{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.989914662529092,
  "eval_steps": 21,
  "global_step": 1288,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0031031807602792862,
      "grad_norm": 7.78125,
      "learning_rate": 1.0000000000000002e-06,
      "loss": 1.7457,
      "step": 1
    },
    {
      "epoch": 0.0031031807602792862,
      "eval_loss": 2.851605176925659,
      "eval_runtime": 36.0554,
      "eval_samples_per_second": 136.124,
      "eval_steps_per_second": 13.618,
      "step": 1
    },
    {
      "epoch": 0.0062063615205585725,
      "grad_norm": 8.8125,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 1.8753,
      "step": 2
    },
    {
      "epoch": 0.009309542280837859,
      "grad_norm": 8.3125,
      "learning_rate": 3e-06,
      "loss": 1.7573,
      "step": 3
    },
    {
      "epoch": 0.012412723041117145,
      "grad_norm": 7.46875,
      "learning_rate": 4.000000000000001e-06,
      "loss": 1.6633,
      "step": 4
    },
    {
      "epoch": 0.015515903801396431,
      "grad_norm": 7.25,
      "learning_rate": 5e-06,
      "loss": 1.7447,
      "step": 5
    },
    {
      "epoch": 0.018619084561675717,
      "grad_norm": 5.03125,
      "learning_rate": 6e-06,
      "loss": 1.6933,
      "step": 6
    },
    {
      "epoch": 0.021722265321955005,
      "grad_norm": 4.75,
      "learning_rate": 7e-06,
      "loss": 1.6926,
      "step": 7
    },
    {
      "epoch": 0.02482544608223429,
      "grad_norm": 4.59375,
      "learning_rate": 8.000000000000001e-06,
      "loss": 1.624,
      "step": 8
    },
    {
      "epoch": 0.027928626842513578,
      "grad_norm": 4.96875,
      "learning_rate": 9e-06,
      "loss": 1.6107,
      "step": 9
    },
    {
      "epoch": 0.031031807602792862,
      "grad_norm": 9.375,
      "learning_rate": 1e-05,
      "loss": 1.497,
      "step": 10
    },
    {
      "epoch": 0.03413498836307215,
      "grad_norm": 4.3125,
      "learning_rate": 9.999984893013967e-06,
      "loss": 1.4678,
      "step": 11
    },
    {
      "epoch": 0.037238169123351435,
      "grad_norm": 3.65625,
      "learning_rate": 9.999939572147153e-06,
      "loss": 1.4111,
      "step": 12
    },
    {
      "epoch": 0.04034134988363072,
      "grad_norm": 3.0625,
      "learning_rate": 9.999864037673424e-06,
      "loss": 1.4662,
      "step": 13
    },
    {
      "epoch": 0.04344453064391001,
      "grad_norm": 3.09375,
      "learning_rate": 9.99975829004922e-06,
      "loss": 1.4718,
      "step": 14
    },
    {
      "epoch": 0.04654771140418929,
      "grad_norm": 3.015625,
      "learning_rate": 9.99962232991355e-06,
      "loss": 1.4397,
      "step": 15
    },
    {
      "epoch": 0.04965089216446858,
      "grad_norm": 2.875,
      "learning_rate": 9.999456158087994e-06,
      "loss": 1.3946,
      "step": 16
    },
    {
      "epoch": 0.05275407292474787,
      "grad_norm": 2.640625,
      "learning_rate": 9.999259775576696e-06,
      "loss": 1.4171,
      "step": 17
    },
    {
      "epoch": 0.055857253685027156,
      "grad_norm": 2.609375,
      "learning_rate": 9.999033183566353e-06,
      "loss": 1.3742,
      "step": 18
    },
    {
      "epoch": 0.05896043444530644,
      "grad_norm": 2.84375,
