189279079437,
      "learning_rate": 1.895758798851409e-05,
      "loss": 0.6539,
      "step": 2220
    },
    {
      "epoch": 2.839747179774382,
      "eval_loss": 0.011458431370556355,
      "eval_runtime": 10.8251,
      "eval_samples_per_second": 45.82,
      "eval_steps_per_second": 5.727,
      "step": 2220
    },
    {
      "epoch": 2.8410272821825746,
      "grad_norm": 0.513734757900238,
      "learning_rate": 1.8942890971908105e-05,
      "loss": 0.6336,
      "step": 2221
    },
    {
      "epoch": 2.842307384590767,
      "grad_norm": 0.5714700818061829,
      "learning_rate": 1.892819605363309e-05,
      "loss": 0.717,
      "step": 2222
    },
    {
      "epoch": 2.8435874869989597,
      "grad_norm": 0.5283195376396179,
      "learning_rate": 1.891350324377531e-05,
      "loss": 0.6757,
      "step": 2223
    },
    {
      "epoch": 2.8448675894071527,
      "grad_norm": 0.5310732126235962,
      "learning_rate": 1.889881255241957e-05,
      "loss": 0.6614,
      "step": 2224
    },
    {
      "epoch": 2.846147691815345,
      "grad_norm": 0.5158543586730957,
      "learning_rate": 1.888412398964925e-05,
      "loss": 0.6406,
      "step": 2225
    },
    {
      "epoch": 2.8474277942235378,
      "grad_norm": 0.5152372717857361,
      "learning_rate": 1.886943756554623e-05,
      "loss": 0.6522,
      "step": 2226
    },
    {
      "epoch": 2.8487078966317307,
      "grad_norm": 0.5706690549850464,
      "learning_rate": 1.885475329019095e-05,
      "loss": 0.6851,
      "step": 2227
    },
    {
      "epoch": 2.8499879990399233,
      "grad_norm": 0.5428133010864258,
      "learning_rate": 1.8840071173662374e-05,
      "loss": 0.6936,
      "step": 2228
    },
    {
      "epoch": 2.851268101448116,
      "grad_norm": 0.6033597588539124,
      "learning_rate": 1.882539122603796e-05,
      "loss": 0.7267,
      "step": 2229
    },
    {
      "epoch": 2.8525482038563084,
      "grad_norm": 0.5333406925201416,
      "learning_rate": 1.88107134573937e-05,
      "loss": 0.7122,
      "step": 2230
    },
    {
      "epoch": 2.853828306264501,
      "grad_norm": 0.5083885192871094,
      "learning_rate": 1.8796037877804098e-05,
      "loss": 0.6591,
      "step": 2231
    },
    {
      "epoch": 2.855108408672694,
      "grad_norm": 0.5966604351997375,
      "learning_rate": 1.8781364497342128e-05,
      "loss": 0.6841,
      "step": 2232
    },
    {
      "epoch": 2.8563885110808864,
      "grad_norm": 0.573322594165802,
      "learning_rate": 1.8766693326079274e-05,
      "loss": 0.6358,
      "step": 2233
    },
    {
      "epoch": 2.857668613489079,
      "grad_norm": 0.5453227758407593,
      "learning_rate": 1.8752024374085503e-05,
      "loss": 0.7122,
      "step": 2234
    },
    {
      "epoch": 2.858948715897272,
      "grad_norm": 0.5487176179885864,
      "learning_rate": 1.873735765142924e-05,
      "loss": 0.6715,
      "step": 2235
    },
    {
      "epoch": 2.8602288183054645,
      "grad_norm": 0.522406280040741,
      "learning_rate": 1.87226931681774e-05,
      "loss": 0.673,
      "step": 2236
    },
    {
      "epoch": 2.861508920713657,
      "grad_norm": 0.5376841425895691,
      "learning_rate": 1.8708030934395373e-05,
      "loss": 0.7006,
      "step": 2237
    },
    {
      "epoch": 2.8627890231218496,
      "grad_norm": 0.5221294164657593,
      "learning_rate": 1.8693370960146958e-05,
      "loss": 0.6842,
      "step": 2238
    },
    {
      "epoch": 2.8640691255300426,
      "grad_norm": 0.5563523173332214,
      "learning_rate": 1.8678713255494455e-05,
      "loss": 0.6909,
      "step": 2239
    },
    {
      "epoch": 2.865349227938235,
      "grad_norm": 0.5184001922607422,
      "learning_rate": 1.8664057830498565e-05,
      "loss": 0.6464,
      "step": 2240
    },
    {
      "epoch": 2.8666293303464276,
      "grad_norm": 0.5413727164268494,
      "learning_rate": 1.8649404695218455e-05,
      "loss": 0.6858,
      "step": 2241
    },
    {
      "epoch": 2.8679094327546206,
      "grad_norm": 0.5493753552436829,
      "learning_rate": 1.8634753859711704e-05,
      "loss": 0.6573,
      "step": 2242
    },
    {
      "epoch": 2.869189535162813,
      "grad_norm": 0.5444915294647217,
      "learning_rate": 1.862010533403432e-05,
      "loss": 0.6871,
      "step": 2243
    },
    {
      "epoch": 2.8704696375710057,
      "grad_norm": 0.5372090935707092,
      "learning_rate": 1.860545912824071e-05,
      "loss": 0.6596,
      "step": 2244
    },
    {
      "epoch": 2.8717497399791982,
      "grad_norm": 0.5529900193214417,
      "learning_rate": 1.8590815252383722e-05,
      "loss": 0.6677,
      "step": 2245
    },
    {
      "epoch": 2.873029842387391,
      "grad_norm": 0.5337818264961243,
      "learning_rate": 1.857617371651456e-05,
      "loss": 0.6618,
      "step": 2246
    },
    {
      "epoch": 2.8743099447955838,
      "grad_norm": 0.5331010818481445,
      "learning_rate": 1.856153453068286e-05,
      "loss": 0.6936,
      "step": 2247
    },
    {
      "epoch": 2.8755900472037763,
      "grad_norm": 0.5968855619430542,
      "learning_rate": 1.8546897704936633e-05,
      "loss": 0.674,
      "step": 2248
    },
    {
      "epoch": 2.876870149611969,
      "grad_norm": 0.550433337688446,
      "learning_rate": 1.8532263249322255e-05,
      "loss": 0.6734,
      "step": 2249
    },
    {
      "epoch": 2.878150252020162,
      "grad_norm": 0.5292633175849915,
      "learning_rate": 1.8517631173884495e-05,
      "loss": 0.6707,
      "step": 2250
    },
    {
      "epoch": 2.878150252020162,
      "eval_loss": 0.011448027566075325,
      "eval_runtime": 10.8256,
      "eval_samples_per_second": 45.817,
      "eval_steps_per_second": 5.727,
      "step": 2250
    },
    {
      "epoch": 2.8794303544283544,
      "grad_norm": 0.523199200630188,
      "learning_rate": 1.850300148866648e-05,
      "loss": 0.641,
      "step": 2251
    },
    {
      "epoch": 2.880710456836547,
      "grad_norm": 0.5419129729270935,
      "learning_rate": 1.8488374203709692e-05,
      "loss": 0.6883,
      "step": 2252
    },
    {
      "epoch": 2.8819905592447395,
      "grad_norm": 0.5356900691986084,
      "learning_rate": 1.8473749329053977e-05,
      "loss": 0.6677,
      "step": 2253
    },
    {
      "epoch": 2.883270661652932,
      "grad_norm": 0.5589077472686768,
      "learning_rate": 1.845912687473753e-05,
      "loss": 0.7183,
      "step": 2254
    },
    {
      "epoch": 2.884550764061125,
      "grad_norm": 0.5814553499221802,
      "learning_rate": 1.8444506850796852e-05,
      "loss": 0.6842,
      "step": 2255
    },
    {
      "epoch": 2.8858308664693175,
      "grad_norm": 0.5884097814559937,
      "learning_rate": 1.8429889267266818e-05,
      "loss": 0.6861,
      "step": 2256
    },
    {
      "epoch": 2.88711096887751,
      "grad_norm": 0.5445072054862976,
      "learning_rate": 1.8415274134180605e-05,
      "loss": 0.69,
      "step": 2257
    },
    {
      "epoch": 2.888391071285703,
      "grad_norm": 0.584773063659668,
      "learning_rate": 1.8400661461569705e-05,
      "loss": 0.6821,
      "step": 2258
    },
    {
      "epoch": 2.8896711736938956,
      "grad_norm": 0.5701531767845154,
      "learning_rate": 1.8386051259463928e-05,
      "loss": 0.6888,
      "step": 2259
    },
    {
      "epoch": 2.890951276102088,
      "grad_norm": 0.5567769408226013,
      "learning_rate": 1.8371443537891406e-05,
      "loss": 0.679,
      "step": 2260
    },
    {
      "epoch": 2.8922313785102807,
      "grad_norm": 0.5713131427764893,
      "learning_rate": 1.835683830687853e-05,
      "loss": 0.6713,
      "step": 2261
    },
    {
      "epoch": 2.893511480918473,
      "grad_norm": 0.5630804300308228,
      "learning_rate": 1.834223557645001e-05,
      "loss": 0.7099,
      "step": 2262
    },
    {
      "epoch": 2.894791583326666,
      "grad_norm": 0.5425707697868347,
      "learning_rate": 1.832763535662884e-05,
      "loss": 0.6889,
      "step": 2263
    },
    {
      "epoch": 2.8960716857348587,
      "grad_norm": 0.5156663656234741,
      "learning_rate": 1.831303765743627e-05,
      "loss": 0.6713,
      "step": 2264
    },
    {
      "epoch": 2.8973517881430517,
      "grad_norm": 0.5536324977874756,
      "learning_rate": 1.829844248889184e-05,
      "loss": 0.6558,
      "step": 2265
    },
    {
      "epoch": 2.8986318905512443,
      "grad_norm": 0.5337991714477539,
      "learning_rate": 1.828384986101335e-05,
      "loss": 0.6679,
      "step": 2266
    },
    {
      "epoch": 2.899911992959437,
      "grad_norm": 0.542484700679779,
      "learning_rate": 1.8269259783816838e-05,
      "loss": 0.6912,
      "step": 2267
    },
    {
      "epoch": 2.9011920953676293,
      "grad_norm": 0.5581287741661072,
      "learning_rate": 1.8254672267316625e-05,
      "loss": 0.6627,
      "step": 2268
    },
    {
      "epoch": 2.902472197775822,
      "grad_norm": 0.5504825115203857,
      "learning_rate": 1.8240087321525236e-05,
      "loss": 0.7211,
      "step": 2269
    },
    {
      "epoch": 2.903752300184015,
      "grad_norm": 0.549676239490509,
      "learning_rate": 1.822550495645346e-05,
      "loss": 0.6678,
      "step": 2270
    },
    {
      "epoch": 2.9050324025922074,
      "grad_norm": 0.5072528719902039,
      "learning_rate": 1.8210925182110305e-05,
      "loss": 0.6432,
      "step": 2271
    },
    {
      "epoch": 2.9063125050004,
      "grad_norm": 0.5294970273971558,
      "learning_rate": 1.8196348008502993e-05,
      "loss": 0.6796,
      "step": 2272
    },
    {
      "epoch": 2.907592607408593,
      "grad_norm": 0.5018666386604309,
      "learning_rate": 1.8181773445636975e-05,
      "loss": 0.6761,
      "step": 2273
    },
    {
      "epoch": 2.9088727098167855,
      "grad_norm": 0.5346142649650574,
      "learning_rate": 1.816720150351591e-05,
      "loss": 0.6618,
      "step": 2274
    },
    {
      "epoch": 2.910152812224978,
      "grad_norm": 0.5576887726783752,
      "learning_rate": 1.815263219214164e-05,
      "loss": 0.6694,
      "step": 2275
    },
    {
      "epoch": 2.9114329146331706,
      "grad_norm": 0.519875705242157,
      "learning_rate": 1.813806552151421e-05,
      "loss": 0.6862,
      "step": 2276
    },
    {
      "epoch": 2.912713017041363,
      "grad_norm": 0.5766414403915405,
      "learning_rate": 1.8123501501631874e-05,
      "loss": 0.6929,
      "step": 2277
    },
    {
      "epoch": 2.913993119449556,
      "grad_norm": 0.5306098461151123,
      "learning_rate": 1.810894014249103e-05,
      "loss": 0.6441,
      "step": 2278
    },
    {
      "epoch": 2.9152732218577486,
      "grad_norm": 0.5158607959747314,
      "learning_rate": 1.8094381454086278e-05,
      "loss": 0.6806,
      "step": 2279
    },
    {
      "epoch": 2.916553324265941,
      "grad_norm": 0.5030341744422913,
      "learning_rate": 1.8079825446410376e-05,
      "loss": 0.6913,
      "step": 2280
    },
    {
      "epoch": 2.916553324265941,
      "eval_loss": 0.011444072239100933,
      "eval_runtime": 10.8172,
      "eval_samples_per_second": 45.853,
      "eval_steps_per_second": 5.732,
      "step": 2280
    },
    {
      "epoch": 2.917833426674134,
      "grad_norm": 0.5056853890419006,
      "learning_rate": 1.806527212945423e-05,
      "loss": 0.6785,
      "step": 2281
    },
    {
      "epoch": 2.9191135290823267,
      "grad_norm": 0.5447837710380554,
      "learning_rate": 1.8050721513206913e-05,
      "loss": 0.6811,
      "step": 2282
    },
    {
      "epoch": 2.9203936314905192,
      "grad_norm": 0.5257372260093689,
      "learning_rate": 1.803617360765565e-05,
      "loss": 0.6601,
      "step": 2283
    },
    {
      "epoch": 2.9216737338987118,
      "grad_norm": 0.5078234672546387,
      "learning_rate": 1.8021628422785796e-05,
      "loss": 0.6813,
      "step": 2284
    },
    {
      "epoch": 2.9229538363069043,
      "grad_norm": 0.5475780963897705,
      "learning_rate": 1.8007085968580825e-05,
      "loss": 0.668,
      "step": 2285
    },
    {
      "epoch": 2.9242339387150973,
      "grad_norm": 0.5548843741416931,
      "learning_rate": 1.7992546255022366e-05,
      "loss": 0.675,
      "step": 2286
    },
    {
      "epoch": 2.92551404112329,
      "grad_norm": 0.5062752366065979,
      "learning_rate": 1.7978009292090145e-05,
      "loss": 0.6595,
      "step": 2287
    },
    {
      "epoch": 2.9267941435314824,
      "grad_norm": 0.5550132393836975,
      "learning_rate": 1.7963475089762002e-05,
      "loss": 0.7002,
      "step": 2288
    },
    {
      "epoch": 2.9280742459396754,
      "grad_norm": 0.5342233180999756,
      "learning_rate": 1.79489436580139e-05,
      "loss": 0.6494,
      "step": 2289
    },
    {
      "epoch": 2.929354348347868,
      "grad_norm": 0.5512034296989441,
      "learning_rate": 1.7934415006819882e-05,
      "loss": 0.6887,
      "step": 2290
    },
    {
      "epoch": 2.9306344507560604,
      "grad_norm": 0.5523194074630737,
      "learning_rate": 1.791988914615208e-05,
      "loss": 0.6685,
      "step": 2291
    },
    {
      "epoch": 2.931914553164253,
      "grad_norm": 0.5492927432060242,
      "learning_rate": 1.790536608598073e-05,
      "loss": 0.6466,
      "step": 2292
    },
    {
      "epoch": 2.9331946555724455,
      "grad_norm": 0.5797692537307739,
      "learning_rate": 1.7890845836274126e-05,
      "loss": 0.6576,
      "step": 2293
    },
    {
      "epoch": 2.9344747579806385,
      "grad_norm": 0.5514007210731506,
      "learning_rate": 1.7876328406998644e-05,
      "loss": 0.6667,
      "step": 2294
    },
    {
      "epoch": 2.935754860388831,
      "grad_norm": 0.5545125603675842,
      "learning_rate": 1.786181380811872e-05,
      "loss": 0.6881,
      "step": 2295
    },
    {
      "epoch": 2.937034962797024,
      "grad_norm": 0.5694110989570618,
      "learning_rate": 1.7847302049596855e-05,
      "loss": 0.6972,
      "step": 2296
    },
    {
      "epoch": 2.9383150652052166,
      "grad_norm": 0.5477825999259949,
      "learning_rate": 1.7832793141393584e-05,
      "loss": 0.6806,
      "step": 2297
    },
    {
      "epoch": 2.939595167613409,
      "grad_norm": 0.5117954015731812,
      "learning_rate": 1.7818287093467504e-05,
      "loss": 0.6962,
      "step": 2298
    },
    {
      "epoch": 2.9408752700216017,
      "grad_norm": 0.5466132760047913,
      "learning_rate": 1.780378391577524e-05,
      "loss": 0.6643,
      "step": 2299
    },
    {
      "epoch": 2.942155372429794,
      "grad_norm": 0.5463626980781555,
      "learning_rate": 1.778928361827144e-05,
      "loss": 0.6869,
      "step": 2300
    },
    {
      "epoch": 2.943435474837987,
      "grad_norm": 0.5090941786766052,
      "learning_rate": 1.7774786210908784e-05,
      "loss": 0.6402,
      "step": 2301
    },
    {
      "epoch": 2.9447155772461797,
      "grad_norm": 0.5924723148345947,
      "learning_rate": 1.7760291703637973e-05,
      "loss": 0.6985,
      "step": 2302
    },
    {
      "epoch": 2.9459956796543723,
      "grad_norm": 0.5959092378616333,
      "learning_rate": 1.7745800106407717e-05,
      "loss": 0.6777,
      "step": 2303
    },
    {
      "epoch": 2.9472757820625652,
      "grad_norm": 0.5699868202209473,
      "learning_rate": 1.7731311429164707e-05,
      "loss": 0.7121,
      "step": 2304
    },
    {
      "epoch": 2.948555884470758,
      "grad_norm": 0.5955172777175903,
      "learning_rate": 1.771682568185365e-05,
      "loss": 0.6519,
      "step": 2305
    },
    {
      "epoch": 2.9498359868789503,
      "grad_norm": 0.6218169331550598,
      "learning_rate": 1.770234287441725e-05,
      "loss": 0.6957,
      "step": 2306
    },
    {
      "epoch": 2.951116089287143,
      "grad_norm": 0.559373676776886,
      "learning_rate": 1.768786301679616e-05,
      "loss": 0.6966,
      "step": 2307
    },
    {
      "epoch": 2.9523961916953354,
      "grad_norm": 0.5644216537475586,
      "learning_rate": 1.7673386118929044e-05,
      "loss": 0.7107,
      "step": 2308
    },
    {
      "epoch": 2.9536762941035284,
      "grad_norm": 0.5214970111846924,
      "learning_rate": 1.7658912190752522e-05,
      "loss": 0.6266,
      "step": 2309
    },
    {
      "epoch": 2.954956396511721,
      "grad_norm": 0.5662040114402771,
      "learning_rate": 1.7644441242201162e-05,
      "loss": 0.6738,
      "step": 2310
    },
    {
      "epoch": 2.954956396511721,
      "eval_loss": 0.011429225094616413,
      "eval_runtime": 10.8327,
      "eval_samples_per_second": 45.787,
      "eval_steps_per_second": 5.723,
      "step": 2310
    },
    {
      "epoch": 2.9562364989199135,
      "grad_norm": 0.5169128179550171,
      "learning_rate": 1.76299732832075e-05,
      "loss": 0.6668,
      "step": 2311
    },
    {
      "epoch": 2.9575166013281065,
      "grad_norm": 0.5528967380523682,
      "learning_rate": 1.761550832370203e-05,
      "loss": 0.6774,
      "step": 2312
    },
    {
      "epoch": 2.958796703736299,
      "grad_norm": 0.5568346977233887,
      "learning_rate": 1.760104637361316e-05,
      "loss": 0.6672,
      "step": 2313
    },
    {
      "epoch": 2.9600768061444915,
      "grad_norm": 0.5672390460968018,
      "learning_rate": 1.758658744286725e-05,
      "loss": 0.6915,
      "step": 2314
    },
    {
      "epoch": 2.961356908552684,
      "grad_norm": 0.5150502324104309,
      "learning_rate": 1.75721315413886e-05,
      "loss": 0.7044,
      "step": 2315
    },
    {
      "epoch": 2.9626370109608766,
      "grad_norm": 0.5412737131118774,
      "learning_rate": 1.75576786790994e-05,
      "loss": 0.7196,
      "step": 2316
    },
    {
      "epoch": 2.9639171133690696,
      "grad_norm": 0.5048683285713196,
      "learning_rate": 1.754322886591977e-05,
      "loss": 0.6404,
      "step": 2317
    },
    {
      "epoch": 2.965197215777262,
      "grad_norm": 0.48774898052215576,
      "learning_rate": 1.7528782111767754e-05,
      "loss": 0.6685,
      "step": 2318
    },
    {
      "epoch": 2.9664773181854547,
      "grad_norm": 0.5408198237419128,
      "learning_rate": 1.751433842655926e-05,
      "loss": 0.6782,
      "step": 2319
    },
    {
      "epoch": 2.9677574205936477,
      "grad_norm": 0.5126347541809082,
      "learning_rate": 1.7499897820208125e-05,
      "loss": 0.6698,
      "step": 2320
    },
    {
      "epoch": 2.96903752300184,
      "grad_norm": 0.5047155618667603,
      "learning_rate": 1.7485460302626042e-05,
      "loss": 0.6528,
      "step": 2321
    },
    {
      "epoch": 2.9703176254100327,
      "grad_norm": 0.5210356116294861,
      "learning_rate": 1.7471025883722604e-05,
      "loss": 0.6761,
      "step": 2322
    },
    {
      "epoch": 2.9715977278182253,
      "grad_norm": 0.5186281204223633,
      "learning_rate": 1.7456594573405284e-05,
      "loss": 0.7083,
      "step": 2323
    },
    {
      "epoch": 2.9728778302264183,
      "grad_norm": 0.5127356648445129,
      "learning_rate": 1.7442166381579392e-05,
      "loss": 0.6877,
      "step": 2324
    },
    {
      "epoch": 2.974157932634611,
      "grad_norm": 0.5343574285507202,
      "learning_rate": 1.7427741318148124e-05,
      "loss": 0.7014,
      "step": 2325
    },
    {
      "epoch": 2.9754380350428034,
      "grad_norm": 0.4927723705768585,
      "learning_rate": 1.7413319393012524e-05,
      "loss": 0.6777,
      "step": 2326
    },
    {
      "epoch": 2.9767181374509963,
      "grad_norm": 0.4936085343360901,
      "learning_rate": 1.7398900616071457e-05,
      "loss": 0.6631,
      "step": 2327
    },
    {
      "epoch": 2.977998239859189,
      "grad_norm": 0.4894747734069824,
      "learning_rate": 1.738448499722167e-05,
      "loss": 0.6378,
      "step": 2328
    },
    {
      "epoch": 2.9792783422673814,
      "grad_norm": 0.5036857724189758,
      "learning_rate": 1.7370072546357713e-05,
      "loss": 0.683,
      "step": 2329
    },
    {
      "epoch": 2.980558444675574,
      "grad_norm": 0.5254474878311157,
      "learning_rate": 1.7355663273371965e-05,
      "loss": 0.624,
      "step": 2330
    },
    {
      "epoch": 2.9818385470837665,
      "grad_norm": 0.502440333366394,
      "learning_rate": 1.7341257188154625e-05,
      "loss": 0.6759,
      "step": 2331
    },
    {
      "epoch": 2.9831186494919595,
      "grad_norm": 0.520214319229126,
      "learning_rate": 1.7326854300593723e-05,
      "loss": 0.6709,
      "step": 2332
    },
    {
      "epoch": 2.984398751900152,
      "grad_norm": 0.5521348714828491,
      "learning_rate": 1.7312454620575055e-05,
      "loss": 0.6755,
      "step": 2333
    },
    {
      "epoch": 2.9856788543083446,
      "grad_norm": 0.5944056510925293,
      "learning_rate": 1.729805815798225e-05,
      "loss": 0.7021,
      "step": 2334
    },
    {
      "epoch": 2.9869589567165375,
      "grad_norm": 0.596458375453949,
      "learning_rate": 1.728366492269672e-05,
      "loss": 0.6757,
      "step": 2335
    },
    {
      "epoch": 2.98823905912473,
      "grad_norm": 0.5317464470863342,
      "learning_rate": 1.7269274924597644e-05,
      "loss": 0.7086,
      "step": 2336
    },
    {
      "epoch": 2.9895191615329226,
      "grad_norm": 0.5277230739593506,
      "learning_rate": 1.725488817356201e-05,
      "loss": 0.6537,
      "step": 2337
    },
    {
      "epoch": 2.990799263941115,
      "grad_norm": 0.5766561031341553,
      "learning_rate": 1.7240504679464554e-05,
      "loss": 0.6919,
      "step": 2338
    },
    {
      "epoch": 2.9920793663493077,
      "grad_norm": 0.5419392585754395,
      "learning_rate": 1.722612445217778e-05,
      "loss": 0.6429,
      "step": 2339
    },
    {
      "epoch": 2.9933594687575007,
      "grad_norm": 0.5629022121429443,
      "learning_rate": 1.7211747501571957e-05,
      "loss": 0.6483,
      "step": 2340
    },
    {
      "epoch": 2.9933594687575007,
      "eval_loss": 0.011423509567975998,
      "eval_runtime": 10.8251,
      "eval_samples_per_second": 45.819,
      "eval_steps_per_second": 5.727,
      "step": 2340
    },
    {
      "epoch": 2.9946395711656932,
      "grad_norm": 0.5854880809783936,
      "learning_rate": 1.7197373837515104e-05,
      "loss": 0.6896,
      "step": 2341
    },
    {
      "epoch": 2.9959196735738858,
      "grad_norm": 0.4965278208255768,
      "learning_rate": 1.718300346987297e-05,
      "loss": 0.6661,
      "step": 2342
    },
    {
      "epoch": 2.9971997759820788,
      "grad_norm": 0.5521373152732849,
      "learning_rate": 1.7168636408509064e-05,
      "loss": 0.6534,
      "step": 2343
    },
    {
      "epoch": 2.9984798783902713,
      "grad_norm": 0.5576989054679871,
      "learning_rate": 1.715427266328461e-05,
      "loss": 0.6832,
      "step": 2344
    },
    {
      "epoch": 2.999759980798464,
      "grad_norm": 0.5305948853492737,
      "learning_rate": 1.7139912244058555e-05,
      "loss": 0.6656,
      "step": 2345
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.9938316345214844,
      "learning_rate": 1.7125555160687582e-05,
      "loss": 0.6321,
      "step": 2346
    },
    {
      "epoch": 3.0012801024081925,
      "grad_norm": 0.7422385215759277,
      "learning_rate": 1.7111201423026054e-05,
      "loss": 0.6476,
      "step": 2347
    },
    {
      "epoch": 3.0025602048163855,
      "grad_norm": 0.6276028752326965,
      "learning_rate": 1.7096851040926062e-05,
      "loss": 0.6424,
      "step": 2348
    },
    {
      "epoch": 3.003840307224578,
      "grad_norm": 0.5436055064201355,
      "learning_rate": 1.708250402423739e-05,
      "loss": 0.6255,
      "step": 2349
    },
    {
      "epoch": 3.0051204096327706,
      "grad_norm": 0.6255354881286621,
      "learning_rate": 1.70681603828075e-05,
      "loss": 0.6323,
      "step": 2350
    },
    {
      "epoch": 3.006400512040963,
      "grad_norm": 0.6582286357879639,
      "learning_rate": 1.7053820126481545e-05,
      "loss": 0.6366,
      "step": 2351
    },
    {
      "epoch": 3.007680614449156,
      "grad_norm": 0.5947375893592834,
      "learning_rate": 1.703948326510237e-05,
      "loss": 0.6242,
      "step": 2352
    },
    {
      "epoch": 3.0089607168573487,
      "grad_norm": 0.603613018989563,
      "learning_rate": 1.7025149808510456e-05,
      "loss": 0.674,
      "step": 2353
    },
    {
      "epoch": 3.010240819265541,
      "grad_norm": 0.5693184733390808,
      "learning_rate": 1.7010819766543972e-05,
      "loss": 0.6536,
      "step": 2354
    },
    {
      "epoch": 3.0115209216737338,
      "grad_norm": 0.5672442317008972,
      "learning_rate": 1.6996493149038747e-05,
      "loss": 0.6322,
      "step": 2355
    },
    {
      "epoch": 3.0128010240819267,
      "grad_norm": 0.637457549571991,
      "learning_rate": 1.698216996582823e-05,
      "loss": 0.6834,
      "step": 2356
    },
    {
      "epoch": 3.0140811264901193,
      "grad_norm": 0.5192814469337463,
      "learning_rate": 1.6967850226743544e-05,
      "loss": 0.6562,
      "step": 2357
    },
    {
      "epoch": 3.015361228898312,
      "grad_norm": 0.5664059519767761,
      "learning_rate": 1.6953533941613446e-05,
      "loss": 0.6536,
      "step": 2358
    },
    {
      "epoch": 3.0166413313065044,
      "grad_norm": 0.559282660484314,
      "learning_rate": 1.693922112026429e-05,
      "loss": 0.6477,
      "step": 2359
    },
    {
      "epoch": 3.0179214337146973,
      "grad_norm": 0.5464085340499878,
      "learning_rate": 1.692491177252009e-05,
      "loss": 0.5935,
      "step": 2360
    },
    {
      "epoch": 3.01920153612289,
      "grad_norm": 0.5866004824638367,
      "learning_rate": 1.691060590820246e-05,
      "loss": 0.6732,
      "step": 2361
    },
    {
      "epoch": 3.0204816385310824,
      "grad_norm": 0.5507637858390808,
      "learning_rate": 1.6896303537130617e-05,
      "loss": 0.6411,
      "step": 2362
    },
    {
      "epoch": 3.021761740939275,
      "grad_norm": 0.5781289339065552,
      "learning_rate": 1.6882004669121394e-05,
      "loss": 0.6463,
      "step": 2363
    },
    {
      "epoch": 3.023041843347468,
      "grad_norm": 0.5602301955223083,
      "learning_rate": 1.6867709313989217e-05,
      "loss": 0.6076,
      "step": 2364
    },
    {
      "epoch": 3.0243219457556605,
      "grad_norm": 0.5312731266021729,
      "learning_rate": 1.6853417481546084e-05,
      "loss": 0.6828,
      "step": 2365
    },
    {
      "epoch": 3.025602048163853,
      "grad_norm": 0.5267046093940735,
      "learning_rate": 1.6839129181601596e-05,
      "loss": 0.6334,
      "step": 2366
    },
    {
      "epoch": 3.0268821505720456,
      "grad_norm": 0.5690827369689941,
      "learning_rate": 1.682484442396292e-05,
      "loss": 0.6406,
      "step": 2367
    },
    {
      "epoch": 3.0281622529802386,
      "grad_norm": 0.5078542828559875,
      "learning_rate": 1.681056321843479e-05,
      "loss": 0.6609,
      "step": 2368
    },
    {
      "epoch": 3.029442355388431,
      "grad_norm": 0.5197672843933105,
      "learning_rate": 1.6796285574819505e-05,
      "loss": 0.6719,
      "step": 2369
    },
    {
      "epoch": 3.0307224577966236,
      "grad_norm": 0.5278779864311218,
      "learning_rate": 1.6782011502916925e-05,
      "loss": 0.6652,
      "step": 2370
    },
    {
      "epoch": 3.0307224577966236,
      "eval_loss": 0.011501915752887726,
      "eval_runtime": 10.8261,
      "eval_samples_per_second": 45.815,
      "eval_steps_per_second": 5.727,
      "step": 2370
    },
    {
      "epoch": 3.032002560204816,
      "grad_norm": 0.53105628490448,
      "learning_rate": 1.6767741012524444e-05,
      "loss": 0.6375,
      "step": 2371
    },
    {
      "epoch": 3.033282662613009,
      "grad_norm": 0.5455028414726257,
      "learning_rate": 1.6753474113437014e-05,
      "loss": 0.648,
      "step": 2372
    },
    {
      "epoch": 3.0345627650212017,
      "grad_norm": 0.5050350427627563,
      "learning_rate": 1.6739210815447107e-05,
      "loss": 0.656,
      "step": 2373
    },
    {
      "epoch": 3.0358428674293942,
      "grad_norm": 0.5428577065467834,
      "learning_rate": 1.672495112834474e-05,
      "loss": 0.6447,
      "step": 2374
    },
    {
      "epoch": 3.0371229698375872,
      "grad_norm": 0.5513570308685303,
      "learning_rate": 1.671069506191743e-05,
      "loss": 0.6423,
      "step": 2375
    },
    {
      "epoch": 3.0384030722457798,
      "grad_norm": 0.5326774716377258,
      "learning_rate": 1.6696442625950223e-05,
      "loss": 0.6622,
      "step": 2376
    },
    {
      "epoch": 3.0396831746539723,
      "grad_norm": 0.516811192035675,
      "learning_rate": 1.6682193830225678e-05,
      "loss": 0.632,
      "step": 2377
    },
    {
      "epoch": 3.040963277062165,
      "grad_norm": 0.5091072916984558,
      "learning_rate": 1.6667948684523847e-05,
      "loss": 0.6082,
      "step": 2378
    },
    {
      "epoch": 3.042243379470358,
      "grad_norm": 0.49962544441223145,
      "learning_rate": 1.6653707198622272e-05,
      "loss": 0.6579,
      "step": 2379
    },
    {
      "epoch": 3.0435234818785504,
      "grad_norm": 0.4801453948020935,
      "learning_rate": 1.6639469382295992e-05,
      "loss": 0.6097,
      "step": 2380
    },
    {
      "epoch": 3.044803584286743,
      "grad_norm": 0.5038813352584839,
      "learning_rate": 1.662523524531753e-05,
      "loss": 0.6301,
      "step": 2381
    },
    {
      "epoch": 3.0460836866949355,
      "grad_norm": 0.5112163424491882,
      "learning_rate": 1.6611004797456866e-05,
      "loss": 0.6524,
      "step": 2382
    },
    {
      "epoch": 3.0473637891031284,
      "grad_norm": 0.537038266658783,
      "learning_rate": 1.659677804848147e-05,
      "loss": 0.6646,
      "step": 2383
    },
    {
      "epoch": 3.048643891511321,
      "grad_norm": 0.5388227701187134,
      "learning_rate": 1.658255500815626e-05,
      "loss": 0.659,
      "step": 2384
    },
    {
      "epoch": 3.0499239939195135,
      "grad_norm": 0.49667227268218994,
      "learning_rate": 1.6568335686243608e-05,
      "loss": 0.6349,
      "step": 2385
    },
    {
      "epoch": 3.051204096327706,
      "grad_norm": 0.5519853234291077,
      "learning_rate": 1.6554120092503346e-05,
      "loss": 0.6339,
      "step": 2386
    },
    {
      "epoch": 3.052484198735899,
      "grad_norm": 0.5274022221565247,
      "learning_rate": 1.653990823669273e-05,
      "loss": 0.6435,
      "step": 2387
    },
    {
      "epoch": 3.0537643011440916,
      "grad_norm": 0.5133228302001953,
      "learning_rate": 1.652570012856646e-05,
      "loss": 0.6492,
      "step": 2388
    },
    {
      "epoch": 3.055044403552284,
      "grad_norm": 0.5335580110549927,
      "learning_rate": 1.651149577787666e-05,
      "loss": 0.6459,
      "step": 2389
    },
    {
      "epoch": 3.0563245059604767,
      "grad_norm": 0.509559154510498,
      "learning_rate": 1.649729519437289e-05,
      "loss": 0.6581,
      "step": 2390
    },
    {
      "epoch": 3.0576046083686697,
      "grad_norm": 0.5015418529510498,
      "learning_rate": 1.6483098387802098e-05,
      "loss": 0.678,
      "step": 2391
    },
    {
      "epoch": 3.058884710776862,
      "grad_norm": 0.5484011173248291,
      "learning_rate": 1.6468905367908653e-05,
      "loss": 0.6452,
      "step": 2392
    },
    {
      "epoch": 3.0601648131850547,
      "grad_norm": 0.5278120636940002,
      "learning_rate": 1.6454716144434334e-05,
      "loss": 0.6251,
      "step": 2393
    },
    {
      "epoch": 3.0614449155932473,
      "grad_norm": 0.5070874691009521,
      "learning_rate": 1.6440530727118295e-05,
      "loss": 0.6237,
      "step": 2394
    },
    {
      "epoch": 3.0627250180014403,
      "grad_norm": 0.5475306510925293,
      "learning_rate": 1.6426349125697097e-05,
      "loss": 0.6619,
      "step": 2395
    },
    {
      "epoch": 3.064005120409633,
      "grad_norm": 0.5989736318588257,
      "learning_rate": 1.6412171349904675e-05,
      "loss": 0.6766,
      "step": 2396
    },
    {
      "epoch": 3.0652852228178253,
      "grad_norm": 0.5467805862426758,
      "learning_rate": 1.6397997409472325e-05,
      "loss": 0.6418,
      "step": 2397
    },
    {
      "epoch": 3.066565325226018,
      "grad_norm": 0.6003584265708923,
      "learning_rate": 1.6383827314128732e-05,
      "loss": 0.6556,
      "step": 2398
    },
    {
      "epoch": 3.067845427634211,
      "grad_norm": 0.5687540769577026,
      "learning_rate": 1.6369661073599913e-05,
      "loss": 0.677,
      "step": 2399
    },
    {
      "epoch": 3.0691255300424034,
      "grad_norm": 0.5815536379814148,
      "learning_rate": 1.6355498697609275e-05,
      "loss": 0.6457,
      "step": 2400
    },
    {
      "epoch": 3.0691255300424034,
      "eval_loss": 0.011508642695844173,
      "eval_runtime": 10.8338,
      "eval_samples_per_second": 45.783,
      "eval_steps_per_second": 5.723,
      "step": 2400
    },
    {
      "epoch": 3.070405632450596,
      "grad_norm": 0.5543200373649597,
      "learning_rate": 1.6341340195877557e-05,
      "loss": 0.6363,
      "step": 2401
    },
    {
      "epoch": 3.0716857348587885,
      "grad_norm": 0.5092668533325195,
      "learning_rate": 1.6327185578122823e-05,
      "loss": 0.6362,
      "step": 2402
    },
    {
      "epoch": 3.0729658372669815,
      "grad_norm": 0.5534361004829407,
      "learning_rate": 1.631303485406049e-05,
      "loss": 0.705,
      "step": 2403
    },
    {
      "epoch": 3.074245939675174,
      "grad_norm": 0.5444126129150391,
      "learning_rate": 1.62988880334033e-05,
      "loss": 0.6531,
      "step": 2404
    },
    {
      "epoch": 3.0755260420833666,
      "grad_norm": 0.5639611482620239,
      "learning_rate": 1.6284745125861313e-05,
      "loss": 0.686,
      "step": 2405
    },
    {
      "epoch": 3.0768061444915595,
      "grad_norm": 0.5092193484306335,
      "learning_rate": 1.6270606141141896e-05,
      "loss": 0.6227,
      "step": 2406
    },
    {
      "epoch": 3.078086246899752,
      "grad_norm": 0.5228305459022522,
      "learning_rate": 1.6256471088949742e-05,
      "loss": 0.6722,
      "step": 2407
    },
    {
      "epoch": 3.0793663493079446,
      "grad_norm": 0.4967285096645355,
      "learning_rate": 1.624233997898682e-05,
      "loss": 0.6753,
      "step": 2408
    },
    {
      "epoch": 3.080646451716137,
      "grad_norm": 0.5037521719932556,
      "learning_rate": 1.6228212820952414e-05,
      "loss": 0.6445,
      "step": 2409
    },
    {
      "epoch": 3.08192655412433,
      "grad_norm": 0.5268568396568298,
      "learning_rate": 1.62140896245431e-05,
      "loss": 0.6346,
      "step": 2410
    },
    {
      "epoch": 3.0832066565325227,
      "grad_norm": 0.49345672130584717,
      "learning_rate": 1.6199970399452697e-05,
      "loss": 0.6156,
      "step": 2411
    },
    {
      "epoch": 3.0844867589407152,
      "grad_norm": 0.5403556227684021,
      "learning_rate": 1.6185855155372338e-05,
      "loss": 0.6568,
      "step": 2412
    },
    {
      "epoch": 3.0857668613489078,
      "grad_norm": 0.5360240340232849,
      "learning_rate": 1.6171743901990414e-05,
      "loss": 0.654,
      "step": 2413
    },
    {
      "epoch": 3.0870469637571007,
      "grad_norm": 0.5525732040405273,
      "learning_rate": 1.6157636648992566e-05,
      "loss": 0.6403,
      "step": 2414
    },
    {
      "epoch": 3.0883270661652933,
      "grad_norm": 0.5173043012619019,
      "learning_rate": 1.6143533406061692e-05,
      "loss": 0.6336,
      "step": 2415
    },
    {
      "epoch": 3.089607168573486,
      "grad_norm": 0.5381692051887512,
      "learning_rate": 1.6129434182877952e-05,
      "loss": 0.6541,
      "step": 2416
    },
    {
      "epoch": 3.0908872709816784,
      "grad_norm": 0.5557066202163696,
      "learning_rate": 1.611533898911872e-05,
      "loss": 0.6644,
      "step": 2417
    },
    {
      "epoch": 3.0921673733898714,
      "grad_norm": 0.5595386624336243,
      "learning_rate": 1.610124783445864e-05,
      "loss": 0.6232,
      "step": 2418
    },
    {
      "epoch": 3.093447475798064,
      "grad_norm": 0.5165018439292908,
      "learning_rate": 1.6087160728569557e-05,
      "loss": 0.6227,
      "step": 2419
    },
    {
      "epoch": 3.0947275782062564,
      "grad_norm": 0.5257919430732727,
      "learning_rate": 1.6073077681120532e-05,
      "loss": 0.6236,
      "step": 2420
    },
    {
      "epoch": 3.096007680614449,
      "grad_norm": 0.5543671250343323,
      "learning_rate": 1.6058998701777864e-05,
      "loss": 0.644,
      "step": 2421
    },
    {
      "epoch": 3.097287783022642,
      "grad_norm": 0.5471598505973816,
      "learning_rate": 1.6044923800205055e-05,
      "loss": 0.6632,
      "step": 2422
    },
    {
      "epoch": 3.0985678854308345,
      "grad_norm": 0.47705143690109253,
      "learning_rate": 1.6030852986062782e-05,
      "loss": 0.6168,
      "step": 2423
    },
    {
      "epoch": 3.099847987839027,
      "grad_norm": 0.5634862780570984,
      "learning_rate": 1.6016786269008954e-05,
      "loss": 0.6512,
      "step": 2424
    },
    {
      "epoch": 3.1011280902472196,
      "grad_norm": 0.5108270645141602,
      "learning_rate": 1.600272365869864e-05,
      "loss": 0.6425,
      "step": 2425
    },
    {
      "epoch": 3.1024081926554126,
      "grad_norm": 0.51654052734375,
      "learning_rate": 1.5988665164784093e-05,
      "loss": 0.6681,
      "step": 2426
    },
    {
      "epoch": 3.103688295063605,
      "grad_norm": 0.5479849576950073,
      "learning_rate": 1.5974610796914765e-05,
      "loss": 0.6162,
      "step": 2427
    },
    {
      "epoch": 3.1049683974717976,
      "grad_norm": 0.5156897306442261,
      "learning_rate": 1.5960560564737247e-05,
      "loss": 0.6803,
      "step": 2428
    },
    {
      "epoch": 3.10624849987999,
      "grad_norm": 0.5287712216377258,
      "learning_rate": 1.5946514477895303e-05,
      "loss": 0.6313,
      "step": 2429
    },
    {
      "epoch": 3.107528602288183,
      "grad_norm": 0.5856785178184509,
      "learning_rate": 1.5932472546029863e-05,
      "loss": 0.66,
      "step": 2430
    },
    {
      "epoch": 3.107528602288183,
      "eval_loss": 0.01151246763765812,
      "eval_runtime": 10.8137,
      "eval_samples_per_second": 45.868,
      "eval_steps_per_second": 5.733,
      "step": 2430
    },
    {
      "epoch": 3.1088087046963757,
      "grad_norm": 0.5303325653076172,
      "learning_rate": 1.5918434778778974e-05,
      "loss": 0.6467,
      "step": 2431
    },
    {
      "epoch": 3.1100888071045683,
      "grad_norm": 0.6068860292434692,
      "learning_rate": 1.5904401185777855e-05,
      "loss": 0.6841,
      "step": 2432
    },
    {
      "epoch": 3.111368909512761,
      "grad_norm": 0.5345388650894165,
      "learning_rate": 1.5890371776658858e-05,
      "loss": 0.6411,
      "step": 2433
    },
    {
      "epoch": 3.112649011920954,
      "grad_norm": 0.5276821851730347,
      "learning_rate": 1.5876346561051438e-05,
      "loss": 0.6408,
      "step": 2434
    },
    {
      "epoch": 3.1139291143291463,
      "grad_norm": 0.5618433356285095,
      "learning_rate": 1.5862325548582193e-05,
      "loss": 0.6744,
      "step": 2435
    },
    {
      "epoch": 3.115209216737339,
      "grad_norm": 0.5100088119506836,
      "learning_rate": 1.584830874887484e-05,
      "loss": 0.6772,
      "step": 2436
    },
    {
      "epoch": 3.116489319145532,
      "grad_norm": 0.5173203945159912,
      "learning_rate": 1.5834296171550187e-05,
      "loss": 0.6595,
      "step": 2437
    },
    {
      "epoch": 3.1177694215537244,
      "grad_norm": 0.5615100860595703,
      "learning_rate": 1.5820287826226145e-05,
      "loss": 0.6464,
      "step": 2438
    },
    {
      "epoch": 3.119049523961917,
      "grad_norm": 0.5282062292098999,
      "learning_rate": 1.5806283722517747e-05,
      "loss": 0.6454,
      "step": 2439
    },
    {
      "epoch": 3.1203296263701095,
      "grad_norm": 0.4987587332725525,
      "learning_rate": 1.579228387003708e-05,
      "loss": 0.6473,
      "step": 2440
    },
    {
      "epoch": 3.1216097287783025,
      "grad_norm": 0.5156669020652771,
      "learning_rate": 1.577828827839333e-05,
      "loss": 0.6576,
      "step": 2441
    },
    {
      "epoch": 3.122889831186495,
      "grad_norm": 0.5347732901573181,
      "learning_rate": 1.5764296957192763e-05,
      "loss": 0.6158,
      "step": 2442
    },
    {
      "epoch": 3.1241699335946875,
      "grad_norm": 0.5573805570602417,
      "learning_rate": 1.5750309916038707e-05,
      "loss": 0.6272,
      "step": 2443
    },
    {
      "epoch": 3.12545003600288,
      "grad_norm": 0.49766889214515686,
      "learning_rate": 1.5736327164531543e-05,
      "loss": 0.6148,
      "step": 2444
    },
    {
      "epoch": 3.126730138411073,
      "grad_norm": 0.5209107398986816,
      "learning_rate": 1.572234871226873e-05,
      "loss": 0.6452,
      "step": 2445
    },
    {
      "epoch": 3.1280102408192656,
      "grad_norm": 0.5295728445053101,
      "learning_rate": 1.5708374568844758e-05,
      "loss": 0.6283,
      "step": 2446
    },
    {
      "epoch": 3.129290343227458,
      "grad_norm": 0.5458559393882751,