      "learning_rate": 9.998776383426217e-06,
      "loss": 1.4403,
      "step": 19
    },
    {
      "epoch": 0.062063615205585725,
      "grad_norm": 2.546875,
      "learning_rate": 9.998489376708073e-06,
      "loss": 1.3268,
      "step": 20
    },
    {
      "epoch": 0.06516679596586501,
      "grad_norm": 2.8125,
      "learning_rate": 9.998172165146248e-06,
      "loss": 1.3743,
      "step": 21
    },
    {
      "epoch": 0.06516679596586501,
      "eval_loss": 2.691330909729004,
      "eval_runtime": 239.485,
      "eval_samples_per_second": 20.494,
      "eval_steps_per_second": 2.05,
      "step": 21
    },
    {
      "epoch": 0.0682699767261443,
      "grad_norm": 2.71875,
      "learning_rate": 9.997824750657586e-06,
      "loss": 1.3792,
      "step": 22
    },
    {
      "epoch": 0.07137315748642359,
      "grad_norm": 2.65625,
      "learning_rate": 9.997447135341439e-06,
      "loss": 1.3903,
      "step": 23
    },
    {
      "epoch": 0.07447633824670287,
      "grad_norm": 2.5,
      "learning_rate": 9.99703932147966e-06,
      "loss": 1.3214,
      "step": 24
    },
    {
      "epoch": 0.07757951900698215,
      "grad_norm": 2.453125,
      "learning_rate": 9.996601311536586e-06,
      "loss": 1.3832,
      "step": 25
    },
    {
      "epoch": 0.08068269976726145,
      "grad_norm": 2.59375,
      "learning_rate": 9.996133108159019e-06,
      "loss": 1.3616,
      "step": 26
    },
    {
      "epoch": 0.08378588052754073,
      "grad_norm": 2.453125,
      "learning_rate": 9.995634714176215e-06,
      "loss": 1.3388,
      "step": 27
    },
    {
      "epoch": 0.08688906128782002,
      "grad_norm": 2.34375,
      "learning_rate": 9.995106132599869e-06,
      "loss": 1.3497,
      "step": 28
    },
    {
      "epoch": 0.0899922420480993,
      "grad_norm": 2.40625,
      "learning_rate": 9.994547366624089e-06,
      "loss": 1.3252,
      "step": 29
    },
    {
      "epoch": 0.09309542280837858,
      "grad_norm": 2.5,
      "learning_rate": 9.993958419625382e-06,
      "loss": 1.3557,
      "step": 30
    },
    {
      "epoch": 0.09619860356865788,
      "grad_norm": 2.4375,
      "learning_rate": 9.993339295162637e-06,
      "loss": 1.3278,
      "step": 31
    },
    {
      "epoch": 0.09930178432893716,
      "grad_norm": 2.296875,
      "learning_rate": 9.992689996977091e-06,
      "loss": 1.2994,
      "step": 32
    },
    {
      "epoch": 0.10240496508921644,
      "grad_norm": 2.453125,
      "learning_rate": 9.992010528992326e-06,
      "loss": 1.3511,
      "step": 33
    },
    {
      "epoch": 0.10550814584949574,
      "grad_norm": 2.609375,
      "learning_rate": 9.99130089531422e-06,
      "loss": 1.3588,
      "step": 34
    },
    {
      "epoch": 0.10861132660977502,
      "grad_norm": 2.359375,
      "learning_rate": 9.990561100230949e-06,
      "loss": 1.3294,
      "step": 35
    },
    {
      "epoch": 0.11171450737005431,
      "grad_norm": 2.609375,
      "learning_rate": 9.98979114821294e-06,
      "loss": 1.3197,
      "step": 36
    },
    {
      "epoch": 0.11481768813033359,
      "grad_norm": 2.515625,
      "learning_rate": 9.988991043912857e-06,
      "loss": 1.3121,
      "step": 37
    },
    {