      "learning_rate": 1.5694404743851162e-05,
      "loss": 0.673,
      "step": 2447
    },
    {
      "epoch": 3.1305704456356507,
      "grad_norm": 0.5030727982521057,
      "learning_rate": 1.5680439246876528e-05,
      "loss": 0.6546,
      "step": 2448
    },
    {
      "epoch": 3.1318505480438437,
      "grad_norm": 0.5325101017951965,
      "learning_rate": 1.5666478087506445e-05,
      "loss": 0.6171,
      "step": 2449
    },
    {
      "epoch": 3.133130650452036,
      "grad_norm": 0.5380441546440125,
      "learning_rate": 1.5652521275323547e-05,
      "loss": 0.6626,
      "step": 2450
    },
    {
      "epoch": 3.1344107528602287,
      "grad_norm": 0.5023500323295593,
      "learning_rate": 1.5638568819907475e-05,
      "loss": 0.6462,
      "step": 2451
    },
    {
      "epoch": 3.1356908552684213,
      "grad_norm": 0.5191518664360046,
      "learning_rate": 1.5624620730834878e-05,
      "loss": 0.6669,
      "step": 2452
    },
    {
      "epoch": 3.1369709576766143,
      "grad_norm": 0.5531533360481262,
      "learning_rate": 1.5610677017679417e-05,
      "loss": 0.6702,
      "step": 2453
    },
    {
      "epoch": 3.138251060084807,
      "grad_norm": 0.5363332033157349,
      "learning_rate": 1.5596737690011735e-05,
      "loss": 0.6578,
      "step": 2454
    },
    {
      "epoch": 3.1395311624929993,
      "grad_norm": 0.530849814414978,
      "learning_rate": 1.5582802757399485e-05,
      "loss": 0.6559,
      "step": 2455
    },
    {
      "epoch": 3.140811264901192,
      "grad_norm": 0.507783055305481,
      "learning_rate": 1.5568872229407282e-05,
      "loss": 0.652,
      "step": 2456
    },
    {
      "epoch": 3.142091367309385,
      "grad_norm": 0.5618488788604736,
      "learning_rate": 1.5554946115596733e-05,
      "loss": 0.6378,
      "step": 2457
    },
    {
      "epoch": 3.1433714697175774,
      "grad_norm": 0.531067967414856,
      "learning_rate": 1.5541024425526406e-05,
      "loss": 0.6661,
      "step": 2458
    },
    {
      "epoch": 3.14465157212577,
      "grad_norm": 0.547110915184021,
      "learning_rate": 1.5527107168751845e-05,
      "loss": 0.6627,
      "step": 2459
    },
    {
      "epoch": 3.145931674533963,
      "grad_norm": 0.5522833466529846,
      "learning_rate": 1.5513194354825523e-05,
      "loss": 0.6631,
      "step": 2460
    },
    {
      "epoch": 3.145931674533963,
      "eval_loss": 0.011496835388243198,
      "eval_runtime": 10.8308,
      "eval_samples_per_second": 45.795,
      "eval_steps_per_second": 5.724,
      "step": 2460
    },
    {
      "epoch": 3.1472117769421555,
      "grad_norm": 0.5810582041740417,
      "learning_rate": 1.54992859932969e-05,
      "loss": 0.6893,
      "step": 2461
    },
    {
      "epoch": 3.148491879350348,
      "grad_norm": 0.5589973330497742,
      "learning_rate": 1.5485382093712368e-05,
      "loss": 0.6431,
      "step": 2462
    },
    {
      "epoch": 3.1497719817585406,
      "grad_norm": 0.5137026309967041,
      "learning_rate": 1.547148266561523e-05,
      "loss": 0.6451,
      "step": 2463
    },
    {
      "epoch": 3.1510520841667335,
      "grad_norm": 0.5745928287506104,
      "learning_rate": 1.545758771854575e-05,
      "loss": 0.6455,
      "step": 2464
    },
    {
      "epoch": 3.152332186574926,
      "grad_norm": 0.5704997777938843,
      "learning_rate": 1.5443697262041126e-05,
      "loss": 0.6519,
      "step": 2465
    },
    {
      "epoch": 3.1536122889831186,
      "grad_norm": 0.5672619342803955,
      "learning_rate": 1.5429811305635433e-05,
      "loss": 0.6707,
      "step": 2466
    },
    {
      "epoch": 3.154892391391311,
      "grad_norm": 0.544376790523529,
      "learning_rate": 1.541592985885968e-05,
      "loss": 0.6582,
      "step": 2467
    },
    {
      "epoch": 3.156172493799504,
      "grad_norm": 0.5694115161895752,
      "learning_rate": 1.54020529312418e-05,
      "loss": 0.6253,
      "step": 2468
    },
    {
      "epoch": 3.1574525962076967,
      "grad_norm": 0.5433099865913391,
      "learning_rate": 1.538818053230658e-05,
      "loss": 0.6587,
      "step": 2469
    },
    {
      "epoch": 3.1587326986158892,
      "grad_norm": 0.5109469890594482,
      "learning_rate": 1.5374312671575736e-05,
      "loss": 0.6336,
      "step": 2470
    },
    {
      "epoch": 3.1600128010240818,
      "grad_norm": 0.5574575066566467,
      "learning_rate": 1.5360449358567864e-05,
      "loss": 0.6303,
      "step": 2471
    },
    {
      "epoch": 3.1612929034322748,
      "grad_norm": 0.5440110564231873,
      "learning_rate": 1.5346590602798413e-05,
      "loss": 0.652,
      "step": 2472
    },
    {
      "epoch": 3.1625730058404673,
      "grad_norm": 0.5183259844779968,
      "learning_rate": 1.5332736413779726e-05,
      "loss": 0.659,
      "step": 2473
    },
    {
      "epoch": 3.16385310824866,
      "grad_norm": 0.5226305723190308,
      "learning_rate": 1.5318886801021018e-05,
      "loss": 0.6368,
      "step": 2474
    },
    {
      "epoch": 3.1651332106568524,
      "grad_norm": 0.5394190549850464,
      "learning_rate": 1.5305041774028345e-05,
      "loss": 0.6709,
      "step": 2475
    },
    {
      "epoch": 3.1664133130650454,
      "grad_norm": 0.5406842231750488,
      "learning_rate": 1.529120134230462e-05,
      "loss": 0.6476,
      "step": 2476
    },
    {
      "epoch": 3.167693415473238,
      "grad_norm": 0.5195798873901367,
      "learning_rate": 1.527736551534961e-05,
      "loss": 0.6945,
      "step": 2477
    },
    {
      "epoch": 3.1689735178814304,
      "grad_norm": 0.5152619481086731,
      "learning_rate": 1.526353430265991e-05,
      "loss": 0.6142,
      "step": 2478
    },
    {
      "epoch": 3.170253620289623,
      "grad_norm": 0.5421910285949707,
      "learning_rate": 1.524970771372896e-05,
      "loss": 0.6577,
      "step": 2479
    },
    {
      "epoch": 3.171533722697816,
      "grad_norm": 0.534690797328949,
      "learning_rate": 1.5235885758047011e-05,
      "loss": 0.6539,
      "step": 2480
    },
    {
      "epoch": 3.1728138251060085,
      "grad_norm": 0.4806743562221527,
      "learning_rate": 1.522206844510115e-05,
      "loss": 0.6135,
      "step": 2481
    },
    {
      "epoch": 3.174093927514201,
      "grad_norm": 0.5362149477005005,
      "learning_rate": 1.5208255784375276e-05,
      "loss": 0.6881,
      "step": 2482
    },
    {
      "epoch": 3.1753740299223936,
      "grad_norm": 0.5271737575531006,
      "learning_rate": 1.5194447785350073e-05,
      "loss": 0.675,
      "step": 2483
    },
    {
      "epoch": 3.1766541323305866,
      "grad_norm": 0.5258865356445312,
      "learning_rate": 1.5180644457503056e-05,
      "loss": 0.6225,
      "step": 2484
    },
    {
      "epoch": 3.177934234738779,
      "grad_norm": 0.5148061513900757,
      "learning_rate": 1.5166845810308518e-05,
      "loss": 0.6428,
      "step": 2485
    },
    {
      "epoch": 3.1792143371469717,
      "grad_norm": 0.5288801193237305,
      "learning_rate": 1.5153051853237526e-05,
      "loss": 0.6603,
      "step": 2486
    },
    {
      "epoch": 3.180494439555164,
      "grad_norm": 0.5338370203971863,
      "learning_rate": 1.5139262595757954e-05,
      "loss": 0.6308,
      "step": 2487
    },
    {
      "epoch": 3.181774541963357,
      "grad_norm": 0.5008540153503418,
      "learning_rate": 1.5125478047334445e-05,
      "loss": 0.6177,
      "step": 2488
    },
    {
      "epoch": 3.1830546443715497,
      "grad_norm": 0.5101842880249023,
      "learning_rate": 1.5111698217428385e-05,
      "loss": 0.6187,
      "step": 2489
    },
    {
      "epoch": 3.1843347467797423,
      "grad_norm": 0.5729373097419739,
      "learning_rate": 1.5097923115497955e-05,
      "loss": 0.6646,
      "step": 2490
    },
    {
      "epoch": 3.1843347467797423,
      "eval_loss": 0.011502954177558422,
      "eval_runtime": 10.8173,
      "eval_samples_per_second": 45.852,
      "eval_steps_per_second": 5.732,
      "step": 2490
    },
    {
      "epoch": 3.1856148491879352,
      "grad_norm": 0.5281634330749512,
      "learning_rate": 1.5084152750998079e-05,
      "loss": 0.6225,
      "step": 2491
    },
    {
      "epoch": 3.186894951596128,
      "grad_norm": 0.5156639218330383,
      "learning_rate": 1.5070387133380408e-05,
      "loss": 0.658,
      "step": 2492
    },
    {
      "epoch": 3.1881750540043203,
      "grad_norm": 0.5498717427253723,
      "learning_rate": 1.5056626272093358e-05,
      "loss": 0.6519,
      "step": 2493
    },
    {
      "epoch": 3.189455156412513,
      "grad_norm": 0.5112549066543579,
      "learning_rate": 1.5042870176582094e-05,
      "loss": 0.6372,
      "step": 2494
    },
    {
      "epoch": 3.190735258820706,
      "grad_norm": 0.4875716269016266,
      "learning_rate": 1.5029118856288465e-05,
      "loss": 0.6735,
      "step": 2495
    },
    {
      "epoch": 3.1920153612288984,
      "grad_norm": 0.5193595290184021,
      "learning_rate": 1.501537232065108e-05,
      "loss": 0.6487,
      "step": 2496
    },
    {
      "epoch": 3.193295463637091,
      "grad_norm": 0.5427823066711426,
      "learning_rate": 1.500163057910526e-05,
      "loss": 0.6656,
      "step": 2497
    },
    {
      "epoch": 3.1945755660452835,
      "grad_norm": 0.5591933131217957,
      "learning_rate": 1.4987893641083014e-05,
      "loss": 0.6421,
      "step": 2498
    },
    {
      "epoch": 3.1958556684534765,
      "grad_norm": 0.517242431640625,
      "learning_rate": 1.4974161516013071e-05,
      "loss": 0.6358,
      "step": 2499
    },
    {
      "epoch": 3.197135770861669,
      "grad_norm": 0.535478413105011,
      "learning_rate": 1.496043421332087e-05,
      "loss": 0.631,
      "step": 2500
    },
    {
      "epoch": 3.1984158732698615,
      "grad_norm": 0.5194719433784485,
      "learning_rate": 1.4946711742428495e-05,
      "loss": 0.6253,
      "step": 2501
    },
    {
      "epoch": 3.199695975678054,
      "grad_norm": 0.5094932317733765,
      "learning_rate": 1.4932994112754755e-05,
      "loss": 0.6394,
      "step": 2502
    },
    {
      "epoch": 3.200976078086247,
      "grad_norm": 0.504401683807373,
      "learning_rate": 1.4919281333715136e-05,
      "loss": 0.629,
      "step": 2503
    },
    {
      "epoch": 3.2022561804944396,
      "grad_norm": 0.5324363708496094,
      "learning_rate": 1.4905573414721759e-05,
      "loss": 0.6078,
      "step": 2504
    },
    {
      "epoch": 3.203536282902632,
      "grad_norm": 0.5347485542297363,
      "learning_rate": 1.4891870365183447e-05,
      "loss": 0.6302,
      "step": 2505
    },
    {
      "epoch": 3.2048163853108247,
      "grad_norm": 0.5270418524742126,
      "learning_rate": 1.4878172194505656e-05,
      "loss": 0.6754,
      "step": 2506
    },
    {
      "epoch": 3.2060964877190177,
      "grad_norm": 0.5032305121421814,
      "learning_rate": 1.4864478912090508e-05,
      "loss": 0.649,
      "step": 2507
    },
    {
      "epoch": 3.20737659012721,
      "grad_norm": 0.5125587582588196,
      "learning_rate": 1.4850790527336773e-05,
      "loss": 0.6588,
      "step": 2508
    },
    {
      "epoch": 3.2086566925354028,
      "grad_norm": 0.5229718685150146,
      "learning_rate": 1.4837107049639832e-05,
      "loss": 0.6512,
      "step": 2509
    },
    {
      "epoch": 3.2099367949435953,
      "grad_norm": 0.5067161321640015,
      "learning_rate": 1.4823428488391724e-05,
      "loss": 0.644,
      "step": 2510
    },
    {
      "epoch": 3.2112168973517883,
      "grad_norm": 0.5176657438278198,
      "learning_rate": 1.4809754852981115e-05,
      "loss": 0.6723,
      "step": 2511
    },
    {
      "epoch": 3.212496999759981,
      "grad_norm": 0.5523592829704285,
      "learning_rate": 1.4796086152793274e-05,
      "loss": 0.6277,
      "step": 2512
    },
    {
      "epoch": 3.2137771021681734,
      "grad_norm": 0.5369203686714172,
      "learning_rate": 1.4782422397210083e-05,
      "loss": 0.6766,
      "step": 2513
    },
    {
      "epoch": 3.215057204576366,
      "grad_norm": 0.5011036992073059,
      "learning_rate": 1.4768763595610059e-05,
      "loss": 0.6428,
      "step": 2514
    },
    {
      "epoch": 3.216337306984559,
      "grad_norm": 0.49802538752555847,
      "learning_rate": 1.4755109757368264e-05,
      "loss": 0.6285,
      "step": 2515
    },
    {
      "epoch": 3.2176174093927514,
      "grad_norm": 0.5186784267425537,
      "learning_rate": 1.4741460891856415e-05,
      "loss": 0.6369,
      "step": 2516
    },
    {
      "epoch": 3.218897511800944,
      "grad_norm": 0.5105106234550476,
      "learning_rate": 1.4727817008442777e-05,
      "loss": 0.6184,
      "step": 2517
    },
    {
      "epoch": 3.2201776142091365,
      "grad_norm": 0.5055716037750244,
      "learning_rate": 1.4714178116492198e-05,
      "loss": 0.6534,
      "step": 2518
    },
    {
      "epoch": 3.2214577166173295,
      "grad_norm": 0.4734814465045929,
      "learning_rate": 1.4700544225366114e-05,
      "loss": 0.6255,
      "step": 2519
    },
    {
      "epoch": 3.222737819025522,
      "grad_norm": 0.502964973449707,
      "learning_rate": 1.4686915344422522e-05,
      "loss": 0.6442,
      "step": 2520
    },
    {
      "epoch": 3.222737819025522,
      "eval_loss": 0.01147892139852047,
      "eval_runtime": 10.8149,
      "eval_samples_per_second": 45.863,
      "eval_steps_per_second": 5.733,
      "step": 2520
    },
    {
      "epoch": 3.2240179214337146,
      "grad_norm": 0.5003054738044739,
      "learning_rate": 1.4673291483015977e-05,
      "loss": 0.6418,
      "step": 2521
    },
    {
      "epoch": 3.2252980238419076,
      "grad_norm": 0.5213746428489685,
      "learning_rate": 1.4659672650497591e-05,
      "loss": 0.6658,
      "step": 2522
    },
    {
      "epoch": 3.2265781262501,
      "grad_norm": 0.4791144132614136,
      "learning_rate": 1.464605885621503e-05,
      "loss": 0.6302,
      "step": 2523
    },
    {
      "epoch": 3.2278582286582926,
      "grad_norm": 0.5048968195915222,
      "learning_rate": 1.463245010951249e-05,
      "loss": 0.6486,
      "step": 2524
    },
    {
      "epoch": 3.229138331066485,
      "grad_norm": 0.5158227682113647,
      "learning_rate": 1.4618846419730713e-05,
      "loss": 0.6538,
      "step": 2525
    },
    {
      "epoch": 3.230418433474678,
      "grad_norm": 0.4812990427017212,
      "learning_rate": 1.4605247796206965e-05,
      "loss": 0.6315,
      "step": 2526
    },
    {
      "epoch": 3.2316985358828707,
      "grad_norm": 0.50927734375,
      "learning_rate": 1.459165424827504e-05,
      "loss": 0.6979,
      "step": 2527
    },
    {
      "epoch": 3.2329786382910632,
      "grad_norm": 0.5096977353096008,
      "learning_rate": 1.4578065785265234e-05,
      "loss": 0.6565,
      "step": 2528
    },
    {
      "epoch": 3.234258740699256,
      "grad_norm": 0.5484464764595032,
      "learning_rate": 1.456448241650437e-05,
      "loss": 0.6493,
      "step": 2529
    },
    {
      "epoch": 3.2355388431074488,
      "grad_norm": 0.52590411901474,
      "learning_rate": 1.455090415131577e-05,
      "loss": 0.649,
      "step": 2530
    },
    {
      "epoch": 3.2368189455156413,
      "grad_norm": 0.53493332862854,
      "learning_rate": 1.4537330999019248e-05,
      "loss": 0.6335,
      "step": 2531
    },
    {
      "epoch": 3.238099047923834,
      "grad_norm": 0.520137369632721,
      "learning_rate": 1.4523762968931103e-05,
      "loss": 0.6315,
      "step": 2532
    },
    {
      "epoch": 3.2393791503320264,
      "grad_norm": 0.5328630208969116,
      "learning_rate": 1.4510200070364135e-05,
      "loss": 0.6767,
      "step": 2533
    },
    {
      "epoch": 3.2406592527402194,
      "grad_norm": 0.5490903854370117,
      "learning_rate": 1.4496642312627614e-05,
      "loss": 0.6824,
      "step": 2534
    },
    {
      "epoch": 3.241939355148412,
      "grad_norm": 0.5077177882194519,
      "learning_rate": 1.4483089705027267e-05,
      "loss": 0.6279,
      "step": 2535
    },
    {
      "epoch": 3.2432194575566045,
      "grad_norm": 0.5176362991333008,
      "learning_rate": 1.4469542256865312e-05,
      "loss": 0.6216,
      "step": 2536
    },
    {
      "epoch": 3.244499559964797,
      "grad_norm": 0.5356772541999817,
      "learning_rate": 1.4455999977440414e-05,
      "loss": 0.666,
      "step": 2537
    },
    {
      "epoch": 3.24577966237299,
      "grad_norm": 0.5240214467048645,
      "learning_rate": 1.4442462876047682e-05,
      "loss": 0.643,
      "step": 2538
    },
    {
      "epoch": 3.2470597647811825,
      "grad_norm": 0.5250520706176758,
      "learning_rate": 1.4428930961978678e-05,
      "loss": 0.6494,
      "step": 2539
    },
    {
      "epoch": 3.248339867189375,
      "grad_norm": 0.5164676308631897,
      "learning_rate": 1.4415404244521419e-05,
      "loss": 0.6438,
      "step": 2540
    },
    {
      "epoch": 3.2496199695975676,
      "grad_norm": 0.48520180583000183,
      "learning_rate": 1.4401882732960317e-05,
      "loss": 0.6583,
      "step": 2541
    },
    {
      "epoch": 3.2509000720057606,
      "grad_norm": 0.5541596412658691,
      "learning_rate": 1.4388366436576253e-05,
      "loss": 0.6597,
      "step": 2542
    },
    {
      "epoch": 3.252180174413953,
      "grad_norm": 0.5362801551818848,
      "learning_rate": 1.4374855364646503e-05,
      "loss": 0.6521,
      "step": 2543
    },
    {
      "epoch": 3.2534602768221457,
      "grad_norm": 0.5064684748649597,
      "learning_rate": 1.4361349526444754e-05,
      "loss": 0.6117,
      "step": 2544
    },
    {
      "epoch": 3.2547403792303387,
      "grad_norm": 0.509677529335022,
      "learning_rate": 1.4347848931241122e-05,
      "loss": 0.6209,
      "step": 2545
    },
    {
      "epoch": 3.256020481638531,
      "grad_norm": 0.5341014862060547,
      "learning_rate": 1.4334353588302113e-05,
      "loss": 0.6763,
      "step": 2546
    },
    {
      "epoch": 3.2573005840467237,
      "grad_norm": 0.5059588551521301,
      "learning_rate": 1.4320863506890619e-05,
      "loss": 0.6113,
      "step": 2547
    },
    {
      "epoch": 3.2585806864549163,
      "grad_norm": 0.5250412225723267,
      "learning_rate": 1.4307378696265927e-05,
      "loss": 0.6918,
      "step": 2548
    },
    {
      "epoch": 3.259860788863109,
      "grad_norm": 0.5155032873153687,
      "learning_rate": 1.4293899165683725e-05,
      "loss": 0.637,
      "step": 2549
    },
    {
      "epoch": 3.261140891271302,
      "grad_norm": 0.5140616297721863,
      "learning_rate": 1.4280424924396043e-05,
      "loss": 0.6472,
      "step": 2550
    },
    {
      "epoch": 3.261140891271302,
      "eval_loss": 0.011485250666737556,
      "eval_runtime": 10.8206,
      "eval_samples_per_second": 45.839,
      "eval_steps_per_second": 5.73,
      "step": 2550
    },
    {
      "epoch": 3.2624209936794943,
      "grad_norm": 0.5688918232917786,
      "learning_rate": 1.42669559816513e-05,
      "loss": 0.6877,
      "step": 2551
    },
    {
      "epoch": 3.263701096087687,
      "grad_norm": 0.5144367814064026,
      "learning_rate": 1.4253492346694287e-05,
      "loss": 0.5954,
      "step": 2552
    },
    {
      "epoch": 3.26498119849588,
      "grad_norm": 0.516718864440918,
      "learning_rate": 1.4240034028766125e-05,
      "loss": 0.6494,
      "step": 2553
    },
    {
      "epoch": 3.2662613009040724,
      "grad_norm": 0.48907017707824707,
      "learning_rate": 1.4226581037104312e-05,
      "loss": 0.6202,
      "step": 2554
    },
    {
      "epoch": 3.267541403312265,
      "grad_norm": 0.5194664001464844,
      "learning_rate": 1.4213133380942678e-05,
      "loss": 0.6482,
      "step": 2555
    },
    {
      "epoch": 3.2688215057204575,
      "grad_norm": 0.5060011148452759,
      "learning_rate": 1.4199691069511387e-05,
      "loss": 0.6194,
      "step": 2556
    },
    {
      "epoch": 3.2701016081286505,
      "grad_norm": 0.5188465118408203,
      "learning_rate": 1.4186254112036944e-05,
      "loss": 0.6615,
      "step": 2557
    },
    {
      "epoch": 3.271381710536843,
      "grad_norm": 0.5025769472122192,
      "learning_rate": 1.4172822517742178e-05,
      "loss": 0.6684,
      "step": 2558
    },
    {
      "epoch": 3.2726618129450356,
      "grad_norm": 0.5170837044715881,
      "learning_rate": 1.4159396295846222e-05,
      "loss": 0.6904,
      "step": 2559
    },
    {
      "epoch": 3.273941915353228,
      "grad_norm": 0.45241689682006836,
      "learning_rate": 1.4145975455564544e-05,
      "loss": 0.6405,
      "step": 2560
    },
    {