      "epoch": 0.11792086889061287,
      "grad_norm": 2.28125,
      "learning_rate": 9.988160792165562e-06,
      "loss": 1.2625,
      "step": 38
    },
    {
      "epoch": 0.12102404965089217,
      "grad_norm": 2.53125,
      "learning_rate": 9.987300397988098e-06,
      "loss": 1.3433,
      "step": 39
    },
    {
      "epoch": 0.12412723041117145,
      "grad_norm": 2.53125,
      "learning_rate": 9.98640986657965e-06,
      "loss": 1.3424,
      "step": 40
    },
    {
      "epoch": 0.12723041117145073,
      "grad_norm": 2.484375,
      "learning_rate": 9.985489203321517e-06,
      "loss": 1.4157,
      "step": 41
    },
    {
      "epoch": 0.13033359193173003,
      "grad_norm": 2.234375,
      "learning_rate": 9.984538413777077e-06,
      "loss": 1.3049,
      "step": 42
    },
    {
      "epoch": 0.13033359193173003,
      "eval_loss": 2.2133007049560547,
      "eval_runtime": 35.872,
      "eval_samples_per_second": 136.82,
      "eval_steps_per_second": 13.688,
      "step": 42
    },
    {
      "epoch": 0.13343677269200932,
      "grad_norm": 2.375,
      "learning_rate": 9.983557503691755e-06,
      "loss": 1.3171,
      "step": 43
    },
    {
      "epoch": 0.1365399534522886,
      "grad_norm": 2.359375,
      "learning_rate": 9.982546478992989e-06,
      "loss": 1.2672,
      "step": 44
    },
    {
      "epoch": 0.13964313421256788,
      "grad_norm": 2.265625,
      "learning_rate": 9.981505345790194e-06,
      "loss": 1.3015,
      "step": 45
    },
    {
      "epoch": 0.14274631497284718,
      "grad_norm": 2.421875,
      "learning_rate": 9.980434110374725e-06,
      "loss": 1.3395,
      "step": 46
    },
    {
      "epoch": 0.14584949573312644,
      "grad_norm": 2.375,
      "learning_rate": 9.979332779219835e-06,
      "loss": 1.3385,
      "step": 47
    },
    {
      "epoch": 0.14895267649340574,
      "grad_norm": 2.421875,
      "learning_rate": 9.978201358980645e-06,
      "loss": 1.3163,
      "step": 48
    },
    {
      "epoch": 0.15205585725368503,
      "grad_norm": 2.28125,
      "learning_rate": 9.977039856494091e-06,
      "loss": 1.2862,
      "step": 49
    },
    {
      "epoch": 0.1551590380139643,
      "grad_norm": 2.28125,
      "learning_rate": 9.975848278778897e-06,
      "loss": 1.252,
      "step": 50
    },
    {
      "epoch": 0.1582622187742436,
      "grad_norm": 2.40625,
      "learning_rate": 9.97462663303552e-06,
      "loss": 1.3021,
      "step": 51
    },
    {
      "epoch": 0.1613653995345229,
      "grad_norm": 2.421875,
      "learning_rate": 9.973374926646117e-06,
      "loss": 1.3471,
      "step": 52
    },
    {
      "epoch": 0.16446858029480219,
      "grad_norm": 2.453125,
      "learning_rate": 9.972093167174488e-06,
      "loss": 1.3728,
      "step": 53
    },
    {
      "epoch": 0.16757176105508145,
      "grad_norm": 2.359375,
      "learning_rate": 9.970781362366046e-06,
      "loss": 1.324,
      "step": 54
    },
    {
      "epoch": 0.17067494181536075,
      "grad_norm": 2.28125,
      "learning_rate": 9.969439520147754e-06,
      "loss": 1.2654,
      "step": 55
    },
    {
      "epoch": 0.17377812257564004,
      "grad_norm": 2.484375,
      "learning_rate": 9.968067648628092e-06,