      "epoch": 3.275222017761421,
      "grad_norm": 0.503831684589386,
      "learning_rate": 1.4132560006108911e-05,
      "loss": 0.6379,
      "step": 2561
    },
    {
      "epoch": 3.2765021201696136,
      "grad_norm": 0.5172925591468811,
      "learning_rate": 1.4119149956687369e-05,
      "loss": 0.5998,
      "step": 2562
    },
    {
      "epoch": 3.277782222577806,
      "grad_norm": 0.5172489285469055,
      "learning_rate": 1.4105745316504282e-05,
      "loss": 0.6599,
      "step": 2563
    },
    {
      "epoch": 3.2790623249859987,
      "grad_norm": 0.5125643014907837,
      "learning_rate": 1.4092346094760308e-05,
      "loss": 0.6366,
      "step": 2564
    },
    {
      "epoch": 3.2803424273941917,
      "grad_norm": 0.5013106465339661,
      "learning_rate": 1.4078952300652353e-05,
      "loss": 0.6429,
      "step": 2565
    },
    {
      "epoch": 3.2816225298023842,
      "grad_norm": 0.5247995257377625,
      "learning_rate": 1.4065563943373626e-05,
      "loss": 0.6329,
      "step": 2566
    },
    {
      "epoch": 3.2829026322105768,
      "grad_norm": 0.5207878947257996,
      "learning_rate": 1.4052181032113596e-05,
      "loss": 0.6384,
      "step": 2567
    },
    {
      "epoch": 3.2841827346187698,
      "grad_norm": 0.5144835710525513,
      "learning_rate": 1.4038803576057985e-05,
      "loss": 0.6779,
      "step": 2568
    },
    {
      "epoch": 3.2854628370269623,
      "grad_norm": 0.48141875863075256,
      "learning_rate": 1.4025431584388779e-05,
      "loss": 0.6067,
      "step": 2569
    },
    {
      "epoch": 3.286742939435155,
      "grad_norm": 0.4974786639213562,
      "learning_rate": 1.4012065066284218e-05,
      "loss": 0.6101,
      "step": 2570
    },
    {
      "epoch": 3.2880230418433474,
      "grad_norm": 0.4903869330883026,
      "learning_rate": 1.3998704030918784e-05,
      "loss": 0.6572,
      "step": 2571
    },
    {
      "epoch": 3.28930314425154,
      "grad_norm": 0.5125056505203247,
      "learning_rate": 1.3985348487463181e-05,
      "loss": 0.6627,
      "step": 2572
    },
    {
      "epoch": 3.290583246659733,
      "grad_norm": 0.5043832659721375,
      "learning_rate": 1.397199844508436e-05,
      "loss": 0.6403,
      "step": 2573
    },
    {
      "epoch": 3.2918633490679254,
      "grad_norm": 0.5195678472518921,
      "learning_rate": 1.3958653912945494e-05,
      "loss": 0.6521,
      "step": 2574
    },
    {
      "epoch": 3.293143451476118,
      "grad_norm": 0.48719099164009094,
      "learning_rate": 1.3945314900205959e-05,
      "loss": 0.6282,
      "step": 2575
    },
    {
      "epoch": 3.294423553884311,
      "grad_norm": 0.5019971132278442,
      "learning_rate": 1.3931981416021361e-05,
      "loss": 0.6179,
      "step": 2576
    },
    {
      "epoch": 3.2957036562925035,
      "grad_norm": 0.5336318612098694,
      "learning_rate": 1.3918653469543508e-05,
      "loss": 0.6344,
      "step": 2577
    },
    {
      "epoch": 3.296983758700696,
      "grad_norm": 0.512867271900177,
      "learning_rate": 1.3905331069920393e-05,
      "loss": 0.6314,
      "step": 2578
    },
    {
      "epoch": 3.2982638611088886,
      "grad_norm": 0.5135596394538879,
      "learning_rate": 1.3892014226296219e-05,
      "loss": 0.6474,
      "step": 2579
    },
    {
      "epoch": 3.299543963517081,
      "grad_norm": 0.5097352862358093,
      "learning_rate": 1.3878702947811369e-05,
      "loss": 0.6458,
      "step": 2580
    },
    {
      "epoch": 3.299543963517081,
      "eval_loss": 0.011472688056528568,
      "eval_runtime": 10.8197,
      "eval_samples_per_second": 45.842,
      "eval_steps_per_second": 5.73,
      "step": 2580
    },
    {
      "epoch": 3.300824065925274,
      "grad_norm": 0.5281054377555847,
      "learning_rate": 1.38653972436024e-05,
      "loss": 0.6547,
      "step": 2581
    },
    {
      "epoch": 3.3021041683334666,
      "grad_norm": 0.5279375910758972,
      "learning_rate": 1.3852097122802048e-05,
      "loss": 0.6368,
      "step": 2582
    },
    {
      "epoch": 3.303384270741659,
      "grad_norm": 0.4660930931568146,
      "learning_rate": 1.3838802594539228e-05,
      "loss": 0.5961,
      "step": 2583
    },
    {
      "epoch": 3.304664373149852,
      "grad_norm": 0.565133810043335,
      "learning_rate": 1.3825513667938995e-05,
      "loss": 0.6666,
      "step": 2584
    },
    {
      "epoch": 3.3059444755580447,
      "grad_norm": 0.5189217925071716,
      "learning_rate": 1.3812230352122567e-05,
      "loss": 0.6162,
      "step": 2585
    },
    {
      "epoch": 3.3072245779662373,
      "grad_norm": 0.5098384022712708,
      "learning_rate": 1.3798952656207329e-05,
      "loss": 0.6198,
      "step": 2586
    },
    {
      "epoch": 3.30850468037443,
      "grad_norm": 0.5222168564796448,
      "learning_rate": 1.378568058930678e-05,
      "loss": 0.6415,
      "step": 2587
    },
    {
      "epoch": 3.309784782782623,
      "grad_norm": 0.5284304022789001,
      "learning_rate": 1.3772414160530569e-05,
      "loss": 0.6771,
      "step": 2588
    },
    {
      "epoch": 3.3110648851908153,
      "grad_norm": 0.5394641160964966,
      "learning_rate": 1.3759153378984487e-05,
      "loss": 0.673,
      "step": 2589
    },
    {
      "epoch": 3.312344987599008,
      "grad_norm": 0.5227346420288086,
      "learning_rate": 1.3745898253770418e-05,
      "loss": 0.6511,
      "step": 2590
    },
    {
      "epoch": 3.3136250900072004,
      "grad_norm": 0.5089024901390076,
      "learning_rate": 1.3732648793986397e-05,
      "loss": 0.6723,
      "step": 2591
    },
    {
      "epoch": 3.3149051924153934,
      "grad_norm": 0.4928726553916931,
      "learning_rate": 1.3719405008726553e-05,
      "loss": 0.616,
      "step": 2592
    },
    {
      "epoch": 3.316185294823586,
      "grad_norm": 0.55116206407547,
      "learning_rate": 1.3706166907081113e-05,
      "loss": 0.6562,
      "step": 2593
    },
    {
      "epoch": 3.3174653972317785,
      "grad_norm": 0.5938664078712463,
      "learning_rate": 1.3692934498136422e-05,
      "loss": 0.6519,
      "step": 2594
    },
    {
      "epoch": 3.318745499639971,
      "grad_norm": 0.538509726524353,
      "learning_rate": 1.3679707790974917e-05,
      "loss": 0.6417,
      "step": 2595
    },
    {
      "epoch": 3.320025602048164,
      "grad_norm": 0.5012615919113159,
      "learning_rate": 1.3666486794675092e-05,
      "loss": 0.659,
      "step": 2596
    },
    {
      "epoch": 3.3213057044563565,
      "grad_norm": 0.5400484204292297,
      "learning_rate": 1.365327151831155e-05,
      "loss": 0.6484,
      "step": 2597
    },
    {
      "epoch": 3.322585806864549,
      "grad_norm": 0.604289174079895,
      "learning_rate": 1.364006197095497e-05,
      "loss": 0.6116,
      "step": 2598
    },
    {
      "epoch": 3.323865909272742,
      "grad_norm": 0.5269507169723511,
      "learning_rate": 1.3626858161672074e-05,
      "loss": 0.6393,
      "step": 2599
    },
    {
      "epoch": 3.3251460116809346,
      "grad_norm": 0.525745153427124,
      "learning_rate": 1.3613660099525665e-05,
      "loss": 0.632,
      "step": 2600
    },
    {
      "epoch": 3.326426114089127,
      "grad_norm": 0.5607315897941589,
      "learning_rate": 1.3600467793574605e-05,
      "loss": 0.6375,
      "step": 2601
    },
    {
      "epoch": 3.3277062164973197,
      "grad_norm": 0.5357958078384399,
      "learning_rate": 1.3587281252873781e-05,
      "loss": 0.6474,
      "step": 2602
    },
    {
      "epoch": 3.328986318905512,
      "grad_norm": 0.5517680644989014,
      "learning_rate": 1.3574100486474152e-05,
      "loss": 0.6332,
      "step": 2603
    },
    {
      "epoch": 3.330266421313705,
      "grad_norm": 0.5497151613235474,
      "learning_rate": 1.3560925503422696e-05,
      "loss": 0.6697,
      "step": 2604
    },
    {
      "epoch": 3.3315465237218977,
      "grad_norm": 0.5158340334892273,
      "learning_rate": 1.3547756312762422e-05,
      "loss": 0.6599,
      "step": 2605
    },
    {
      "epoch": 3.3328266261300903,
      "grad_norm": 0.5367072820663452,
      "learning_rate": 1.3534592923532364e-05,
      "loss": 0.6555,
      "step": 2606
    },
    {
      "epoch": 3.3341067285382833,
      "grad_norm": 0.5325165390968323,
      "learning_rate": 1.352143534476759e-05,
      "loss": 0.6397,
      "step": 2607
    },
    {
      "epoch": 3.335386830946476,
      "grad_norm": 0.5078912973403931,
      "learning_rate": 1.350828358549915e-05,
      "loss": 0.6309,
      "step": 2608
    },
    {
      "epoch": 3.3366669333546684,
      "grad_norm": 0.5415282845497131,
      "learning_rate": 1.3495137654754137e-05,
      "loss": 0.6722,
      "step": 2609
    },
    {
      "epoch": 3.337947035762861,
      "grad_norm": 0.5061556696891785,
      "learning_rate": 1.34819975615556e-05,
      "loss": 0.6203,
      "step": 2610
    },
    {
      "epoch": 3.337947035762861,
      "eval_loss": 0.011465904302895069,
      "eval_runtime": 10.8358,
      "eval_samples_per_second": 45.774,
      "eval_steps_per_second": 5.722,
      "step": 2610
    },
    {
      "epoch": 3.339227138171054,
      "grad_norm": 0.49840816855430603,
      "learning_rate": 1.3468863314922614e-05,
      "loss": 0.6177,
      "step": 2611
    },
    {
      "epoch": 3.3405072405792464,
      "grad_norm": 0.5172483325004578,
      "learning_rate": 1.3455734923870232e-05,
      "loss": 0.6419,
      "step": 2612
    },
    {
      "epoch": 3.341787342987439,
      "grad_norm": 0.5256161093711853,
      "learning_rate": 1.3442612397409482e-05,
      "loss": 0.6568,
      "step": 2613
    },
    {
      "epoch": 3.3430674453956315,
      "grad_norm": 0.5415335297584534,
      "learning_rate": 1.3429495744547377e-05,
      "loss": 0.6708,
      "step": 2614
    },
    {
      "epoch": 3.3443475478038245,
      "grad_norm": 0.5305517315864563,
      "learning_rate": 1.3416384974286895e-05,
      "loss": 0.6868,
      "step": 2615
    },
    {
      "epoch": 3.345627650212017,
      "grad_norm": 0.5212390422821045,
      "learning_rate": 1.3403280095626965e-05,
      "loss": 0.6799,
      "step": 2616
    },
    {
      "epoch": 3.3469077526202096,
      "grad_norm": 0.5267554521560669,
      "learning_rate": 1.3390181117562483e-05,
      "loss": 0.6157,
      "step": 2617
    },
    {
      "epoch": 3.348187855028402,
      "grad_norm": 0.5484142899513245,
      "learning_rate": 1.3377088049084307e-05,
      "loss": 0.654,
      "step": 2618
    },
    {
      "epoch": 3.349467957436595,
      "grad_norm": 0.5149903297424316,
      "learning_rate": 1.3364000899179205e-05,
      "loss": 0.6387,
      "step": 2619
    },
    {
      "epoch": 3.3507480598447876,
      "grad_norm": 0.5208218693733215,
      "learning_rate": 1.3350919676829915e-05,
      "loss": 0.653,
      "step": 2620
    },
    {
      "epoch": 3.35202816225298,
      "grad_norm": 0.5140671730041504,
      "learning_rate": 1.3337844391015096e-05,
      "loss": 0.6861,
      "step": 2621
    },
    {
      "epoch": 3.3533082646611727,
      "grad_norm": 0.5186113119125366,
      "learning_rate": 1.3324775050709321e-05,
      "loss": 0.6382,
      "step": 2622
    },
    {
      "epoch": 3.3545883670693657,
      "grad_norm": 0.523269534111023,
      "learning_rate": 1.3311711664883098e-05,
      "loss": 0.6498,
      "step": 2623
    },
    {
      "epoch": 3.3558684694775582,
      "grad_norm": 0.5470902919769287,
      "learning_rate": 1.3298654242502844e-05,
      "loss": 0.651,
      "step": 2624
    },
    {
      "epoch": 3.3571485718857508,
      "grad_norm": 0.5154792070388794,
      "learning_rate": 1.3285602792530869e-05,
      "loss": 0.6426,
      "step": 2625
    },
    {
      "epoch": 3.3584286742939433,
      "grad_norm": 0.5300922393798828,
      "learning_rate": 1.3272557323925406e-05,
      "loss": 0.6408,
      "step": 2626
    },
    {
      "epoch": 3.3597087767021363,
      "grad_norm": 0.5152605772018433,
      "learning_rate": 1.3259517845640576e-05,
      "loss": 0.6036,
      "step": 2627
    },
    {
      "epoch": 3.360988879110329,
      "grad_norm": 0.525360643863678,
      "learning_rate": 1.3246484366626367e-05,
      "loss": 0.6391,
      "step": 2628
    },
    {
      "epoch": 3.3622689815185214,
      "grad_norm": 0.503139317035675,
      "learning_rate": 1.3233456895828677e-05,
      "loss": 0.6347,
      "step": 2629
    },
    {
      "epoch": 3.3635490839267144,
      "grad_norm": 0.5070618391036987,
      "learning_rate": 1.3220435442189275e-05,
      "loss": 0.6235,
      "step": 2630
    },
    {
      "epoch": 3.364829186334907,
      "grad_norm": 0.5270324945449829,
      "learning_rate": 1.320742001464578e-05,
      "loss": 0.6244,
      "step": 2631
    },
    {
      "epoch": 3.3661092887430994,
      "grad_norm": 0.5384184718132019,
      "learning_rate": 1.3194410622131698e-05,
      "loss": 0.6447,
      "step": 2632
    },
    {
      "epoch": 3.367389391151292,
      "grad_norm": 0.548175036907196,
      "learning_rate": 1.3181407273576388e-05,
      "loss": 0.6591,
      "step": 2633
    },
    {
      "epoch": 3.3686694935594845,
      "grad_norm": 0.5638697147369385,
      "learning_rate": 1.3168409977905044e-05,
      "loss": 0.6661,
      "step": 2634
    },
    {
      "epoch": 3.3699495959676775,
      "grad_norm": 0.5330691933631897,
      "learning_rate": 1.315541874403873e-05,
      "loss": 0.6423,
      "step": 2635
    },
    {
      "epoch": 3.37122969837587,
      "grad_norm": 0.5639760494232178,
      "learning_rate": 1.3142433580894328e-05,
      "loss": 0.674,
      "step": 2636
    },
    {
      "epoch": 3.3725098007840626,
      "grad_norm": 0.5288316607475281,
      "learning_rate": 1.3129454497384565e-05,
      "loss": 0.6564,
      "step": 2637
    },
    {
      "epoch": 3.3737899031922556,
      "grad_norm": 0.49552685022354126,
      "learning_rate": 1.3116481502418e-05,
      "loss": 0.6562,
      "step": 2638
    },
    {
      "epoch": 3.375070005600448,
      "grad_norm": 0.5514019727706909,
      "learning_rate": 1.310351460489899e-05,
      "loss": 0.6928,
      "step": 2639
    },
    {
      "epoch": 3.3763501080086407,
      "grad_norm": 0.5216597318649292,
      "learning_rate": 1.3090553813727733e-05,
      "loss": 0.6511,
      "step": 2640
    },
    {
      "epoch": 3.3763501080086407,
      "eval_loss": 0.01146740186959505,
      "eval_runtime": 10.816,
      "eval_samples_per_second": 45.858,
      "eval_steps_per_second": 5.732,
      "step": 2640
    },
    {
      "epoch": 3.377630210416833,
      "grad_norm": 0.5338744521141052,
      "learning_rate": 1.3077599137800232e-05,
      "loss": 0.6437,
      "step": 2641
    },
    {
      "epoch": 3.378910312825026,
      "grad_norm": 0.5240645408630371,
      "learning_rate": 1.3064650586008268e-05,
      "loss": 0.6593,
      "step": 2642
    },
    {
      "epoch": 3.3801904152332187,
      "grad_norm": 0.5555424094200134,
      "learning_rate": 1.3051708167239441e-05,
      "loss": 0.6483,
      "step": 2643
    },
    {
      "epoch": 3.3814705176414113,
      "grad_norm": 0.5192124247550964,
      "learning_rate": 1.3038771890377151e-05,
      "loss": 0.6597,
      "step": 2644
    },
    {
      "epoch": 3.382750620049604,
      "grad_norm": 0.5266752243041992,
      "learning_rate": 1.3025841764300548e-05,
      "loss": 0.6681,
      "step": 2645
    },
    {
      "epoch": 3.384030722457797,
      "grad_norm": 0.5485540628433228,
      "learning_rate": 1.3012917797884593e-05,
      "loss": 0.6594,
      "step": 2646
    },
    {
      "epoch": 3.3853108248659893,
      "grad_norm": 0.5444266200065613,
      "learning_rate": 1.3000000000000006e-05,
      "loss": 0.6864,
      "step": 2647
    },
    {
      "epoch": 3.386590927274182,
      "grad_norm": 0.4996925890445709,
      "learning_rate": 1.2987088379513261e-05,
      "loss": 0.6583,
      "step": 2648
    },
    {
      "epoch": 3.3878710296823744,
      "grad_norm": 0.49421724677085876,
      "learning_rate": 1.2974182945286615e-05,
      "loss": 0.6437,
      "step": 2649
    },
    {
      "epoch": 3.3891511320905674,
      "grad_norm": 0.5644569396972656,
      "learning_rate": 1.2961283706178076e-05,
      "loss": 0.6565,
      "step": 2650
    },
    {
      "epoch": 3.39043123449876,
      "grad_norm": 0.5538458824157715,
      "learning_rate": 1.2948390671041373e-05,
      "loss": 0.6698,
      "step": 2651
    },
    {
      "epoch": 3.3917113369069525,
      "grad_norm": 0.5349107384681702,
      "learning_rate": 1.2935503848726003e-05,
      "loss": 0.6256,
      "step": 2652
    },
    {
      "epoch": 3.392991439315145,
      "grad_norm": 0.5281985402107239,
      "learning_rate": 1.2922623248077204e-05,
      "loss": 0.6339,
      "step": 2653
    },
    {
      "epoch": 3.394271541723338,
      "grad_norm": 0.5274403095245361,
      "learning_rate": 1.2909748877935913e-05,
      "loss": 0.6299,
      "step": 2654
    },
    {
      "epoch": 3.3955516441315305,
      "grad_norm": 0.5260682106018066,
      "learning_rate": 1.2896880747138813e-05,
      "loss": 0.6728,
      "step": 2655
    },
    {
      "epoch": 3.396831746539723,
      "grad_norm": 0.591275155544281,
      "learning_rate": 1.2884018864518313e-05,
      "loss": 0.6733,
      "step": 2656
    },
    {
      "epoch": 3.3981118489479156,
      "grad_norm": 0.603613555431366,
      "learning_rate": 1.28711632389025e-05,
      "loss": 0.6924,
      "step": 2657
    },
    {
      "epoch": 3.3993919513561086,
      "grad_norm": 0.5234869122505188,
      "learning_rate": 1.2858313879115198e-05,
      "loss": 0.6393,
      "step": 2658
    },
    {
      "epoch": 3.400672053764301,
      "grad_norm": 0.5910223126411438,
      "learning_rate": 1.2845470793975925e-05,
      "loss": 0.6349,
      "step": 2659
    },
    {
      "epoch": 3.4019521561724937,
      "grad_norm": 0.6169006824493408,
      "learning_rate": 1.2832633992299872e-05,
      "loss": 0.6649,
      "step": 2660
    },
    {
      "epoch": 3.4032322585806867,
      "grad_norm": 0.528428316116333,
      "learning_rate": 1.2819803482897944e-05,
      "loss": 0.676,
      "step": 2661
    },
    {
      "epoch": 3.404512360988879,
      "grad_norm": 0.5155686736106873,
      "learning_rate": 1.2806979274576703e-05,
      "loss": 0.6176,
      "step": 2662
    },
    {
      "epoch": 3.4057924633970718,
      "grad_norm": 0.6005485653877258,
      "learning_rate": 1.2794161376138403e-05,
      "loss": 0.6673,
      "step": 2663
    },
    {
      "epoch": 3.4070725658052643,
      "grad_norm": 0.5823571085929871,
      "learning_rate": 1.2781349796380968e-05,
      "loss": 0.66,
      "step": 2664
    },
    {
      "epoch": 3.408352668213457,
      "grad_norm": 0.5107295513153076,
      "learning_rate": 1.2768544544097967e-05,
      "loss": 0.6478,
      "step": 2665
    },
    {
      "epoch": 3.40963277062165,
      "grad_norm": 0.5336031913757324,
      "learning_rate": 1.2755745628078642e-05,
      "loss": 0.649,
      "step": 2666
    },
    {
      "epoch": 3.4109128730298424,
      "grad_norm": 0.5728003978729248,
      "learning_rate": 1.2742953057107888e-05,
      "loss": 0.6503,
      "step": 2667
    },
    {
      "epoch": 3.412192975438035,
      "grad_norm": 0.5376524925231934,
      "learning_rate": 1.2730166839966227e-05,
      "loss": 0.6417,
      "step": 2668
    },
    {
      "epoch": 3.413473077846228,
      "grad_norm": 0.4836972951889038,
      "learning_rate": 1.271738698542984e-05,
      "loss": 0.6223,
      "step": 2669
    },
    {
      "epoch": 3.4147531802544204,
      "grad_norm": 0.5275030732154846,
      "learning_rate": 1.2704613502270533e-05,
      "loss": 0.6453,
      "step": 2670
    },
    {
      "epoch": 3.4147531802544204,
      "eval_loss": 0.011455184780061245,
      "eval_runtime": 10.8108,
      "eval_samples_per_second": 45.88,
      "eval_steps_per_second": 5.735,
      "step": 2670
    },
    {
      "epoch": 3.416033282662613,
      "grad_norm": 0.53963702917099,
      "learning_rate": 1.2691846399255727e-05,
      "loss": 0.6553,
      "step": 2671
    },
    {
      "epoch": 3.4173133850708055,
      "grad_norm": 0.4867890179157257,
      "learning_rate": 1.2679085685148484e-05,
      "loss": 0.5999,
      "step": 2672
    },
    {
      "epoch": 3.4185934874789985,
      "grad_norm": 0.5570497512817383,
      "learning_rate": 1.2666331368707475e-05,
      "loss": 0.6594,
      "step": 2673
    },
    {
      "epoch": 3.419873589887191,
      "grad_norm": 0.5171359777450562,