      "loss": 1.3198,
      "step": 56
    },
    {
      "epoch": 0.1768813033359193,
      "grad_norm": 2.359375,
      "learning_rate": 9.966665756096996e-06,
      "loss": 1.3077,
      "step": 57
    },
    {
      "epoch": 0.1799844840961986,
      "grad_norm": 2.28125,
      "learning_rate": 9.965233851025816e-06,
      "loss": 1.2822,
      "step": 58
    },
    {
      "epoch": 0.1830876648564779,
      "grad_norm": 2.328125,
      "learning_rate": 9.963771942067257e-06,
      "loss": 1.2697,
      "step": 59
    },
    {
      "epoch": 0.18619084561675717,
      "grad_norm": 2.390625,
      "learning_rate": 9.962280038055337e-06,
      "loss": 1.2677,
      "step": 60
    },
    {
      "epoch": 0.18929402637703646,
      "grad_norm": 2.359375,
      "learning_rate": 9.960758148005323e-06,
      "loss": 1.2904,
      "step": 61
    },
    {
      "epoch": 0.19239720713731576,
      "grad_norm": 2.625,
      "learning_rate": 9.959206281113685e-06,
      "loss": 1.268,
      "step": 62
    },
    {
      "epoch": 0.19550038789759502,
      "grad_norm": 2.453125,
      "learning_rate": 9.957624446758034e-06,
      "loss": 1.2851,
      "step": 63
    },
    {
      "epoch": 0.19550038789759502,
      "eval_loss": 2.2618393898010254,
      "eval_runtime": 35.8504,
      "eval_samples_per_second": 136.902,
      "eval_steps_per_second": 13.696,
      "step": 63
    },
    {
      "epoch": 0.19860356865787432,
      "grad_norm": 2.265625,
      "learning_rate": 9.956012654497073e-06,
      "loss": 1.3279,
      "step": 64
    },
    {
      "epoch": 0.2017067494181536,
      "grad_norm": 2.421875,
      "learning_rate": 9.95437091407053e-06,
      "loss": 1.3197,
      "step": 65
    },
    {
      "epoch": 0.20480993017843288,
      "grad_norm": 2.328125,
      "learning_rate": 9.952699235399104e-06,
      "loss": 1.3258,
      "step": 66
    },
    {
      "epoch": 0.20791311093871218,
      "grad_norm": 2.40625,
      "learning_rate": 9.950997628584404e-06,
      "loss": 1.2951,
      "step": 67
    },
    {
      "epoch": 0.21101629169899147,
      "grad_norm": 2.5625,
      "learning_rate": 9.949266103908894e-06,
      "loss": 1.3182,
      "step": 68
    },
    {
      "epoch": 0.21411947245927077,
      "grad_norm": 2.359375,
      "learning_rate": 9.947504671835819e-06,
      "loss": 1.3408,
      "step": 69
    },
    {
      "epoch": 0.21722265321955003,
      "grad_norm": 2.234375,
      "learning_rate": 9.945713343009154e-06,
      "loss": 1.3018,
      "step": 70
    },
    {
      "epoch": 0.22032583397982933,
      "grad_norm": 2.375,
      "learning_rate": 9.943892128253525e-06,
      "loss": 1.2745,
      "step": 71
    },
    {
      "epoch": 0.22342901474010862,
      "grad_norm": 2.453125,
      "learning_rate": 9.942041038574163e-06,
      "loss": 1.2899,
      "step": 72
    },
    {
      "epoch": 0.2265321955003879,
      "grad_norm": 2.390625,
      "learning_rate": 9.94016008515682e-06,
      "loss": 1.2288,
      "step": 73
    },
    {
      "epoch": 0.22963537626066718,
      "grad_norm": 2.265625,
      "learning_rate": 9.938249279367714e-06,
      "loss": 1.2715,
      "step": 74
    },
    {
      "epoch": 0.23273855702094648,