      "learning_rate": 1.2653583458686964e-05,
      "loss": 0.6529,
      "step": 2674
    },
    {
      "epoch": 3.4211536922953836,
      "grad_norm": 0.5122377872467041,
      "learning_rate": 1.2640841963836836e-05,
      "loss": 0.6122,
      "step": 2675
    },
    {
      "epoch": 3.422433794703576,
      "grad_norm": 0.5108626484870911,
      "learning_rate": 1.2628106892902575e-05,
      "loss": 0.6291,
      "step": 2676
    },
    {
      "epoch": 3.423713897111769,
      "grad_norm": 0.49061429500579834,
      "learning_rate": 1.2615378254625235e-05,
      "loss": 0.6277,
      "step": 2677
    },
    {
      "epoch": 3.4249939995199616,
      "grad_norm": 0.5067298412322998,
      "learning_rate": 1.2602656057741469e-05,
      "loss": 0.6387,
      "step": 2678
    },
    {
      "epoch": 3.426274101928154,
      "grad_norm": 0.5020197033882141,
      "learning_rate": 1.2589940310983519e-05,
      "loss": 0.6141,
      "step": 2679
    },
    {
      "epoch": 3.4275542043363467,
      "grad_norm": 0.5395712852478027,
      "learning_rate": 1.2577231023079176e-05,
      "loss": 0.6547,
      "step": 2680
    },
    {
      "epoch": 3.4288343067445397,
      "grad_norm": 0.5163862109184265,
      "learning_rate": 1.2564528202751811e-05,
      "loss": 0.6267,
      "step": 2681
    },
    {
      "epoch": 3.4301144091527322,
      "grad_norm": 0.5071302652359009,
      "learning_rate": 1.2551831858720368e-05,
      "loss": 0.6522,
      "step": 2682
    },
    {
      "epoch": 3.431394511560925,
      "grad_norm": 0.5683829188346863,
      "learning_rate": 1.253914199969931e-05,
      "loss": 0.665,
      "step": 2683
    },
    {
      "epoch": 3.4326746139691178,
      "grad_norm": 0.5407935976982117,
      "learning_rate": 1.2526458634398683e-05,
      "loss": 0.646,
      "step": 2684
    },
    {
      "epoch": 3.4339547163773103,
      "grad_norm": 0.4977572560310364,
      "learning_rate": 1.2513781771524073e-05,
      "loss": 0.6393,
      "step": 2685
    },
    {
      "epoch": 3.435234818785503,
      "grad_norm": 0.5278409719467163,
      "learning_rate": 1.2501111419776579e-05,
      "loss": 0.6315,
      "step": 2686
    },
    {
      "epoch": 3.4365149211936954,
      "grad_norm": 0.5229158401489258,
      "learning_rate": 1.2488447587852859e-05,
      "loss": 0.663,
      "step": 2687
    },
    {
      "epoch": 3.437795023601888,
      "grad_norm": 0.508394181728363,
      "learning_rate": 1.2475790284445064e-05,
      "loss": 0.6556,
      "step": 2688
    },
    {
      "epoch": 3.439075126010081,
      "grad_norm": 0.509078860282898,
      "learning_rate": 1.2463139518240896e-05,
      "loss": 0.6802,
      "step": 2689
    },
    {
      "epoch": 3.4403552284182735,
      "grad_norm": 0.49052590131759644,
      "learning_rate": 1.2450495297923566e-05,
      "loss": 0.6257,
      "step": 2690
    },
    {
      "epoch": 3.441635330826466,
      "grad_norm": 0.5256490111351013,
      "learning_rate": 1.2437857632171762e-05,
      "loss": 0.6561,
      "step": 2691
    },
    {
      "epoch": 3.442915433234659,
      "grad_norm": 0.49466726183891296,
      "learning_rate": 1.2425226529659702e-05,
      "loss": 0.638,
      "step": 2692
    },
    {
      "epoch": 3.4441955356428515,
      "grad_norm": 0.512749433517456,
      "learning_rate": 1.2412601999057099e-05,
      "loss": 0.6479,
      "step": 2693
    },
    {
      "epoch": 3.445475638051044,
      "grad_norm": 0.5261350274085999,
      "learning_rate": 1.239998404902914e-05,
      "loss": 0.6593,
      "step": 2694
    },
    {
      "epoch": 3.4467557404592366,
      "grad_norm": 0.5123868584632874,
      "learning_rate": 1.2387372688236499e-05,
      "loss": 0.632,
      "step": 2695
    },
    {
      "epoch": 3.448035842867429,
      "grad_norm": 0.5266541838645935,
      "learning_rate": 1.2374767925335343e-05,
      "loss": 0.6584,
      "step": 2696
    },
    {
      "epoch": 3.449315945275622,
      "grad_norm": 0.5414618253707886,
      "learning_rate": 1.2362169768977284e-05,
      "loss": 0.6433,
      "step": 2697
    },
    {
      "epoch": 3.4505960476838147,
      "grad_norm": 0.5119932293891907,
      "learning_rate": 1.2349578227809417e-05,
      "loss": 0.6532,
      "step": 2698
    },
    {
      "epoch": 3.451876150092007,
      "grad_norm": 0.49727195501327515,
      "learning_rate": 1.2336993310474306e-05,
      "loss": 0.6137,
      "step": 2699
    },
    {
      "epoch": 3.4531562525002,
      "grad_norm": 0.48369887471199036,
      "learning_rate": 1.2324415025609939e-05,
      "loss": 0.5994,
      "step": 2700
    },
    {
      "epoch": 3.4531562525002,
      "eval_loss": 0.011453227140009403,
      "eval_runtime": 10.84,
      "eval_samples_per_second": 45.757,
      "eval_steps_per_second": 5.72,
      "step": 2700
    },
    {
      "epoch": 3.4544363549083927,
      "grad_norm": 0.5059986114501953,
      "learning_rate": 1.2311843381849773e-05,
      "loss": 0.6355,
      "step": 2701
    },
    {
      "epoch": 3.4557164573165853,
      "grad_norm": 0.5174420475959778,
      "learning_rate": 1.2299278387822713e-05,
      "loss": 0.6414,
      "step": 2702
    },
    {
      "epoch": 3.456996559724778,
      "grad_norm": 0.5212859511375427,
      "learning_rate": 1.2286720052153069e-05,
      "loss": 0.6569,
      "step": 2703
    },
    {
      "epoch": 3.458276662132971,
      "grad_norm": 0.5277246832847595,
      "learning_rate": 1.2274168383460613e-05,
      "loss": 0.678,
      "step": 2704
    },
    {
      "epoch": 3.4595567645411633,
      "grad_norm": 0.5248231887817383,
      "learning_rate": 1.2261623390360535e-05,
      "loss": 0.6425,
      "step": 2705
    },
    {
      "epoch": 3.460836866949356,
      "grad_norm": 0.5126795768737793,
      "learning_rate": 1.224908508146342e-05,
      "loss": 0.6548,
      "step": 2706
    },
    {
      "epoch": 3.4621169693575484,
      "grad_norm": 0.48268821835517883,
      "learning_rate": 1.2236553465375289e-05,
      "loss": 0.6461,
      "step": 2707
    },
    {
      "epoch": 3.4633970717657414,
      "grad_norm": 0.5046666860580444,
      "learning_rate": 1.2224028550697571e-05,
      "loss": 0.6482,
      "step": 2708
    },
    {
      "epoch": 3.464677174173934,
      "grad_norm": 0.5129714012145996,
      "learning_rate": 1.2211510346027066e-05,
      "loss": 0.6383,
      "step": 2709
    },
    {
      "epoch": 3.4659572765821265,
      "grad_norm": 0.5030511021614075,
      "learning_rate": 1.2198998859956004e-05,
      "loss": 0.6613,
      "step": 2710
    },
    {
      "epoch": 3.467237378990319,
      "grad_norm": 0.5071579217910767,
      "learning_rate": 1.218649410107199e-05,
      "loss": 0.6331,
      "step": 2711
    },
    {
      "epoch": 3.468517481398512,
      "grad_norm": 0.460328072309494,
      "learning_rate": 1.2173996077957997e-05,
      "loss": 0.6461,
      "step": 2712
    },
    {
      "epoch": 3.4697975838067046,
      "grad_norm": 0.5210601687431335,
      "learning_rate": 1.2161504799192394e-05,
      "loss": 0.6331,
      "step": 2713
    },
    {
      "epoch": 3.471077686214897,
      "grad_norm": 0.5281786322593689,
      "learning_rate": 1.214902027334892e-05,
      "loss": 0.6774,
      "step": 2714
    },
    {
      "epoch": 3.47235778862309,
      "grad_norm": 0.5002842545509338,
      "learning_rate": 1.2136542508996658e-05,
      "loss": 0.6504,
      "step": 2715
    },
    {
      "epoch": 3.4736378910312826,
      "grad_norm": 0.5501790642738342,
      "learning_rate": 1.2124071514700079e-05,
      "loss": 0.6822,
      "step": 2716
    },
    {
      "epoch": 3.474917993439475,
      "grad_norm": 0.4851137101650238,
      "learning_rate": 1.2111607299018995e-05,
      "loss": 0.626,
      "step": 2717
    },
    {
      "epoch": 3.4761980958476677,
      "grad_norm": 0.4627394676208496,
      "learning_rate": 1.2099149870508548e-05,
      "loss": 0.6263,
      "step": 2718
    },
    {
      "epoch": 3.4774781982558602,
      "grad_norm": 0.5304125547409058,
      "learning_rate": 1.2086699237719249e-05,
      "loss": 0.6281,
      "step": 2719
    },
    {
      "epoch": 3.4787583006640532,
      "grad_norm": 0.5508079528808594,
      "learning_rate": 1.2074255409196937e-05,
      "loss": 0.6424,
      "step": 2720
    },
    {
      "epoch": 3.4800384030722458,
      "grad_norm": 0.4871533513069153,
      "learning_rate": 1.2061818393482763e-05,
      "loss": 0.6421,
      "step": 2721
    },
    {
      "epoch": 3.4813185054804383,
      "grad_norm": 0.5122039318084717,
      "learning_rate": 1.2049388199113218e-05,
      "loss": 0.6372,
      "step": 2722
    },
    {
      "epoch": 3.4825986078886313,
      "grad_norm": 0.5093011260032654,
      "learning_rate": 1.2036964834620123e-05,
      "loss": 0.643,
      "step": 2723
    },
    {
      "epoch": 3.483878710296824,
      "grad_norm": 0.49953410029411316,
      "learning_rate": 1.2024548308530576e-05,
      "loss": 0.658,
      "step": 2724
    },
    {
      "epoch": 3.4851588127050164,
      "grad_norm": 0.5171728134155273,
      "learning_rate": 1.2012138629367013e-05,
      "loss": 0.6037,
      "step": 2725
    },
    {
      "epoch": 3.486438915113209,
      "grad_norm": 0.4938555657863617,
      "learning_rate": 1.1999735805647165e-05,
      "loss": 0.6135,
      "step": 2726
    },
    {
      "epoch": 3.487719017521402,
      "grad_norm": 0.5075860023498535,
      "learning_rate": 1.1987339845884036e-05,
      "loss": 0.6341,
      "step": 2727
    },
    {
      "epoch": 3.4889991199295944,
      "grad_norm": 0.5005190372467041,
      "learning_rate": 1.1974950758585941e-05,
      "loss": 0.6515,
      "step": 2728
    },
    {
      "epoch": 3.490279222337787,
      "grad_norm": 0.5378018617630005,
      "learning_rate": 1.1962568552256476e-05,
      "loss": 0.6805,
      "step": 2729
    },
    {
      "epoch": 3.4915593247459795,
      "grad_norm": 0.5122777223587036,
      "learning_rate": 1.1950193235394507e-05,
      "loss": 0.6556,
      "step": 2730
    },
    {
      "epoch": 3.4915593247459795,
      "eval_loss": 0.011446916498243809,
      "eval_runtime": 10.8125,
      "eval_samples_per_second": 45.873,
      "eval_steps_per_second": 5.734,
      "step": 2730
    },
    {
      "epoch": 3.4928394271541725,
      "grad_norm": 0.4979049563407898,
      "learning_rate": 1.193782481649417e-05,
      "loss": 0.6541,
      "step": 2731
    },
    {
      "epoch": 3.494119529562365,
      "grad_norm": 0.521952748298645,
      "learning_rate": 1.192546330404487e-05,
      "loss": 0.6411,
      "step": 2732
    },
    {
      "epoch": 3.4953996319705576,
      "grad_norm": 0.5584136843681335,
      "learning_rate": 1.1913108706531284e-05,
      "loss": 0.6401,
      "step": 2733
    },
    {
      "epoch": 3.49667973437875,
      "grad_norm": 0.543571412563324,
      "learning_rate": 1.1900761032433308e-05,
      "loss": 0.6724,
      "step": 2734
    },
    {
      "epoch": 3.497959836786943,
      "grad_norm": 0.5464795231819153,
      "learning_rate": 1.1888420290226127e-05,
      "loss": 0.6272,
      "step": 2735
    },
    {
      "epoch": 3.4992399391951357,
      "grad_norm": 0.47890669107437134,
      "learning_rate": 1.187608648838014e-05,
      "loss": 0.6469,
      "step": 2736
    },
    {
      "epoch": 3.500520041603328,
      "grad_norm": 0.5000232458114624,
      "learning_rate": 1.1863759635360994e-05,
      "loss": 0.6453,
      "step": 2737
    },
    {
      "epoch": 3.501800144011521,
      "grad_norm": 0.5134245753288269,
      "learning_rate": 1.185143973962956e-05,
      "loss": 0.6464,
      "step": 2738
    },
    {
      "epoch": 3.5030802464197137,
      "grad_norm": 0.5332192778587341,
      "learning_rate": 1.1839126809641953e-05,
      "loss": 0.6527,
      "step": 2739
    },
    {
      "epoch": 3.5043603488279063,
      "grad_norm": 0.4757227599620819,
      "learning_rate": 1.1826820853849465e-05,
      "loss": 0.6408,
      "step": 2740
    },
    {
      "epoch": 3.505640451236099,
      "grad_norm": 0.46543270349502563,
      "learning_rate": 1.1814521880698649e-05,
      "loss": 0.6258,
      "step": 2741
    },
    {
      "epoch": 3.5069205536442913,
      "grad_norm": 0.5437485575675964,
      "learning_rate": 1.1802229898631246e-05,
      "loss": 0.6337,
      "step": 2742
    },
    {
      "epoch": 3.5082006560524843,
      "grad_norm": 0.5336166620254517,
      "learning_rate": 1.1789944916084177e-05,
      "loss": 0.6424,
      "step": 2743
    },
    {
      "epoch": 3.509480758460677,
      "grad_norm": 0.48311272263526917,
      "learning_rate": 1.1777666941489594e-05,
      "loss": 0.6377,
      "step": 2744
    },
    {
      "epoch": 3.5107608608688694,
      "grad_norm": 0.5338751673698425,
      "learning_rate": 1.176539598327482e-05,
      "loss": 0.6573,
      "step": 2745
    },
    {
      "epoch": 3.5120409632770624,
      "grad_norm": 0.5336971282958984,
      "learning_rate": 1.1753132049862362e-05,
      "loss": 0.6708,
      "step": 2746
    },
    {
      "epoch": 3.513321065685255,
      "grad_norm": 0.4926799535751343,
      "learning_rate": 1.1740875149669907e-05,
      "loss": 0.6375,
      "step": 2747
    },
    {
      "epoch": 3.5146011680934475,
      "grad_norm": 0.5325849056243896,
      "learning_rate": 1.1728625291110323e-05,
      "loss": 0.6544,
      "step": 2748
    },
    {
      "epoch": 3.51588127050164,
      "grad_norm": 0.4882565438747406,
      "learning_rate": 1.1716382482591629e-05,
      "loss": 0.6412,
      "step": 2749
    },
    {
      "epoch": 3.5171613729098326,
      "grad_norm": 0.5467479825019836,
      "learning_rate": 1.170414673251702e-05,
      "loss": 0.6751,
      "step": 2750
    },
    {
      "epoch": 3.5184414753180255,
      "grad_norm": 0.526167631149292,
      "learning_rate": 1.169191804928484e-05,
      "loss": 0.6613,
      "step": 2751
    },
    {
      "epoch": 3.519721577726218,
      "grad_norm": 0.5064482092857361,
      "learning_rate": 1.1679696441288582e-05,
      "loss": 0.6645,
      "step": 2752
    },
    {
      "epoch": 3.5210016801344106,
      "grad_norm": 0.47246575355529785,
      "learning_rate": 1.1667481916916876e-05,
      "loss": 0.5996,
      "step": 2753
    },
    {
      "epoch": 3.5222817825426036,
      "grad_norm": 0.5616881251335144,
      "learning_rate": 1.165527448455351e-05,
      "loss": 0.6659,
      "step": 2754
    },
    {
      "epoch": 3.523561884950796,
      "grad_norm": 0.509503185749054,
      "learning_rate": 1.1643074152577382e-05,
      "loss": 0.6317,
      "step": 2755
    },
    {
      "epoch": 3.5248419873589887,
      "grad_norm": 0.4886564314365387,
      "learning_rate": 1.1630880929362526e-05,
      "loss": 0.6265,
      "step": 2756
    },
    {
      "epoch": 3.5261220897671812,
      "grad_norm": 0.5238721370697021,
      "learning_rate": 1.1618694823278104e-05,
      "loss": 0.6558,
      "step": 2757
    },
    {
      "epoch": 3.5274021921753738,
      "grad_norm": 0.5514553189277649,
      "learning_rate": 1.1606515842688379e-05,
      "loss": 0.654,
      "step": 2758
    },
    {
      "epoch": 3.5286822945835667,
      "grad_norm": 0.5037204027175903,
      "learning_rate": 1.1594343995952728e-05,
      "loss": 0.6883,
      "step": 2759
    },
    {
      "epoch": 3.5299623969917593,
      "grad_norm": 0.5038042664527893,
      "learning_rate": 1.1582179291425644e-05,
      "loss": 0.6657,
      "step": 2760
    },
    {
      "epoch": 3.5299623969917593,
      "eval_loss": 0.011440154165029526,
      "eval_runtime": 10.8133,
      "eval_samples_per_second": 45.869,
      "eval_steps_per_second": 5.734,
      "step": 2760
    },
    {
      "epoch": 3.5312424993999523,
      "grad_norm": 0.5204631686210632,
      "learning_rate": 1.157002173745669e-05,
      "loss": 0.6247,
      "step": 2761
    },
    {
      "epoch": 3.532522601808145,
      "grad_norm": 0.5278136730194092,
      "learning_rate": 1.1557871342390547e-05,
      "loss": 0.6395,
      "step": 2762
    },
    {
      "epoch": 3.5338027042163374,
      "grad_norm": 0.4869731068611145,
      "learning_rate": 1.1545728114566985e-05,
      "loss": 0.6196,
      "step": 2763
    },
    {
      "epoch": 3.53508280662453,
      "grad_norm": 0.49234598875045776,
      "learning_rate": 1.1533592062320822e-05,
      "loss": 0.6644,
      "step": 2764
    },
    {
      "epoch": 3.5363629090327224,
      "grad_norm": 0.5424562692642212,
      "learning_rate": 1.1521463193981983e-05,
      "loss": 0.6567,
      "step": 2765
    },
    {
      "epoch": 3.5376430114409154,
      "grad_norm": 0.5251913070678711,
      "learning_rate": 1.1509341517875452e-05,
      "loss": 0.6352,
      "step": 2766
    },
    {
      "epoch": 3.538923113849108,
      "grad_norm": 0.48312902450561523,
      "learning_rate": 1.1497227042321269e-05,
      "loss": 0.6426,
      "step": 2767
    },
    {
      "epoch": 3.5402032162573005,
      "grad_norm": 0.5144852995872498,
      "learning_rate": 1.1485119775634548e-05,
      "loss": 0.632,
      "step": 2768
    },
    {
      "epoch": 3.5414833186654935,
      "grad_norm": 0.5160422921180725,
      "learning_rate": 1.1473019726125427e-05,
      "loss": 0.6454,
      "step": 2769
    },
    {
      "epoch": 3.542763421073686,
      "grad_norm": 0.49337631464004517,
      "learning_rate": 1.1460926902099123e-05,
      "loss": 0.6459,
      "step": 2770
    },
    {
      "epoch": 3.5440435234818786,
      "grad_norm": 0.4793124496936798,
      "learning_rate": 1.1448841311855881e-05,
      "loss": 0.6547,
      "step": 2771
    },
    {
      "epoch": 3.545323625890071,
      "grad_norm": 0.5613303184509277,
      "learning_rate": 1.1436762963690966e-05,
      "loss": 0.6404,
      "step": 2772
    },
    {
      "epoch": 3.5466037282982636,
      "grad_norm": 0.5109605193138123,
      "learning_rate": 1.142469186589469e-05,
      "loss": 0.6778,
      "step": 2773
    },
    {
      "epoch": 3.5478838307064566,
      "grad_norm": 0.482348769903183,
      "learning_rate": 1.1412628026752395e-05,
      "loss": 0.6193,
      "step": 2774
    },
    {
      "epoch": 3.549163933114649,
      "grad_norm": 0.4726719856262207,
      "learning_rate": 1.1400571454544412e-05,
      "loss": 0.6364,
      "step": 2775
    },
    {
      "epoch": 3.5504440355228417,
      "grad_norm": 0.5502317547798157,
      "learning_rate": 1.1388522157546107e-05,
      "loss": 0.6477,
      "step": 2776
    },
    {
      "epoch": 3.5517241379310347,
      "grad_norm": 0.49400949478149414,
      "learning_rate": 1.1376480144027853e-05,
      "loss": 0.6442,
      "step": 2777
    },
    {
      "epoch": 3.5530042403392272,
      "grad_norm": 0.4953269064426422,
      "learning_rate": 1.136444542225501e-05,
      "loss": 0.6482,
      "step": 2778
    },
    {
      "epoch": 3.55428434274742,
      "grad_norm": 0.5284993052482605,
      "learning_rate": 1.1352418000487941e-05,
      "loss": 0.6645,
      "step": 2779
    },
    {
      "epoch": 3.5555644451556123,
      "grad_norm": 0.46703362464904785,
      "learning_rate": 1.1340397886982007e-05,
      "loss": 0.6266,
      "step": 2780
    },
    {
      "epoch": 3.556844547563805,
      "grad_norm": 0.5016201734542847,
      "learning_rate": 1.1328385089987532e-05,
      "loss": 0.6454,
      "step": 2781
    },
    {
      "epoch": 3.558124649971998,
      "grad_norm": 0.47830647230148315,
      "learning_rate": 1.1316379617749837e-05,
      "loss": 0.6147,
      "step": 2782
    },
    {
      "epoch": 3.5594047523801904,
      "grad_norm": 0.5023000836372375,
      "learning_rate": 1.1304381478509216e-05,
      "loss": 0.6637,
      "step": 2783
    },
    {
      "epoch": 3.560684854788383,
      "grad_norm": 0.47462597489356995,
      "learning_rate": 1.1292390680500903e-05,
      "loss": 0.6588,
      "step": 2784
    },
    {
      "epoch": 3.561964957196576,
      "grad_norm": 0.4568457305431366,
      "learning_rate": 1.1280407231955125e-05,
      "loss": 0.6512,
      "step": 2785
    },
    {
      "epoch": 3.5632450596047684,
      "grad_norm": 0.4648769497871399,
      "learning_rate": 1.1268431141097054e-05,
      "loss": 0.6008,
      "step": 2786
    },
    {
      "epoch": 3.564525162012961,
      "grad_norm": 0.4593886137008667,
      "learning_rate": 1.1256462416146803e-05,
      "loss": 0.6227,
      "step": 2787
    },
    {
      "epoch": 3.5658052644211535,
      "grad_norm": 0.46304184198379517,
      "learning_rate": 1.1244501065319437e-05,
      "loss": 0.6471,
      "step": 2788
    },
    {