      "grad_norm": 2.421875,
      "learning_rate": 9.93630863275345e-06,
      "loss": 1.2608,
      "step": 75
    },
    {
      "epoch": 0.23584173778122575,
      "grad_norm": 2.328125,
      "learning_rate": 9.934338157040953e-06,
      "loss": 1.2675,
      "step": 76
    },
    {
      "epoch": 0.23894491854150504,
      "grad_norm": 2.359375,
      "learning_rate": 9.932337864137406e-06,
      "loss": 1.2886,
      "step": 77
    },
    {
      "epoch": 0.24204809930178434,
      "grad_norm": 2.25,
      "learning_rate": 9.930307766130169e-06,
      "loss": 1.3359,
      "step": 78
    },
    {
      "epoch": 0.2451512800620636,
      "grad_norm": 2.640625,
      "learning_rate": 9.928247875286705e-06,
      "loss": 1.2853,
      "step": 79
    },
    {
      "epoch": 0.2482544608223429,
      "grad_norm": 2.578125,
      "learning_rate": 9.926158204054511e-06,
      "loss": 1.273,
      "step": 80
    },
    {
      "epoch": 0.2513576415826222,
      "grad_norm": 2.421875,
      "learning_rate": 9.924038765061042e-06,
      "loss": 1.1882,
      "step": 81
    },
    {
      "epoch": 0.25446082234290146,
      "grad_norm": 2.3125,
      "learning_rate": 9.921889571113629e-06,
      "loss": 1.2491,
      "step": 82
    },
    {
      "epoch": 0.2575640031031808,
      "grad_norm": 2.390625,
      "learning_rate": 9.91971063519941e-06,
      "loss": 1.3084,
      "step": 83
    },
    {
      "epoch": 0.26066718386346005,
      "grad_norm": 2.375,
      "learning_rate": 9.917501970485252e-06,
      "loss": 1.3514,
      "step": 84
    },
    {
      "epoch": 0.26066718386346005,
      "eval_loss": 2.268841505050659,
      "eval_runtime": 35.7551,
      "eval_samples_per_second": 137.267,
      "eval_steps_per_second": 13.732,
      "step": 84
    },
    {
      "epoch": 0.2637703646237393,
      "grad_norm": 2.3125,
      "learning_rate": 9.915263590317654e-06,
      "loss": 1.2524,
      "step": 85
    },
    {
      "epoch": 0.26687354538401864,
      "grad_norm": 2.484375,
      "learning_rate": 9.912995508222693e-06,
      "loss": 1.227,
      "step": 86
    },
    {
      "epoch": 0.2699767261442979,
      "grad_norm": 2.453125,
      "learning_rate": 9.91069773790592e-06,
      "loss": 1.2823,
      "step": 87
    },
    {
      "epoch": 0.2730799069045772,
      "grad_norm": 2.34375,
      "learning_rate": 9.90837029325229e-06,
      "loss": 1.3209,
      "step": 88
    },
    {
      "epoch": 0.2761830876648565,
      "grad_norm": 2.359375,
      "learning_rate": 9.90601318832607e-06,
      "loss": 1.3291,
      "step": 89
    },
    {
      "epoch": 0.27928626842513576,
      "grad_norm": 2.375,
      "learning_rate": 9.903626437370764e-06,
      "loss": 1.2755,
      "step": 90
    },
    {
      "epoch": 0.28238944918541503,
      "grad_norm": 2.453125,
      "learning_rate": 9.901210054809015e-06,
      "loss": 1.2405,
      "step": 91
    },
    {
      "epoch": 0.28549262994569435,
      "grad_norm": 2.265625,
      "learning_rate": 9.898764055242527e-06,
      "loss": 1.2523,
      "step": 92
    },
    {
      "epoch": 0.2885958107059736,
      "grad_norm": 2.421875,
      "learning_rate": 9.896288453451972e-06,
      "loss": 1.2488,
      "step": 93
    },
    {
      "epoch": 0.2916989914662529,
      "grad_norm": 2.34375,
      "learning_rate": 9.893783264396903e-06,
      "loss": 1.2587,
      "step": 94
    },
    {
      "epoch": 0.2948021722265322,
      "grad_norm": 2.53125,
      "learning_rate": 9.891248503215663e-06,
      "loss": 1.244,
      "step": 95
    },
    {
      "epoch": 0.2979053529868115,
      "grad_norm": 2.328125,
      "learning_rate": 9.888684185225291e-06,
      "loss": 1.2304,
      "step": 96
    },
    {
      "epoch": 0.30100853374709075,
      "grad_norm": 2.296875,
      "learning_rate": 9.886090325921435e-06,
      "loss": 1.2188,
      "step": 97
    },
    {
      "epoch": 0.30411171450737007,
      "grad_norm": 2.296875,
      "learning_rate": 9.883466940978252e-06,
      "loss": 1.2724,
      "step": 98
    },
    {
      "epoch": 0.30721489526764933,
      "grad_norm": 2.484375,
      "learning_rate": 9.88081404624832e-06,
      "loss": 1.2822,
      "step": 99
    },
    {
      "epoch": 0.3103180760279286,
      "grad_norm": 2.515625,
      "learning_rate": 9.878131657762535e-06,
      "loss": 1.305,
      "step": 100
    },
    {
      "epoch": 0.3134212567882079,
      "grad_norm": 2.359375,
      "learning_rate": 9.87541979173002e-06,
      "loss": 1.298,
      "step": 101
    },
    {
      "epoch": 0.3165244375484872,
      "grad_norm": 2.34375,
      "learning_rate": 9.872678464538021e-06,
      "loss": 1.2702,
      "step": 102
    },
    {
      "epoch": 0.31962761830876646,
      "grad_norm": 2.421875,
      "learning_rate": 9.86990769275182e-06,
      "loss": 1.2658,
      "step": 103
    },
    {
      "epoch": 0.3227307990690458,
      "grad_norm": 2.515625,
      "learning_rate": 9.867107493114618e-06,
      "loss": 1.2853,
      "step": 104
    },
    {
      "epoch": 0.32583397982932505,
      "grad_norm": 2.484375,
      "learning_rate": 9.864277882547443e-06,
      "loss": 1.2409,
      "step": 105
    },
    {
      "epoch": 0.32583397982932505,
      "eval_loss": 2.1522727012634277,
      "eval_runtime": 35.8062,
      "eval_samples_per_second": 137.071,
      "eval_steps_per_second": 13.713,
      "step": 105
    },
    {
      "epoch": 0.32893716058960437,
      "grad_norm": 2.421875,
      "learning_rate": 9.861418878149056e-06,
      "loss": 1.2621,
      "step": 106
    },
    {
      "epoch": 0.33204034134988364,
      "grad_norm": 2.515625,
      "learning_rate": 9.858530497195829e-06,
      "loss": 1.2893,
      "step": 107
    },
    {
      "epoch": 0.3351435221101629,
      "grad_norm": 2.515625,
      "learning_rate": 9.855612757141654e-06,
      "loss": 1.3063,
      "step": 108
    },
    {
      "epoch": 0.33824670287044223,
      "grad_norm": 2.328125,
      "learning_rate": 9.852665675617837e-06,
      "loss": 1.2593,
      "step": 109
    },
    {
      "epoch": 0.3413498836307215,
      "grad_norm": 2.265625,
      "learning_rate": 9.849689270432983e-06,
      "loss": 1.278,
      "step": 110
    },
    {
      "epoch": 0.34445306439100076,
      "grad_norm": 2.328125,
      "learning_rate": 9.846683559572898e-06,
      "loss": 1.2546,
      "step": 111
    },
    {
      "epoch": 0.3475562451512801,
      "grad_norm": 2.234375,