      "epoch": 3.567085366829346,
      "grad_norm": 0.47933557629585266,
      "learning_rate": 1.1232547096824966e-05,
      "loss": 0.6404,
      "step": 2789
    },
    {
      "epoch": 3.568365469237539,
      "grad_norm": 0.47464719414711,
      "learning_rate": 1.1220600518868311e-05,
      "loss": 0.6511,
      "step": 2790
    },
    {
      "epoch": 3.568365469237539,
      "eval_loss": 0.01143835112452507,
      "eval_runtime": 10.8175,
      "eval_samples_per_second": 45.852,
      "eval_steps_per_second": 5.731,
      "step": 2790
    },
    {
      "epoch": 3.5696455716457316,
      "grad_norm": 0.5069913864135742,
      "learning_rate": 1.1208661339649348e-05,
      "loss": 0.677,
      "step": 2791
    },
    {
      "epoch": 3.5709256740539246,
      "grad_norm": 0.5090366005897522,
      "learning_rate": 1.1196729567362863e-05,
      "loss": 0.663,
      "step": 2792
    },
    {
      "epoch": 3.572205776462117,
      "grad_norm": 0.4997399151325226,
      "learning_rate": 1.118480521019854e-05,
      "loss": 0.6341,
      "step": 2793
    },
    {
      "epoch": 3.5734858788703097,
      "grad_norm": 0.5092949867248535,
      "learning_rate": 1.1172888276341013e-05,
      "loss": 0.6536,
      "step": 2794
    },
    {
      "epoch": 3.574765981278502,
      "grad_norm": 0.525589108467102,
      "learning_rate": 1.1160978773969773e-05,
      "loss": 0.6612,
      "step": 2795
    },
    {
      "epoch": 3.5760460836866947,
      "grad_norm": 0.5021042823791504,
      "learning_rate": 1.1149076711259248e-05,
      "loss": 0.6583,
      "step": 2796
    },
    {
      "epoch": 3.5773261860948877,
      "grad_norm": 0.4602411091327667,
      "learning_rate": 1.1137182096378751e-05,
      "loss": 0.6315,
      "step": 2797
    },
    {
      "epoch": 3.5786062885030803,
      "grad_norm": 0.4679986536502838,
      "learning_rate": 1.1125294937492468e-05,
      "loss": 0.6492,
      "step": 2798
    },
    {
      "epoch": 3.579886390911273,
      "grad_norm": 0.5496346950531006,
      "learning_rate": 1.1113415242759484e-05,
      "loss": 0.6444,
      "step": 2799
    },
    {
      "epoch": 3.581166493319466,
      "grad_norm": 0.49317190051078796,
      "learning_rate": 1.1101543020333757e-05,
      "loss": 0.6536,
      "step": 2800
    },
    {
      "epoch": 3.5824465957276583,
      "grad_norm": 0.5776232481002808,
      "learning_rate": 1.1089678278364108e-05,
      "loss": 0.6702,
      "step": 2801
    },
    {
      "epoch": 3.583726698135851,
      "grad_norm": 0.5339002013206482,
      "learning_rate": 1.1077821024994231e-05,
      "loss": 0.6184,
      "step": 2802
    },
    {
      "epoch": 3.5850068005440434,
      "grad_norm": 0.4782545268535614,
      "learning_rate": 1.1065971268362685e-05,
      "loss": 0.6611,
      "step": 2803
    },
    {
      "epoch": 3.586286902952236,
      "grad_norm": 0.5438824892044067,
      "learning_rate": 1.1054129016602863e-05,
      "loss": 0.6524,
      "step": 2804
    },
    {
      "epoch": 3.587567005360429,
      "grad_norm": 0.5965319275856018,
      "learning_rate": 1.1042294277843029e-05,
      "loss": 0.661,
      "step": 2805
    },
    {
      "epoch": 3.5888471077686215,
      "grad_norm": 0.5224551558494568,
      "learning_rate": 1.1030467060206287e-05,
      "loss": 0.6552,
      "step": 2806
    },
    {
      "epoch": 3.590127210176814,
      "grad_norm": 0.48367223143577576,
      "learning_rate": 1.1018647371810562e-05,
      "loss": 0.6691,
      "step": 2807
    },
    {
      "epoch": 3.591407312585007,
      "grad_norm": 0.5312662124633789,
      "learning_rate": 1.100683522076863e-05,
      "loss": 0.6279,
      "step": 2808
    },
    {
      "epoch": 3.5926874149931995,
      "grad_norm": 0.5946574211120605,
      "learning_rate": 1.099503061518809e-05,
      "loss": 0.6191,
      "step": 2809
    },
    {
      "epoch": 3.593967517401392,
      "grad_norm": 0.515278697013855,
      "learning_rate": 1.0983233563171346e-05,
      "loss": 0.6127,
      "step": 2810
    },
    {
      "epoch": 3.5952476198095846,
      "grad_norm": 0.506566047668457,
      "learning_rate": 1.0971444072815638e-05,
      "loss": 0.6602,
      "step": 2811
    },
    {
      "epoch": 3.596527722217777,
      "grad_norm": 0.5674377679824829,
      "learning_rate": 1.095966215221301e-05,
      "loss": 0.6686,
      "step": 2812
    },
    {
      "epoch": 3.59780782462597,
      "grad_norm": 0.5017653107643127,
      "learning_rate": 1.0947887809450299e-05,
      "loss": 0.6537,
      "step": 2813
    },
    {
      "epoch": 3.5990879270341627,
      "grad_norm": 0.4885178506374359,
      "learning_rate": 1.0936121052609155e-05,
      "loss": 0.6036,
      "step": 2814
    },
    {
      "epoch": 3.6003680294423552,
      "grad_norm": 0.5086835026741028,
      "learning_rate": 1.0924361889766017e-05,
      "loss": 0.6384,
      "step": 2815
    },
    {
      "epoch": 3.601648131850548,
      "grad_norm": 0.4810003340244293,
      "learning_rate": 1.0912610328992103e-05,
      "loss": 0.6589,
      "step": 2816
    },
    {
      "epoch": 3.6029282342587408,
      "grad_norm": 0.5347673296928406,
      "learning_rate": 1.0900866378353424e-05,
      "loss": 0.7077,
      "step": 2817
    },
    {
      "epoch": 3.6042083366669333,
      "grad_norm": 0.5160301327705383,
      "learning_rate": 1.0889130045910772e-05,
      "loss": 0.6366,
      "step": 2818
    },
    {
      "epoch": 3.605488439075126,
      "grad_norm": 0.5085987448692322,
      "learning_rate": 1.0877401339719688e-05,
      "loss": 0.6437,
      "step": 2819
    },
    {
      "epoch": 3.606768541483319,
      "grad_norm": 0.4960525631904602,
      "learning_rate": 1.0865680267830507e-05,
      "loss": 0.6685,
      "step": 2820
    },
    {
      "epoch": 3.606768541483319,
      "eval_loss": 0.011422453448176384,
      "eval_runtime": 10.8201,
      "eval_samples_per_second": 45.84,
      "eval_steps_per_second": 5.73,
      "step": 2820
    },
    {
      "epoch": 3.6080486438915114,
      "grad_norm": 0.49201685190200806,
      "learning_rate": 1.0853966838288296e-05,
      "loss": 0.6598,
      "step": 2821
    },
    {
      "epoch": 3.609328746299704,
      "grad_norm": 0.506462037563324,
      "learning_rate": 1.0842261059132894e-05,
      "loss": 0.6386,
      "step": 2822
    },
    {
      "epoch": 3.610608848707897,
      "grad_norm": 0.4947039783000946,
      "learning_rate": 1.0830562938398893e-05,
      "loss": 0.636,
      "step": 2823
    },
    {
      "epoch": 3.6118889511160894,
      "grad_norm": 0.4936864376068115,
      "learning_rate": 1.081887248411561e-05,
      "loss": 0.6114,
      "step": 2824
    },
    {
      "epoch": 3.613169053524282,
      "grad_norm": 0.5415510535240173,
      "learning_rate": 1.0807189704307115e-05,
      "loss": 0.7048,
      "step": 2825
    },
    {
      "epoch": 3.6144491559324745,
      "grad_norm": 0.5129798054695129,
      "learning_rate": 1.079551460699221e-05,
      "loss": 0.6489,
      "step": 2826
    },
    {
      "epoch": 3.615729258340667,
      "grad_norm": 0.512713611125946,
      "learning_rate": 1.0783847200184408e-05,
      "loss": 0.6219,
      "step": 2827
    },
    {
      "epoch": 3.61700936074886,
      "grad_norm": 0.4892199635505676,
      "learning_rate": 1.077218749189196e-05,
      "loss": 0.6557,
      "step": 2828
    },
    {
      "epoch": 3.6182894631570526,
      "grad_norm": 0.48060908913612366,
      "learning_rate": 1.0760535490117836e-05,
      "loss": 0.6134,
      "step": 2829
    },
    {
      "epoch": 3.619569565565245,
      "grad_norm": 0.49317482113838196,
      "learning_rate": 1.0748891202859695e-05,
      "loss": 0.6694,
      "step": 2830
    },
    {
      "epoch": 3.620849667973438,
      "grad_norm": 0.49855637550354004,
      "learning_rate": 1.0737254638109923e-05,
      "loss": 0.6386,
      "step": 2831
    },
    {
      "epoch": 3.6221297703816306,
      "grad_norm": 0.5070940852165222,
      "learning_rate": 1.07256258038556e-05,
      "loss": 0.6555,
      "step": 2832
    },
    {
      "epoch": 3.623409872789823,
      "grad_norm": 0.4797641634941101,
      "learning_rate": 1.0714004708078485e-05,
      "loss": 0.6129,
      "step": 2833
    },
    {
      "epoch": 3.6246899751980157,
      "grad_norm": 0.5025620460510254,
      "learning_rate": 1.0702391358755041e-05,
      "loss": 0.6756,
      "step": 2834
    },
    {
      "epoch": 3.6259700776062083,
      "grad_norm": 0.48994600772857666,
      "learning_rate": 1.0690785763856421e-05,
      "loss": 0.6337,
      "step": 2835
    },
    {
      "epoch": 3.6272501800144012,
      "grad_norm": 0.507883608341217,
      "learning_rate": 1.0679187931348432e-05,
      "loss": 0.6368,
      "step": 2836
    },
    {
      "epoch": 3.628530282422594,
      "grad_norm": 0.4853685200214386,
      "learning_rate": 1.0667597869191567e-05,
      "loss": 0.6215,
      "step": 2837
    },
    {
      "epoch": 3.6298103848307863,
      "grad_norm": 0.5017647743225098,
      "learning_rate": 1.0656015585340995e-05,
      "loss": 0.5984,
      "step": 2838
    },
    {
      "epoch": 3.6310904872389793,
      "grad_norm": 0.4920690953731537,
      "learning_rate": 1.064444108774652e-05,
      "loss": 0.6285,
      "step": 2839
    },
    {
      "epoch": 3.632370589647172,
      "grad_norm": 0.47269406914711,
      "learning_rate": 1.0632874384352623e-05,
      "loss": 0.6599,
      "step": 2840
    },
    {
      "epoch": 3.6336506920553644,
      "grad_norm": 0.5047468543052673,
      "learning_rate": 1.0621315483098442e-05,
      "loss": 0.6241,
      "step": 2841
    },
    {
      "epoch": 3.634930794463557,
      "grad_norm": 0.539953887462616,
      "learning_rate": 1.0609764391917728e-05,
      "loss": 0.617,
      "step": 2842
    },
    {
      "epoch": 3.6362108968717495,
      "grad_norm": 0.4730015695095062,
      "learning_rate": 1.0598221118738898e-05,
      "loss": 0.5904,
      "step": 2843
    },
    {
      "epoch": 3.6374909992799425,
      "grad_norm": 0.4837453067302704,
      "learning_rate": 1.0586685671485003e-05,
      "loss": 0.6658,
      "step": 2844
    },
    {
      "epoch": 3.638771101688135,
      "grad_norm": 0.49402302503585815,
      "learning_rate": 1.0575158058073707e-05,
      "loss": 0.6557,
      "step": 2845
    },
    {
      "epoch": 3.6400512040963275,
      "grad_norm": 0.47058239579200745,
      "learning_rate": 1.0563638286417305e-05,
      "loss": 0.6006,
      "step": 2846
    },
    {
      "epoch": 3.6413313065045205,
      "grad_norm": 0.5083207488059998,
      "learning_rate": 1.055212636442271e-05,
      "loss": 0.6693,
      "step": 2847
    },
    {
      "epoch": 3.642611408912713,
      "grad_norm": 0.5132161974906921,
      "learning_rate": 1.0540622299991444e-05,
      "loss": 0.6507,
      "step": 2848
    },
    {
      "epoch": 3.6438915113209056,
      "grad_norm": 0.48563486337661743,
      "learning_rate": 1.0529126101019651e-05,
      "loss": 0.6325,
      "step": 2849
    },
    {
      "epoch": 3.645171613729098,
      "grad_norm": 0.505311131477356,
      "learning_rate": 1.0517637775398045e-05,
      "loss": 0.6536,
      "step": 2850
    },
    {
      "epoch": 3.645171613729098,
      "eval_loss": 0.011422289535403252,
      "eval_runtime": 10.8019,
      "eval_samples_per_second": 45.918,
      "eval_steps_per_second": 5.74,
      "step": 2850
    },
    {
      "epoch": 3.646451716137291,
      "grad_norm": 0.524929940700531,
      "learning_rate": 1.0506157331011965e-05,
      "loss": 0.6236,
      "step": 2851
    },
    {
      "epoch": 3.6477318185454837,
      "grad_norm": 0.5175740122795105,
      "learning_rate": 1.0494684775741338e-05,
      "loss": 0.6581,
      "step": 2852
    },
    {
      "epoch": 3.649011920953676,
      "grad_norm": 0.4799844026565552,
      "learning_rate": 1.0483220117460654e-05,
      "loss": 0.6281,
      "step": 2853
    },
    {
      "epoch": 3.650292023361869,
      "grad_norm": 0.46461695432662964,
      "learning_rate": 1.0471763364039e-05,
      "loss": 0.6112,
      "step": 2854
    },
    {
      "epoch": 3.6515721257700617,
      "grad_norm": 0.48377180099487305,
      "learning_rate": 1.0460314523340045e-05,
      "loss": 0.6676,
      "step": 2855
    },
    {
      "epoch": 3.6528522281782543,
      "grad_norm": 0.4818356931209564,
      "learning_rate": 1.0448873603221998e-05,
      "loss": 0.6566,
      "step": 2856
    },
    {
      "epoch": 3.654132330586447,
      "grad_norm": 0.4960228502750397,
      "learning_rate": 1.0437440611537663e-05,
      "loss": 0.6622,
      "step": 2857
    },
    {
      "epoch": 3.6554124329946394,
      "grad_norm": 0.4892972707748413,
      "learning_rate": 1.0426015556134392e-05,
      "loss": 0.6362,
      "step": 2858
    },
    {
      "epoch": 3.6566925354028323,
      "grad_norm": 0.48846346139907837,
      "learning_rate": 1.0414598444854074e-05,
      "loss": 0.6535,
      "step": 2859
    },
    {
      "epoch": 3.657972637811025,
      "grad_norm": 0.51702481508255,
      "learning_rate": 1.0403189285533162e-05,
      "loss": 0.6679,
      "step": 2860
    },
    {
      "epoch": 3.6592527402192174,
      "grad_norm": 0.5167385935783386,
      "learning_rate": 1.0391788086002655e-05,
      "loss": 0.6498,
      "step": 2861
    },
    {
      "epoch": 3.6605328426274104,
      "grad_norm": 0.5271387696266174,
      "learning_rate": 1.0380394854088064e-05,
      "loss": 0.686,
      "step": 2862
    },
    {
      "epoch": 3.661812945035603,
      "grad_norm": 0.49380338191986084,
      "learning_rate": 1.036900959760946e-05,
      "loss": 0.6709,
      "step": 2863
    },
    {
      "epoch": 3.6630930474437955,
      "grad_norm": 0.464388906955719,
      "learning_rate": 1.0357632324381433e-05,
      "loss": 0.6197,
      "step": 2864
    },
    {
      "epoch": 3.664373149851988,
      "grad_norm": 0.5040779709815979,
      "learning_rate": 1.0346263042213068e-05,
      "loss": 0.6411,
      "step": 2865
    },
    {
      "epoch": 3.6656532522601806,
      "grad_norm": 0.4891960322856903,
      "learning_rate": 1.0334901758907998e-05,
      "loss": 0.6368,
      "step": 2866
    },
    {
      "epoch": 3.6669333546683736,
      "grad_norm": 0.5071341395378113,
      "learning_rate": 1.0323548482264356e-05,
      "loss": 0.6285,
      "step": 2867
    },
    {
      "epoch": 3.668213457076566,
      "grad_norm": 0.45557788014411926,
      "learning_rate": 1.0312203220074766e-05,
      "loss": 0.6305,
      "step": 2868
    },
    {
      "epoch": 3.6694935594847586,
      "grad_norm": 0.4724888205528259,
      "learning_rate": 1.0300865980126363e-05,
      "loss": 0.6364,
      "step": 2869
    },
    {
      "epoch": 3.6707736618929516,
      "grad_norm": 0.4951120615005493,
      "learning_rate": 1.0289536770200783e-05,
      "loss": 0.6458,
      "step": 2870
    },
    {
      "epoch": 3.672053764301144,
      "grad_norm": 0.4920676350593567,
      "learning_rate": 1.0278215598074133e-05,
      "loss": 0.6417,
      "step": 2871
    },
    {
      "epoch": 3.6733338667093367,
      "grad_norm": 0.4896887540817261,
      "learning_rate": 1.0266902471517008e-05,
      "loss": 0.6563,
      "step": 2872
    },
    {
      "epoch": 3.6746139691175292,
      "grad_norm": 0.48132839798927307,
      "learning_rate": 1.02555973982945e-05,
      "loss": 0.6233,
      "step": 2873
    },
    {
      "epoch": 3.675894071525722,
      "grad_norm": 0.5013761520385742,
      "learning_rate": 1.0244300386166137e-05,
      "loss": 0.6549,
      "step": 2874
    },
    {
      "epoch": 3.6771741739339148,
      "grad_norm": 0.5326442122459412,
      "learning_rate": 1.0233011442885946e-05,
      "loss": 0.6725,
      "step": 2875
    },
    {
      "epoch": 3.6784542763421073,
      "grad_norm": 0.49917691946029663,
      "learning_rate": 1.0221730576202407e-05,
      "loss": 0.6516,
      "step": 2876
    },
    {
      "epoch": 3.6797343787503003,
      "grad_norm": 0.48615798354148865,
      "learning_rate": 1.0210457793858444e-05,
      "loss": 0.6384,
      "step": 2877
    },
    {
      "epoch": 3.681014481158493,
      "grad_norm": 0.4995875954627991,
      "learning_rate": 1.019919310359144e-05,
      "loss": 0.6386,
      "step": 2878
    },
    {
      "epoch": 3.6822945835666854,
      "grad_norm": 0.5343180298805237,
      "learning_rate": 1.0187936513133245e-05,
      "loss": 0.638,
      "step": 2879
    },
    {
      "epoch": 3.683574685974878,
      "grad_norm": 0.5254859924316406,
      "learning_rate": 1.0176688030210107e-05,
      "loss": 0.6579,
      "step": 2880
    },
    {
      "epoch": 3.683574685974878,
      "eval_loss": 0.011417878791689873,
      "eval_runtime": 10.8112,
      "eval_samples_per_second": 45.878,
      "eval_steps_per_second": 5.735,
      "step": 2880
    },
    {
      "epoch": 3.6848547883830705,
      "grad_norm": 0.5034304261207581,
      "learning_rate": 1.016544766254274e-05,
      "loss": 0.6858,
      "step": 2881
    },
    {
      "epoch": 3.6861348907912634,
      "grad_norm": 0.5118473768234253,
      "learning_rate": 1.0154215417846292e-05,
      "loss": 0.6455,
      "step": 2882
    },
    {
      "epoch": 3.687414993199456,
      "grad_norm": 0.46881163120269775,
      "learning_rate": 1.0142991303830301e-05,
      "loss": 0.6804,
      "step": 2883
    },
    {
      "epoch": 3.6886950956076485,
      "grad_norm": 0.491364061832428,
      "learning_rate": 1.013177532819876e-05,
      "loss": 0.6491,
      "step": 2884
    },
    {
      "epoch": 3.6899751980158415,
      "grad_norm": 0.5065906643867493,
      "learning_rate": 1.0120567498650065e-05,
      "loss": 0.6307,
      "step": 2885
    },
    {
      "epoch": 3.691255300424034,
      "grad_norm": 0.5274394750595093,
      "learning_rate": 1.010936782287702e-05,
      "loss": 0.6584,
      "step": 2886
    },
    {
      "epoch": 3.6925354028322266,
      "grad_norm": 0.48760563135147095,
      "learning_rate": 1.0098176308566817e-05,
      "loss": 0.6349,
      "step": 2887
    },
    {
      "epoch": 3.693815505240419,
      "grad_norm": 0.5175226330757141,
      "learning_rate": 1.0086992963401068e-05,
      "loss": 0.6308,
      "step": 2888
    },
    {
      "epoch": 3.6950956076486117,
      "grad_norm": 0.4898320138454437,
      "learning_rate": 1.0075817795055776e-05,
      "loss": 0.6529,
      "step": 2889
    },
    {
      "epoch": 3.6963757100568047,
      "grad_norm": 0.5060924887657166,
      "learning_rate": 1.0064650811201316e-05,
      "loss": 0.6729,
      "step": 2890
    },
    {
      "epoch": 3.697655812464997,
      "grad_norm": 0.523146390914917,
      "learning_rate": 1.0053492019502457e-05,
      "loss": 0.6432,
      "step": 2891
    },
    {
      "epoch": 3.6989359148731897,
      "grad_norm": 0.5231835246086121,
      "learning_rate": 1.0042341427618357e-05,
      "loss": 0.6803,
      "step": 2892
    },
    {
      "epoch": 3.7002160172813827,
      "grad_norm": 0.4628106653690338,
      "learning_rate": 1.0031199043202506e-05,
      "loss": 0.6436,
      "step": 2893
    },
    {
      "epoch": 3.7014961196895753,
      "grad_norm": 0.5030839443206787,
      "learning_rate": 1.0020064873902803e-05,
      "loss": 0.6836,
      "step": 2894
    },
    {
      "epoch": 3.702776222097768,
      "grad_norm": 0.521946907043457,
      "learning_rate": 1.0008938927361495e-05,
      "loss": 0.6512,
      "step": 2895
    },
    {
      "epoch": 3.7040563245059603,
      "grad_norm": 0.4723970293998718,
      "learning_rate": 9.997821211215172e-06,
      "loss": 0.6504,
      "step": 2896
    },
    {
      "epoch": 3.705336426914153,
      "grad_norm": 0.502835750579834,
      "learning_rate": 9.986711733094785e-06,
      "loss": 0.6635,
      "step": 2897
    },
    {
      "epoch": 3.706616529322346,
      "grad_norm": 0.5013492703437805,
      "learning_rate": 9.975610500625642e-06,
      "loss": 0.6466,
      "step": 2898
    },
    {
      "epoch": 3.7078966317305384,
      "grad_norm": 0.5020231604576111,
      "learning_rate": 9.964517521427366e-06,
      "loss": 0.636,
      "step": 2899
    },
    {
      "epoch": 3.709176734138731,
      "grad_norm": 0.5422610640525818,
      "learning_rate": 9.953432803113931e-06,
      "loss": 0.6306,
      "step": 2900
    },
    {
      "epoch": 3.710456836546924,
      "grad_norm": 0.4779011905193329,
      "learning_rate": 9.94235635329365e-06,
      "loss": 0.6402,
      "step": 2901
    },
    {
      "epoch": 3.7117369389551165,
      "grad_norm": 0.4914515018463135,