      "learning_rate": 9.843648561200476e-06,
      "loss": 1.2909,
      "step": 112
    },
    {
      "epoch": 0.35065942591155935,
      "grad_norm": 2.453125,
      "learning_rate": 9.840584293655584e-06,
      "loss": 1.2801,
      "step": 113
    },
    {
      "epoch": 0.3537626066718386,
      "grad_norm": 2.28125,
      "learning_rate": 9.837490775454965e-06,
      "loss": 1.2458,
      "step": 114
    },
    {
      "epoch": 0.35686578743211794,
      "grad_norm": 2.375,
      "learning_rate": 9.834368025292112e-06,
      "loss": 1.314,
      "step": 115
    },
    {
      "epoch": 0.3599689681923972,
      "grad_norm": 2.453125,
      "learning_rate": 9.831216062037163e-06,
      "loss": 1.2959,
      "step": 116
    },
    {
      "epoch": 0.3630721489526765,
      "grad_norm": 2.296875,
      "learning_rate": 9.828034904736783e-06,
      "loss": 1.2588,
      "step": 117
    },
    {
      "epoch": 0.3661753297129558,
      "grad_norm": 2.265625,
      "learning_rate": 9.82482457261405e-06,
      "loss": 1.2992,
      "step": 118
    },
    {
      "epoch": 0.36927851047323507,
      "grad_norm": 2.375,
      "learning_rate": 9.821585085068346e-06,
      "loss": 1.3095,
      "step": 119
    },
    {
      "epoch": 0.37238169123351433,
      "grad_norm": 2.375,
      "learning_rate": 9.81831646167522e-06,
      "loss": 1.2778,
      "step": 120
    },
    {
      "epoch": 0.37548487199379366,
      "grad_norm": 2.40625,
      "learning_rate": 9.8150187221863e-06,
      "loss": 1.2471,
      "step": 121
    },
    {
      "epoch": 0.3785880527540729,
      "grad_norm": 2.390625,
      "learning_rate": 9.811691886529145e-06,
      "loss": 1.2716,
      "step": 122
    },
    {
      "epoch": 0.3816912335143522,
      "grad_norm": 2.328125,
      "learning_rate": 9.808335974807135e-06,
      "loss": 1.2895,
      "step": 123
    },
    {
      "epoch": 0.3847944142746315,
      "grad_norm": 2.4375,
      "learning_rate": 9.80495100729936e-06,
      "loss": 1.2803,
      "step": 124
    },
    {
      "epoch": 0.3878975950349108,
      "grad_norm": 2.390625,
      "learning_rate": 9.801537004460478e-06,
      "loss": 1.2804,
      "step": 125
    },
    {
      "epoch": 0.39100077579519005,
      "grad_norm": 2.3125,
      "learning_rate": 9.798093986920609e-06,
      "loss": 1.2723,
      "step": 126
    },
    {
      "epoch": 0.39100077579519005,
      "eval_loss": 2.1467559337615967,
      "eval_runtime": 35.7061,
      "eval_samples_per_second": 137.455,
      "eval_steps_per_second": 13.751,
      "step": 126
    },
    {
      "epoch": 0.39410395655546937,
      "grad_norm": 2.46875,
      "learning_rate": 9.7946219754852e-06,
      "loss": 1.2288,
      "step": 127
    },
    {
      "epoch": 0.39720713731574864,
      "grad_norm": 2.265625,
      "learning_rate": 9.791120991134903e-06,
      "loss": 1.2629,
      "step": 128
    },
    {
      "epoch": 0.4003103180760279,
      "grad_norm": 2.46875,
      "learning_rate": 9.787591055025445e-06,
      "loss": 1.236,
      "step": 129
    },
    {
      "epoch": 0.4034134988363072,
      "grad_norm": 2.265625,
      "learning_rate": 9.784032188487507e-06,
      "loss": 1.2822,
      "step": 130
    },
    {