      "learning_rate": 9.931288179569135e-06,
      "loss": 0.6696,
      "step": 2902
    },
    {
      "epoch": 3.713017041363309,
      "grad_norm": 0.47003981471061707,
      "learning_rate": 9.920228289537337e-06,
      "loss": 0.6421,
      "step": 2903
    },
    {
      "epoch": 3.7142971437715016,
      "grad_norm": 0.5225586891174316,
      "learning_rate": 9.909176690789522e-06,
      "loss": 0.6798,
      "step": 2904
    },
    {
      "epoch": 3.7155772461796945,
      "grad_norm": 0.5420576930046082,
      "learning_rate": 9.89813339091125e-06,
      "loss": 0.6905,
      "step": 2905
    },
    {
      "epoch": 3.716857348587887,
      "grad_norm": 0.48183944821357727,
      "learning_rate": 9.887098397482398e-06,
      "loss": 0.6766,
      "step": 2906
    },
    {
      "epoch": 3.7181374509960796,
      "grad_norm": 0.4869091510772705,
      "learning_rate": 9.876071718077142e-06,
      "loss": 0.6234,
      "step": 2907
    },
    {
      "epoch": 3.7194175534042726,
      "grad_norm": 0.49029341340065,
      "learning_rate": 9.865053360263939e-06,
      "loss": 0.6243,
      "step": 2908
    },
    {
      "epoch": 3.720697655812465,
      "grad_norm": 0.48741549253463745,
      "learning_rate": 9.854043331605544e-06,
      "loss": 0.6258,
      "step": 2909
    },
    {
      "epoch": 3.7219777582206577,
      "grad_norm": 0.49600744247436523,
      "learning_rate": 9.843041639659003e-06,
      "loss": 0.6108,
      "step": 2910
    },
    {
      "epoch": 3.7219777582206577,
      "eval_loss": 0.011410364881157875,
      "eval_runtime": 10.8253,
      "eval_samples_per_second": 45.819,
      "eval_steps_per_second": 5.727,
      "step": 2910
    },
    {
      "epoch": 3.7232578606288502,
      "grad_norm": 0.5103457570075989,
      "learning_rate": 9.832048291975616e-06,
      "loss": 0.6527,
      "step": 2911
    },
    {
      "epoch": 3.7245379630370428,
      "grad_norm": 0.4938619136810303,
      "learning_rate": 9.821063296100979e-06,
      "loss": 0.648,
      "step": 2912
    },
    {
      "epoch": 3.7258180654452357,
      "grad_norm": 0.5137816667556763,
      "learning_rate": 9.810086659574948e-06,
      "loss": 0.6243,
      "step": 2913
    },
    {
      "epoch": 3.7270981678534283,
      "grad_norm": 0.5702383518218994,
      "learning_rate": 9.79911838993163e-06,
      "loss": 0.6777,
      "step": 2914
    },
    {
      "epoch": 3.728378270261621,
      "grad_norm": 0.47129806876182556,
      "learning_rate": 9.788158494699405e-06,
      "loss": 0.6548,
      "step": 2915
    },
    {
      "epoch": 3.729658372669814,
      "grad_norm": 0.503034770488739,
      "learning_rate": 9.777206981400904e-06,
      "loss": 0.6464,
      "step": 2916
    },
    {
      "epoch": 3.7309384750780064,
      "grad_norm": 0.5019398331642151,
      "learning_rate": 9.76626385755299e-06,
      "loss": 0.6133,
      "step": 2917
    },
    {
      "epoch": 3.732218577486199,
      "grad_norm": 0.5174933671951294,
      "learning_rate": 9.755329130666781e-06,
      "loss": 0.6369,
      "step": 2918
    },
    {
      "epoch": 3.7334986798943914,
      "grad_norm": 0.49872708320617676,
      "learning_rate": 9.744402808247635e-06,
      "loss": 0.6577,
      "step": 2919
    },
    {
      "epoch": 3.734778782302584,
      "grad_norm": 0.5082205533981323,
      "learning_rate": 9.733484897795122e-06,
      "loss": 0.6412,
      "step": 2920
    },
    {
      "epoch": 3.736058884710777,
      "grad_norm": 0.5398660898208618,
      "learning_rate": 9.722575406803056e-06,
      "loss": 0.6766,
      "step": 2921
    },
    {
      "epoch": 3.7373389871189695,
      "grad_norm": 0.5091952085494995,
      "learning_rate": 9.711674342759474e-06,
      "loss": 0.6488,
      "step": 2922
    },
    {
      "epoch": 3.738619089527162,
      "grad_norm": 0.4718760550022125,
      "learning_rate": 9.700781713146608e-06,
      "loss": 0.6402,
      "step": 2923
    },
    {
      "epoch": 3.739899191935355,
      "grad_norm": 0.49748873710632324,
      "learning_rate": 9.68989752544093e-06,
      "loss": 0.6422,
      "step": 2924
    },
    {
      "epoch": 3.7411792943435476,
      "grad_norm": 0.517709493637085,
      "learning_rate": 9.679021787113083e-06,
      "loss": 0.6643,
      "step": 2925
    },
    {
      "epoch": 3.74245939675174,
      "grad_norm": 0.5044804811477661,
      "learning_rate": 9.668154505627942e-06,
      "loss": 0.6962,
      "step": 2926
    },
    {
      "epoch": 3.7437394991599326,
      "grad_norm": 0.49303412437438965,
      "learning_rate": 9.657295688444567e-06,
      "loss": 0.6537,
      "step": 2927
    },
    {
      "epoch": 3.745019601568125,
      "grad_norm": 0.5132579207420349,
      "learning_rate": 9.646445343016195e-06,
      "loss": 0.6316,
      "step": 2928
    },
    {
      "epoch": 3.746299703976318,
      "grad_norm": 0.5101869106292725,
      "learning_rate": 9.635603476790268e-06,
      "loss": 0.6443,
      "step": 2929
    },
    {
      "epoch": 3.7475798063845107,
      "grad_norm": 0.4614020586013794,
      "learning_rate": 9.624770097208404e-06,
      "loss": 0.6141,
      "step": 2930
    },
    {
      "epoch": 3.7488599087927033,
      "grad_norm": 0.49720925092697144,
      "learning_rate": 9.613945211706377e-06,
      "loss": 0.6154,
      "step": 2931
    },
    {
      "epoch": 3.7501400112008962,
      "grad_norm": 0.4910092055797577,
      "learning_rate": 9.603128827714153e-06,
      "loss": 0.6379,
      "step": 2932
    },
    {
      "epoch": 3.751420113609089,
      "grad_norm": 0.5107735395431519,
      "learning_rate": 9.592320952655863e-06,
      "loss": 0.6444,
      "step": 2933
    },
    {
      "epoch": 3.7527002160172813,
      "grad_norm": 0.4670794904232025,
      "learning_rate": 9.581521593949774e-06,
      "loss": 0.6376,
      "step": 2934
    },
    {
      "epoch": 3.753980318425474,
      "grad_norm": 0.5206142067909241,
      "learning_rate": 9.570730759008331e-06,
      "loss": 0.6711,
      "step": 2935
    },
    {
      "epoch": 3.755260420833667,
      "grad_norm": 0.5948630571365356,
      "learning_rate": 9.559948455238128e-06,
      "loss": 0.6411,
      "step": 2936
    },
    {
      "epoch": 3.7565405232418594,
      "grad_norm": 0.5366334915161133,
      "learning_rate": 9.549174690039884e-06,
      "loss": 0.6315,
      "step": 2937
    },
    {
      "epoch": 3.757820625650052,
      "grad_norm": 0.4711497128009796,
      "learning_rate": 9.538409470808472e-06,
      "loss": 0.6188,
      "step": 2938
    },
    {
      "epoch": 3.759100728058245,
      "grad_norm": 0.48475149273872375,
      "learning_rate": 9.527652804932907e-06,
      "loss": 0.6524,
      "step": 2939
    },
    {
      "epoch": 3.7603808304664375,
      "grad_norm": 0.49723920226097107,
      "learning_rate": 9.516904699796308e-06,
      "loss": 0.6607,
      "step": 2940
    },
    {
      "epoch": 3.7603808304664375,
      "eval_loss": 0.011400481685996056,
      "eval_runtime": 10.8026,
      "eval_samples_per_second": 45.915,
      "eval_steps_per_second": 5.739,
      "step": 2940
    },
    {
      "epoch": 3.76166093287463,
      "grad_norm": 0.48554638028144836,
      "learning_rate": 9.50616516277594e-06,
      "loss": 0.6016,
      "step": 2941
    },
    {
      "epoch": 3.7629410352828225,
      "grad_norm": 0.47717761993408203,
      "learning_rate": 9.495434201243187e-06,
      "loss": 0.6703,
      "step": 2942
    },
    {
      "epoch": 3.764221137691015,
      "grad_norm": 0.4773055613040924,
      "learning_rate": 9.484711822563527e-06,
      "loss": 0.655,
      "step": 2943
    },
    {
      "epoch": 3.765501240099208,
      "grad_norm": 0.4841598868370056,
      "learning_rate": 9.473998034096565e-06,
      "loss": 0.68,
      "step": 2944
    },
    {
      "epoch": 3.7667813425074006,
      "grad_norm": 0.4953411817550659,
      "learning_rate": 9.463292843196011e-06,
      "loss": 0.6081,
      "step": 2945
    },
    {
      "epoch": 3.768061444915593,
      "grad_norm": 0.49032536149024963,
      "learning_rate": 9.452596257209658e-06,
      "loss": 0.6028,
      "step": 2946
    },
    {
      "epoch": 3.769341547323786,
      "grad_norm": 0.5038009881973267,
      "learning_rate": 9.441908283479404e-06,
      "loss": 0.6199,
      "step": 2947
    },
    {
      "epoch": 3.7706216497319787,
      "grad_norm": 0.5327941179275513,
      "learning_rate": 9.431228929341243e-06,
      "loss": 0.6676,
      "step": 2948
    },
    {
      "epoch": 3.771901752140171,
      "grad_norm": 0.5019099116325378,
      "learning_rate": 9.420558202125232e-06,
      "loss": 0.6427,
      "step": 2949
    },
    {
      "epoch": 3.7731818545483637,
      "grad_norm": 0.4896283447742462,
      "learning_rate": 9.40989610915553e-06,
      "loss": 0.6164,
      "step": 2950
    },
    {
      "epoch": 3.7744619569565563,
      "grad_norm": 0.4576800465583801,
      "learning_rate": 9.399242657750346e-06,
      "loss": 0.63,
      "step": 2951
    },
    {
      "epoch": 3.7757420593647493,
      "grad_norm": 0.4906672239303589,
      "learning_rate": 9.388597855221976e-06,
      "loss": 0.6331,
      "step": 2952
    },
    {
      "epoch": 3.777022161772942,
      "grad_norm": 0.4907695949077606,
      "learning_rate": 9.377961708876777e-06,
      "loss": 0.637,
      "step": 2953
    },
    {
      "epoch": 3.7783022641811344,
      "grad_norm": 0.468185693025589,
      "learning_rate": 9.367334226015154e-06,
      "loss": 0.6266,
      "step": 2954
    },
    {
      "epoch": 3.7795823665893273,
      "grad_norm": 0.48238587379455566,
      "learning_rate": 9.356715413931575e-06,
      "loss": 0.6371,
      "step": 2955
    },
    {
      "epoch": 3.78086246899752,
      "grad_norm": 0.5327696800231934,
      "learning_rate": 9.34610527991456e-06,
      "loss": 0.6423,
      "step": 2956
    },
    {
      "epoch": 3.7821425714057124,
      "grad_norm": 0.5187394022941589,
      "learning_rate": 9.335503831246656e-06,
      "loss": 0.6623,
      "step": 2957
    },
    {
      "epoch": 3.783422673813905,
      "grad_norm": 0.48732247948646545,
      "learning_rate": 9.324911075204465e-06,
      "loss": 0.6383,
      "step": 2958
    },
    {
      "epoch": 3.7847027762220975,
      "grad_norm": 0.52056884765625,
      "learning_rate": 9.314327019058621e-06,
      "loss": 0.644,
      "step": 2959
    },
    {
      "epoch": 3.7859828786302905,
      "grad_norm": 0.4897262752056122,
      "learning_rate": 9.303751670073771e-06,
      "loss": 0.6604,
      "step": 2960
    },
    {
      "epoch": 3.787262981038483,
      "grad_norm": 0.48963961005210876,
      "learning_rate": 9.293185035508607e-06,
      "loss": 0.6076,
      "step": 2961
    },
    {
      "epoch": 3.788543083446676,
      "grad_norm": 0.49382975697517395,
      "learning_rate": 9.28262712261583e-06,
      "loss": 0.6795,
      "step": 2962
    },
    {
      "epoch": 3.7898231858548685,
      "grad_norm": 0.4907773435115814,
      "learning_rate": 9.272077938642147e-06,
      "loss": 0.6154,
      "step": 2963
    },
    {
      "epoch": 3.791103288263061,
      "grad_norm": 0.47583645582199097,
      "learning_rate": 9.261537490828283e-06,
      "loss": 0.6431,
      "step": 2964
    },
    {
      "epoch": 3.7923833906712536,
      "grad_norm": 0.4625856280326843,
      "learning_rate": 9.251005786408973e-06,
      "loss": 0.6053,
      "step": 2965
    },
    {
      "epoch": 3.793663493079446,
      "grad_norm": 0.48354119062423706,
      "learning_rate": 9.240482832612931e-06,
      "loss": 0.6581,
      "step": 2966
    },
    {
      "epoch": 3.794943595487639,
      "grad_norm": 0.5079452991485596,
      "learning_rate": 9.229968636662878e-06,
      "loss": 0.6286,
      "step": 2967
    },
    {
      "epoch": 3.7962236978958317,
      "grad_norm": 0.46758586168289185,
      "learning_rate": 9.219463205775532e-06,
      "loss": 0.6346,
      "step": 2968
    },
    {
      "epoch": 3.7975038003040242,
      "grad_norm": 0.4636211395263672,
      "learning_rate": 9.208966547161573e-06,
      "loss": 0.6216,
      "step": 2969
    },
    {
      "epoch": 3.798783902712217,
      "grad_norm": 0.5054153800010681,
      "learning_rate": 9.198478668025675e-06,
      "loss": 0.6312,
      "step": 2970
    },
    {
      "epoch": 3.798783902712217,
      "eval_loss": 0.011398686096072197,
      "eval_runtime": 10.8332,
      "eval_samples_per_second": 45.785,
      "eval_steps_per_second": 5.723,
      "step": 2970
    },
    {
      "epoch": 3.8000640051204098,
      "grad_norm": 0.5667929649353027,
      "learning_rate": 9.18799957556649e-06,
      "loss": 0.6501,
      "step": 2971
    },
    {
      "epoch": 3.8013441075286023,
      "grad_norm": 0.4893105626106262,
      "learning_rate": 9.177529276976619e-06,
      "loss": 0.6199,
      "step": 2972
    },
    {
      "epoch": 3.802624209936795,
      "grad_norm": 0.4836716651916504,
      "learning_rate": 9.167067779442646e-06,
      "loss": 0.6253,
      "step": 2973
    },
    {
      "epoch": 3.8039043123449874,
      "grad_norm": 0.562601625919342,
      "learning_rate": 9.156615090145116e-06,
      "loss": 0.6525,
      "step": 2974
    },
    {
      "epoch": 3.8051844147531804,
      "grad_norm": 0.5605821013450623,
      "learning_rate": 9.146171216258506e-06,
      "loss": 0.6649,
      "step": 2975
    },
    {
      "epoch": 3.806464517161373,
      "grad_norm": 0.49592655897140503,
      "learning_rate": 9.135736164951264e-06,
      "loss": 0.6323,
      "step": 2976
    },
    {
      "epoch": 3.8077446195695654,
      "grad_norm": 0.5060100555419922,
      "learning_rate": 9.125309943385778e-06,
      "loss": 0.6173,
      "step": 2977
    },
    {
      "epoch": 3.8090247219777584,
      "grad_norm": 0.5618041753768921,
      "learning_rate": 9.114892558718363e-06,
      "loss": 0.6597,
      "step": 2978
    },
    {
      "epoch": 3.810304824385951,
      "grad_norm": 0.5486238598823547,
      "learning_rate": 9.104484018099288e-06,
      "loss": 0.6451,
      "step": 2979
    },
    {
      "epoch": 3.8115849267941435,
      "grad_norm": 0.4996105134487152,
      "learning_rate": 9.094084328672732e-06,
      "loss": 0.6629,
      "step": 2980
    },
    {
      "epoch": 3.812865029202336,
      "grad_norm": 0.5543266534805298,
      "learning_rate": 9.083693497576813e-06,
      "loss": 0.6762,
      "step": 2981
    },
    {
      "epoch": 3.8141451316105286,
      "grad_norm": 0.5271042585372925,
      "learning_rate": 9.073311531943569e-06,
      "loss": 0.6123,
      "step": 2982
    },
    {
      "epoch": 3.8154252340187216,
      "grad_norm": 0.4737612307071686,
      "learning_rate": 9.062938438898936e-06,
      "loss": 0.6359,
      "step": 2983
    },
    {
      "epoch": 3.816705336426914,
      "grad_norm": 0.4501494765281677,
      "learning_rate": 9.052574225562778e-06,
      "loss": 0.6531,
      "step": 2984
    },
    {
      "epoch": 3.8179854388351067,
      "grad_norm": 0.5052728652954102,
      "learning_rate": 9.042218899048865e-06,
      "loss": 0.6501,
      "step": 2985
    },
    {
      "epoch": 3.8192655412432996,
      "grad_norm": 0.5092394948005676,
      "learning_rate": 9.031872466464853e-06,
      "loss": 0.6631,
      "step": 2986
    },
    {
      "epoch": 3.820545643651492,
      "grad_norm": 0.48900294303894043,
      "learning_rate": 9.0215349349123e-06,
      "loss": 0.6786,
      "step": 2987
    },
    {
      "epoch": 3.8218257460596847,
      "grad_norm": 0.47884610295295715,
      "learning_rate": 9.011206311486662e-06,
      "loss": 0.6174,
      "step": 2988
    },
    {
      "epoch": 3.8231058484678773,
      "grad_norm": 0.4919913113117218,
      "learning_rate": 9.000886603277266e-06,
      "loss": 0.6386,
      "step": 2989
    },
    {
      "epoch": 3.8243859508760703,
      "grad_norm": 0.4870697855949402,
      "learning_rate": 8.990575817367331e-06,
      "loss": 0.6561,
      "step": 2990
    },
    {
      "epoch": 3.825666053284263,
      "grad_norm": 0.5059792995452881,
      "learning_rate": 8.980273960833955e-06,
      "loss": 0.6541,
      "step": 2991
    },
    {
      "epoch": 3.8269461556924553,
      "grad_norm": 0.5328920483589172,
      "learning_rate": 8.969981040748094e-06,
      "loss": 0.654,
      "step": 2992
    },
    {
      "epoch": 3.8282262581006483,
      "grad_norm": 0.4907168447971344,
      "learning_rate": 8.959697064174575e-06,
      "loss": 0.6495,
      "step": 2993
    },
    {
      "epoch": 3.829506360508841,
      "grad_norm": 0.48205500841140747,
      "learning_rate": 8.9494220381721e-06,
      "loss": 0.6013,
      "step": 2994
    },
    {
      "epoch": 3.8307864629170334,
      "grad_norm": 0.5104532241821289,
      "learning_rate": 8.939155969793205e-06,
      "loss": 0.6487,
      "step": 2995
    },
    {
      "epoch": 3.832066565325226,
      "grad_norm": 0.5226468443870544,
      "learning_rate": 8.92889886608429e-06,
      "loss": 0.6818,
      "step": 2996
    },
    {
      "epoch": 3.8333466677334185,
      "grad_norm": 0.5273947715759277,
      "learning_rate": 8.918650734085608e-06,
      "loss": 0.6172,
      "step": 2997
    },
    {
      "epoch": 3.8346267701416115,
      "grad_norm": 0.5069656372070312,
      "learning_rate": 8.908411580831239e-06,
      "loss": 0.6025,
      "step": 2998
    },
    {
      "epoch": 3.835906872549804,
      "grad_norm": 0.455133318901062,
      "learning_rate": 8.898181413349107e-06,
      "loss": 0.646,
      "step": 2999
    },
    {
      "epoch": 3.8371869749579965,
      "grad_norm": 0.5026136636734009,
      "learning_rate": 8.88796023866098e-06,
      "loss": 0.6332,
      "step": 3000
    },
    {
      "epoch": 3.8371869749579965,
      "eval_loss": 0.011393608525395393,
      "eval_runtime": 10.8083,
      "eval_samples_per_second": 45.891,
      "eval_steps_per_second": 5.736,
      "step": 3000
    },
    {
      "epoch": 3.8384670773661895,
      "grad_norm": 0.5468612313270569,
      "learning_rate": 8.877748063782429e-06,
      "loss": 0.6251,
      "step": 3001
    },
    {
      "epoch": 3.839747179774382,
      "grad_norm": 0.5341756939888,
      "learning_rate": 8.867544895722869e-06,
      "loss": 0.6503,
      "step": 3002
    },
    {
      "epoch": 3.8410272821825746,
      "grad_norm": 0.5142664909362793,
      "learning_rate": 8.857350741485526e-06,
      "loss": 0.6446,
      "step": 3003
    },
    {
      "epoch": 3.842307384590767,
      "grad_norm": 0.5296292304992676,
      "learning_rate": 8.847165608067436e-06,
      "loss": 0.6579,
      "step": 3004
    },
    {
      "epoch": 3.8435874869989597,
      "grad_norm": 0.5008630156517029,
      "learning_rate": 8.83698950245945e-06,
      "loss": 0.6564,
      "step": 3005
    },
    {
      "epoch": 3.8448675894071527,
      "grad_norm": 0.45849448442459106,
      "learning_rate": 8.826822431646208e-06,
      "loss": 0.6251,
      "step": 3006
    },
    {
      "epoch": 3.846147691815345,
      "grad_norm": 0.5119452476501465,
      "learning_rate": 8.816664402606167e-06,
      "loss": 0.6651,
      "step": 3007
    },
    {
      "epoch": 3.8474277942235378,
      "grad_norm": 0.5033029913902283,
      "learning_rate": 8.806515422311575e-06,
      "loss": 0.6202,
      "step": 3008
    },
    {
      "epoch": 3.8487078966317307,
      "grad_norm": 0.5177422165870667,
      "learning_rate": 8.796375497728451e-06,
      "loss": 0.6688,
      "step": 3009
    },
    {
      "epoch": 3.8499879990399233,
      "grad_norm": 0.47897839546203613,
      "learning_rate": 8.78624463581662e-06,
      "loss": 0.6356,
      "step": 3010
    },
    {
      "epoch": 3.851268101448116,
      "grad_norm": 0.5281705260276794,
      "learning_rate": 8.776122843529679e-06,
      "loss": 0.631,
      "step": 3011
    },
    {
      "epoch": 3.8525482038563084,
      "grad_norm": 0.473541259765625,
      "learning_rate": 8.766010127814994e-06,
      "loss": 0.6312,
      "step": 3012
    },
    {
      "epoch": 3.853828306264501,
      "grad_norm": 0.49050384759902954,
      "learning_rate": 8.755906495613706e-06,
      "loss": 0.6659,
      "step": 3013
    },
    {
      "epoch": 3.855108408672694,
      "grad_norm": 0.5245078206062317,
      "learning_rate": 8.74581195386073e-06,
      "loss": 0.6498,
      "step": 3014
    },
    {
      "epoch": 3.8563885110808864,
      "grad_norm": 0.513077437877655,
      "learning_rate": 8.735726509484722e-06,
      "loss": 0.6424,
      "step": 3015
    },
    {
      "epoch": 3.857668613489079,
      "grad_norm": 0.49324536323547363,
      "learning_rate": 8.725650169408106e-06,
      "loss": 0.6318,
      "step": 3016
    },
    {
      "epoch": 3.858948715897272,
      "grad_norm": 0.4788437485694885,
      "learning_rate": 8.715582940547068e-06,
      "loss": 0.6248,
      "step": 3017
    },
    {
      "epoch": 3.8602288183054645,
      "grad_norm": 0.5154752731323242,
      "learning_rate": 8.705524829811508e-06,
      "loss": 0.636,
      "step": 3018
    },
    {
      "epoch": 3.861508920713657,
      "grad_norm": 0.49946969747543335,
      "learning_rate": 8.695475844105098e-06,
      "loss": 0.6666,
      "step": 3019
    },
    {
      "epoch": 3.8627890231218496,
      "grad_norm": 0.4704820215702057,
      "learning_rate": 8.68543599032524e-06,
      "loss": 0.6062,
      "step": 3020
    },
    {
      "epoch": 3.8640691255300426,
      "grad_norm": 0.4827554225921631,
      "learning_rate": 8.675405275363058e-06,
      "loss": 0.6504,
      "step": 3021
    },
    {
      "epoch": 3.865349227938235,
      "grad_norm": 0.4924499988555908,
      "learning_rate": 8.665383706103408e-06,
      "loss": 0.645,
      "step": 3022
    },
    {
      "epoch": 3.8666293303464276,
      "grad_norm": 0.4820888340473175,
      "learning_rate": 8.655371289424879e-06,
      "loss": 0.6391,
      "step": 3023
    },
    {
      "epoch": 3.8679094327546206,
      "grad_norm": 0.4984312057495117,
      "learning_rate": 8.645368032199757e-06,
      "loss": 0.6693,
      "step": 3024
    },
    {
      "epoch": 3.869189535162813,
      "grad_norm": 0.49851006269454956,
      "learning_rate": 8.63537394129406e-06,
      "loss": 0.6516,
      "step": 3025
    },
    {
      "epoch": 3.8704696375710057,
      "grad_norm": 0.4811323881149292,
      "learning_rate": 8.62538902356751e-06,
      "loss": 0.637,
      "step": 3026
    },
    {
      "epoch": 3.8717497399791982,
      "grad_norm": 0.4522285759449005,
      "learning_rate": 8.615413285873523e-06,
      "loss": 0.6116,
      "step": 3027
    },
    {
      "epoch": 3.873029842387391,
      "grad_norm": 0.47147703170776367,
      "learning_rate": 8.605446735059223e-06,
      "loss": 0.6483,
      "step": 3028
    },
    {
      "epoch": 3.8743099447955838,
      "grad_norm": 0.4844609797000885,
      "learning_rate": 8.595489377965438e-06,
      "loss": 0.6573,
      "step": 3029
    },
    {
      "epoch": 3.8755900472037763,
      "grad_norm": 0.4950932562351227,
      "learning_rate": 8.585541221426657e-06,
      "loss": 0.6221,
      "step": 3030
    },
    {
      "epoch": 3.8755900472037763,
      "eval_loss": 0.011392003856599331,
      "eval_runtime": 10.8302,
      "eval_samples_per_second": 45.798,
      "eval_steps_per_second": 5.725,
      "step": 3030
    },
    {
      "epoch": 3.876870149611969,
      "grad_norm": 0.4878495931625366,
      "learning_rate": 8.575602272271077e-06,
      "loss": 0.6601,
      "step": 3031
    },
    {
      "epoch": 3.878150252020162,
      "grad_norm": 0.5103570222854614,
      "learning_rate": 8.565672537320581e-06,
      "loss": 0.6724,
      "step": 3032
    },
    {
      "epoch": 3.8794303544283544,
      "grad_norm": 0.4957294464111328,
      "learning_rate": 8.5557520233907e-06,
      "loss": 0.6779,
      "step": 3033
    },
    {
      "epoch": 3.880710456836547,
      "grad_norm": 0.4894900321960449,
      "learning_rate": 8.545840737290655e-06,
      "loss": 0.6367,
      "step": 3034
    },
    {
      "epoch": 3.8819905592447395,
      "grad_norm": 0.4775257706642151,
      "learning_rate": 8.535938685823344e-06,
      "loss": 0.655,
      "step": 3035
    },
    {
      "epoch": 3.883270661652932,
      "grad_norm": 0.478449285030365,
      "learning_rate": 8.526045875785293e-06,
      "loss": 0.6233,
      "step": 3036
    },
    {
      "epoch": 3.884550764061125,
      "grad_norm": 0.5237434506416321,
      "learning_rate": 8.516162313966717e-06,
      "loss": 0.6654,
      "step": 3037
    },
    {
      "epoch": 3.8858308664693175,
      "grad_norm": 0.4853358864784241,
      "learning_rate": 8.506288007151473e-06,
      "loss": 0.6332,
      "step": 3038
    },
    {
      "epoch": 3.88711096887751,
      "grad_norm": 0.4554169774055481,
      "learning_rate": 8.496422962117051e-06,
      "loss": 0.6329,
      "step": 3039
    },
    {
      "epoch": 3.888391071285703,
      "grad_norm": 0.4704521894454956,
      "learning_rate": 8.486567185634606e-06,
      "loss": 0.6481,
      "step": 3040
    },
    {
      "epoch": 3.8896711736938956,
      "grad_norm": 0.4720918834209442,
      "learning_rate": 8.476720684468921e-06,
      "loss": 0.6281,
      "step": 3041
    },
    {
      "epoch": 3.890951276102088,
      "grad_norm": 0.4820510149002075,
      "learning_rate": 8.466883465378411e-06,
      "loss": 0.6298,
      "step": 3042
    },
    {
      "epoch": 3.8922313785102807,
      "grad_norm": 0.49557843804359436,
      "learning_rate": 8.45705553511512e-06,
      "loss": 0.6662,
      "step": 3043
    },
    {
      "epoch": 3.893511480918473,
      "grad_norm": 0.525687575340271,
      "learning_rate": 8.44723690042472e-06,
      "loss": 0.6491,
      "step": 3044
    },
    {
      "epoch": 3.894791583326666,
      "grad_norm": 0.4745238125324249,
      "learning_rate": 8.43742756804651e-06,
      "loss": 0.5982,
      "step": 3045
    },
    {
      "epoch": 3.8960716857348587,
      "grad_norm": 0.48005083203315735,
      "learning_rate": 8.427627544713381e-06,
      "loss": 0.6445,
      "step": 3046
    },
    {
      "epoch": 3.8973517881430517,
      "grad_norm": 0.4923861622810364,
      "learning_rate": 8.417836837151857e-06,
      "loss": 0.6565,
      "step": 3047
    },
    {
      "epoch": 3.8986318905512443,
      "grad_norm": 0.48761045932769775,
      "learning_rate": 8.408055452082064e-06,
      "loss": 0.6536,
      "step": 3048
    },
    {
      "epoch": 3.899911992959437,
      "grad_norm": 0.5227696895599365,
      "learning_rate": 8.398283396217712e-06,
      "loss": 0.6227,
      "step": 3049
    },
    {
      "epoch": 3.9011920953676293,
      "grad_norm": 0.4789896309375763,
      "learning_rate": 8.388520676266132e-06,
      "loss": 0.6195,
      "step": 3050
    },
    {
      "epoch": 3.902472197775822,
      "grad_norm": 0.4864027202129364,
      "learning_rate": 8.378767298928237e-06,
      "loss": 0.6092,
      "step": 3051
    },
    {
      "epoch": 3.903752300184015,
      "grad_norm": 0.47270238399505615,
      "learning_rate": 8.369023270898515e-06,
      "loss": 0.6627,
      "step": 3052
    },
    {
      "epoch": 3.9050324025922074,
      "grad_norm": 0.47152993083000183,
      "learning_rate": 8.359288598865052e-06,
      "loss": 0.6263,
      "step": 3053
    },
    {
      "epoch": 3.9063125050004,
      "grad_norm": 0.46962761878967285,
      "learning_rate": 8.349563289509516e-06,
      "loss": 0.5927,
      "step": 3054
    },
    {
      "epoch": 3.907592607408593,
      "grad_norm": 0.4662138521671295,
      "learning_rate": 8.339847349507132e-06,
      "loss": 0.6271,
      "step": 3055
    },
    {
      "epoch": 3.9088727098167855,
      "grad_norm": 0.4927668273448944,
      "learning_rate": 8.330140785526706e-06,
      "loss": 0.6389,
      "step": 3056
    },
    {
      "epoch": 3.910152812224978,
      "grad_norm": 0.45494070649147034,
      "learning_rate": 8.32044360423061e-06,
      "loss": 0.6574,
      "step": 3057
    },
    {
      "epoch": 3.9114329146331706,
      "grad_norm": 0.5050803422927856,
      "learning_rate": 8.310755812274764e-06,
      "loss": 0.6438,
      "step": 3058
    },
    {
      "epoch": 3.912713017041363,
      "grad_norm": 0.5328724980354309,
      "learning_rate": 8.301077416308651e-06,
      "loss": 0.6236,
      "step": 3059
    },
    {
      "epoch": 3.913993119449556,
      "grad_norm": 0.4846177399158478,
      "learning_rate": 8.291408422975315e-06,
      "loss": 0.6257,
      "step": 3060
    },
    {
      "epoch": 3.913993119449556,
      "eval_loss": 0.011388003826141357,
      "eval_runtime": 10.8209,
      "eval_samples_per_second": 45.837,
      "eval_steps_per_second": 5.73,
      "step": 3060
    },
    {
      "epoch": 3.9152732218577486,
      "grad_norm": 0.48647886514663696,
      "learning_rate": 8.281748838911323e-06,
      "loss": 0.6474,
      "step": 3061
    },
    {
      "epoch": 3.916553324265941,
      "grad_norm": 0.4750177264213562,
      "learning_rate": 8.272098670746801e-06,
      "loss": 0.6538,
      "step": 3062
    },
    {
      "epoch": 3.917833426674134,
      "grad_norm": 0.49009081721305847,
      "learning_rate": 8.262457925105414e-06,
      "loss": 0.6126,
      "step": 3063
    },
    {
      "epoch": 3.9191135290823267,
      "grad_norm": 0.5038008689880371,
      "learning_rate": 8.252826608604343e-06,
      "loss": 0.6666,
      "step": 3064
    },
    {
      "epoch": 3.9203936314905192,
      "grad_norm": 0.49638527631759644,
      "learning_rate": 8.24320472785431e-06,
      "loss": 0.6465,
      "step": 3065
    },
    {
      "epoch": 3.9216737338987118,
      "grad_norm": 0.4825122654438019,
      "learning_rate": 8.23359228945957e-06,
      "loss": 0.6486,
      "step": 3066
    },
    {
      "epoch": 3.9229538363069043,
      "grad_norm": 0.5027803778648376,
      "learning_rate": 8.223989300017867e-06,
      "loss": 0.6504,
      "step": 3067
    },
    {
      "epoch": 3.9242339387150973,
      "grad_norm": 0.49459660053253174,
      "learning_rate": 8.214395766120484e-06,
      "loss": 0.6478,
      "step": 3068
    },
    {
      "epoch": 3.92551404112329,
      "grad_norm": 0.4860994815826416,
      "learning_rate": 8.204811694352216e-06,
      "loss": 0.6398,
      "step": 3069
    },
    {
      "epoch": 3.9267941435314824,
      "grad_norm": 0.4880172312259674,
      "learning_rate": 8.19523709129134e-06,
      "loss": 0.6631,
      "step": 3070
    },
    {
      "epoch": 3.9280742459396754,
      "grad_norm": 0.4944741427898407,
      "learning_rate": 8.185671963509655e-06,
      "loss": 0.6414,
      "step": 3071
    },
    {
      "epoch": 3.929354348347868,
      "grad_norm": 0.5079156160354614,
      "learning_rate": 8.176116317572457e-06,
      "loss": 0.6452,
      "step": 3072
    },
    {
      "epoch": 3.9306344507560604,
      "grad_norm": 0.4638914465904236,
      "learning_rate": 8.166570160038514e-06,
      "loss": 0.643,
      "step": 3073
    },
    {
      "epoch": 3.931914553164253,
      "grad_norm": 0.4641512334346771,
      "learning_rate": 8.1570334974601e-06,
      "loss": 0.6451,
      "step": 3074
    },
    {
      "epoch": 3.9331946555724455,
      "grad_norm": 0.4876812696456909,
      "learning_rate": 8.147506336382973e-06,
      "loss": 0.6115,
      "step": 3075
    },
    {
      "epoch": 3.9344747579806385,
      "grad_norm": 0.49157753586769104,
      "learning_rate": 8.137988683346347e-06,
      "loss": 0.6451,
      "step": 3076
    },
    {
      "epoch": 3.935754860388831,
      "grad_norm": 0.4958047568798065,
      "learning_rate": 8.128480544882934e-06,
      "loss": 0.6406,
      "step": 3077
    },
    {
      "epoch": 3.937034962797024,
      "grad_norm": 0.4504125714302063,
      "learning_rate": 8.118981927518915e-06,
      "loss": 0.6036,
      "step": 3078
    },
    {
      "epoch": 3.9383150652052166,
      "grad_norm": 0.509773313999176,
      "learning_rate": 8.109492837773912e-06,
      "loss": 0.6729,
      "step": 3079
    },
    {
      "epoch": 3.939595167613409,
      "grad_norm": 0.4869062900543213,
      "learning_rate": 8.10001328216103e-06,
      "loss": 0.6204,
      "step": 3080
    },
    {
      "epoch": 3.9408752700216017,
      "grad_norm": 0.5071339011192322,
      "learning_rate": 8.090543267186829e-06,
      "loss": 0.6344,
      "step": 3081
    },
    {
      "epoch": 3.942155372429794,
      "grad_norm": 0.5110708475112915,
      "learning_rate": 8.081082799351301e-06,
      "loss": 0.6936,
      "step": 3082
    },
    {
      "epoch": 3.943435474837987,
      "grad_norm": 0.4968855381011963,
      "learning_rate": 8.071631885147914e-06,
      "loss": 0.6476,
      "step": 3083
    },
    {
      "epoch": 3.9447155772461797,
      "grad_norm": 0.47699102759361267,
      "learning_rate": 8.062190531063549e-06,
      "loss": 0.6432,
      "step": 3084
    },
    {
      "epoch": 3.9459956796543723,
      "grad_norm": 0.4785839319229126,
      "learning_rate": 8.052758743578547e-06,
      "loss": 0.6552,
      "step": 3085
    },
    {
      "epoch": 3.9472757820625652,
      "grad_norm": 0.4930052161216736,
      "learning_rate": 8.043336529166681e-06,
      "loss": 0.7104,
      "step": 3086
    },
    {
      "epoch": 3.948555884470758,
      "grad_norm": 0.46638068556785583,
      "learning_rate": 8.033923894295132e-06,
      "loss": 0.6452,
      "step": 3087
    },
    {
      "epoch": 3.9498359868789503,
      "grad_norm": 0.47538697719573975,
      "learning_rate": 8.024520845424532e-06,
      "loss": 0.6318,
      "step": 3088
    },
    {
      "epoch": 3.951116089287143,
      "grad_norm": 0.5044569373130798,
      "learning_rate": 8.015127389008923e-06,
      "loss": 0.6442,
      "step": 3089
    },
    {
      "epoch": 3.9523961916953354,
      "grad_norm": 0.4961550533771515,
      "learning_rate": 8.005743531495756e-06,
      "loss": 0.6402,
      "step": 3090
    },
    {
      "epoch": 3.9523961916953354,
      "eval_loss": 0.011380051262676716,
      "eval_runtime": 10.826,
      "eval_samples_per_second": 45.816,
      "eval_steps_per_second": 5.727,
      "step": 3090
    },
    {
      "epoch": 3.9536762941035284,
      "grad_norm": 0.47219613194465637,
      "learning_rate": 7.996369279325901e-06,
      "loss": 0.6121,
      "step": 3091
    },
    {
      "epoch": 3.954956396511721,
      "grad_norm": 0.4593760371208191,
      "learning_rate": 7.987004638933646e-06,
      "loss": 0.617,
      "step": 3092
    },
    {
      "epoch": 3.9562364989199135,
      "grad_norm": 0.4869404733181,
      "learning_rate": 7.977649616746653e-06,
      "loss": 0.6563,
      "step": 3093
    },
    {
      "epoch": 3.9575166013281065,
      "grad_norm": 0.4596230685710907,
      "learning_rate": 7.968304219186006e-06,
      "loss": 0.6202,
      "step": 3094
    },
    {
      "epoch": 3.958796703736299,
      "grad_norm": 0.49334338307380676,
      "learning_rate": 7.958968452666184e-06,
      "loss": 0.6464,
      "step": 3095
    },
    {
      "epoch": 3.9600768061444915,
      "grad_norm": 0.48544076085090637,
      "learning_rate": 7.949642323595035e-06,
      "loss": 0.6398,
      "step": 3096
    },
    {
      "epoch": 3.961356908552684,
      "grad_norm": 0.4898148775100708,
      "learning_rate": 7.940325838373808e-06,
      "loss": 0.6404,
      "step": 3097
    },
    {
      "epoch": 3.9626370109608766,
      "grad_norm": 0.4826902151107788,
      "learning_rate": 7.931019003397135e-06,
      "loss": 0.6437,
      "step": 3098
    },
    {
      "epoch": 3.9639171133690696,
      "grad_norm": 0.5142403244972229,
      "learning_rate": 7.921721825053013e-06,
      "loss": 0.6522,
      "step": 3099
    },
    {
      "epoch": 3.965197215777262,
      "grad_norm": 0.49967753887176514,
      "learning_rate": 7.912434309722815e-06,
      "loss": 0.6611,
      "step": 3100
    },
    {
      "epoch": 3.9664773181854547,
      "grad_norm": 0.4844744801521301,
      "learning_rate": 7.90315646378129e-06,
      "loss": 0.6419,
      "step": 3101
    },
    {
      "epoch": 3.9677574205936477,
      "grad_norm": 0.46609875559806824,
      "learning_rate": 7.893888293596534e-06,
      "loss": 0.6159,
      "step": 3102
    },
    {
      "epoch": 3.96903752300184,
      "grad_norm": 0.4745890498161316,
      "learning_rate": 7.884629805530017e-06,
      "loss": 0.6336,
      "step": 3103
    },
    {
      "epoch": 3.9703176254100327,
      "grad_norm": 0.49275532364845276,
      "learning_rate": 7.87538100593656e-06,
      "loss": 0.6114,
      "step": 3104
    },
    {
      "epoch": 3.9715977278182253,
      "grad_norm": 0.4750758111476898,
      "learning_rate": 7.866141901164324e-06,
      "loss": 0.6301,
      "step": 3105
    },
    {
      "epoch": 3.9728778302264183,
      "grad_norm": 0.49187296628952026,
      "learning_rate": 7.856912497554821e-06,
      "loss": 0.6313,
      "step": 3106
    },
    {
      "epoch": 3.974157932634611,
      "grad_norm": 0.5150437355041504,
      "learning_rate": 7.847692801442921e-06,
      "loss": 0.6512,
      "step": 3107
    },
    {
      "epoch": 3.9754380350428034,
      "grad_norm": 0.4669630229473114,
      "learning_rate": 7.8384828191568e-06,
      "loss": 0.6256,
      "step": 3108
    },
    {
      "epoch": 3.9767181374509963,
      "grad_norm": 0.46803414821624756,
      "learning_rate": 7.829282557017994e-06,
      "loss": 0.6541,
      "step": 3109
    },
    {
      "epoch": 3.977998239859189,
      "grad_norm": 0.4534625709056854,
      "learning_rate": 7.82009202134135e-06,
      "loss": 0.6428,
      "step": 3110
    },
    {
      "epoch": 3.9792783422673814,
      "grad_norm": 0.5128123760223389,
      "learning_rate": 7.810911218435044e-06,
      "loss": 0.673,
      "step": 3111
    },
    {
      "epoch": 3.980558444675574,
      "grad_norm": 0.5048739314079285,
      "learning_rate": 7.801740154600584e-06,
      "loss": 0.6655,
      "step": 3112
    },
    {
      "epoch": 3.9818385470837665,
      "grad_norm": 0.4649309515953064,
      "learning_rate": 7.792578836132767e-06,
      "loss": 0.6362,
      "step": 3113
    },
    {
      "epoch": 3.9831186494919595,
      "grad_norm": 0.4508923888206482,
      "learning_rate": 7.783427269319726e-06,
      "loss": 0.6256,
      "step": 3114
    },
    {
      "epoch": 3.984398751900152,
      "grad_norm": 0.48068177700042725,
      "learning_rate": 7.77428546044289e-06,
      "loss": 0.687,
      "step": 3115
    },
    {
      "epoch": 3.9856788543083446,
      "grad_norm": 0.471524178981781,
      "learning_rate": 7.765153415776987e-06,
      "loss": 0.6514,
      "step": 3116
    },
    {
      "epoch": 3.9869589567165375,
      "grad_norm": 0.4320808947086334,
      "learning_rate": 7.756031141590051e-06,
      "loss": 0.626,
      "step": 3117
    },
    {
      "epoch": 3.98823905912473,
      "grad_norm": 0.494357168674469,
      "learning_rate": 7.74691864414341e-06,
      "loss": 0.6736,
      "step": 3118
    },
    {
      "epoch": 3.9895191615329226,
      "grad_norm": 0.49010246992111206,
      "learning_rate": 7.737815929691665e-06,
      "loss": 0.6847,
      "step": 3119
    },
    {
      "epoch": 3.990799263941115,
      "grad_norm": 0.47675013542175293,
      "learning_rate": 7.728723004482726e-06,
      "loss": 0.6426,
      "step": 3120
    },
    {
      "epoch": 3.990799263941115,
      "eval_loss": 0.011370080523192883,
      "eval_runtime": 10.8414,
      "eval_samples_per_second": 45.751,
      "eval_steps_per_second": 5.719,
      "step": 3120
    },
    {
      "epoch": 3.9920793663493077,
      "grad_norm": 0.4744720757007599,
      "learning_rate": 7.71963987475777e-06,
      "loss": 0.6755,
      "step": 3121
    },
    {
      "epoch": 3.9933594687575007,
      "grad_norm": 0.48414307832717896,
      "learning_rate": 7.71056654675125e-06,
      "loss": 0.6348,
      "step": 3122
    },
    {
      "epoch": 3.9946395711656932,
      "grad_norm": 0.48627740144729614,
      "learning_rate": 7.701503026690898e-06,
      "loss": 0.6638,
      "step": 3123
    },
    {
      "epoch": 3.9959196735738858,
      "grad_norm": 0.4748673141002655,
      "learning_rate": 7.692449320797715e-06,
      "loss": 0.6536,
      "step": 3124
    },
    {
      "epoch": 3.9971997759820788,
      "grad_norm": 0.5028969049453735,
      "learning_rate": 7.68340543528595e-06,
      "loss": 0.6614,
      "step": 3125
    },
    {
      "epoch": 3.9984798783902713,
      "grad_norm": 0.5049706101417542,
      "learning_rate": 7.674371376363128e-06,
      "loss": 0.6705,
      "step": 3126
    },
    {
      "epoch": 3.999759980798464,
      "grad_norm": 0.5031850337982178,
      "learning_rate": 7.665347150230029e-06,
      "loss": 0.6389,
      "step": 3127
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.9680603742599487,
      "learning_rate": 7.656332763080672e-06,
      "loss": 0.6112,
      "step": 3128
    }
  ],
  "logging_steps": 1,
  "max_steps": 3910,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 100,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 4.3101892528528425e+18,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}