      "epoch": 0.4065166795965865,
      "grad_norm": 2.265625,
      "learning_rate": 9.780444413026581e-06,
      "loss": 1.2273,
      "step": 131
    },
    {
      "epoch": 0.40961986035686576,
      "grad_norm": 2.25,
      "learning_rate": 9.776827750322865e-06,
      "loss": 1.2815,
      "step": 132
    },
    {
      "epoch": 0.4127230411171451,
      "grad_norm": 2.390625,
      "learning_rate": 9.773182222231104e-06,
      "loss": 1.2404,
      "step": 133
    },
    {
      "epoch": 0.41582622187742435,
      "grad_norm": 2.3125,
      "learning_rate": 9.769507850780474e-06,
      "loss": 1.279,
      "step": 134
    },
    {
      "epoch": 0.4189294026377037,
      "grad_norm": 2.546875,
      "learning_rate": 9.76580465817445e-06,
      "loss": 1.2547,
      "step": 135
    },
    {
      "epoch": 0.42203258339798294,
      "grad_norm": 2.375,
      "learning_rate": 9.762072666790658e-06,
      "loss": 1.2009,
      "step": 136
    },
    {
      "epoch": 0.4251357641582622,
      "grad_norm": 2.3125,
      "learning_rate": 9.758311899180759e-06,
      "loss": 1.2987,
      "step": 137
    },
    {
      "epoch": 0.42823894491854153,
      "grad_norm": 2.171875,
      "learning_rate": 9.754522378070296e-06,
      "loss": 1.2658,
      "step": 138
    },
    {
      "epoch": 0.4313421256788208,
      "grad_norm": 2.21875,
      "learning_rate": 9.750704126358568e-06,
      "loss": 1.2039,
      "step": 139
    },
    {
      "epoch": 0.43444530643910007,
      "grad_norm": 2.375,
      "learning_rate": 9.746857167118486e-06,
      "loss": 1.3077,
      "step": 140
    },
    {
      "epoch": 0.4375484871993794,
      "grad_norm": 2.859375,
      "learning_rate": 9.742981523596431e-06,
      "loss": 1.3127,
      "step": 141
    },
    {
      "epoch": 0.44065166795965865,
      "grad_norm": 2.3125,
      "learning_rate": 9.73907721921212e-06,
      "loss": 1.2383,
      "step": 142
    },
    {
      "epoch": 0.4437548487199379,
      "grad_norm": 2.34375,
      "learning_rate": 9.735144277558466e-06,
      "loss": 1.2219,
      "step": 143
    },
    {
      "epoch": 0.44685802948021724,
      "grad_norm": 2.453125,
      "learning_rate": 9.73118272240142e-06,
      "loss": 1.2442,
      "step": 144
    },
    {
      "epoch": 0.4499612102404965,
      "grad_norm": 2.234375,
      "learning_rate": 9.727192577679852e-06,
      "loss": 1.2864,
      "step": 145
    },
    {
      "epoch": 0.4530643910007758,
      "grad_norm": 2.40625,
      "learning_rate": 9.723173867505383e-06,
      "loss": 1.2431,
      "step": 146
    },
    {
      "epoch": 0.4561675717610551,
      "grad_norm": 2.390625,
      "learning_rate": 9.719126616162252e-06,
      "loss": 1.25,
      "step": 147
    },
    {
      "epoch": 0.4561675717610551,
      "eval_loss": 2.0751290321350098,
      "eval_runtime": 35.8312,
      "eval_samples_per_second": 136.976,
      "eval_steps_per_second": 13.703,
      "step": 147
    },
    {
      "epoch": 0.45927075252133437,
      "grad_norm": 2.375,
      "learning_rate": 9.715050848107167e-06,
      "loss": 1.2228,
      "step": 148
    },
    {
      "epoch": 0.46237393328161364,
      "grad_norm": 2.390625,
      "learning_rate": 9.710946587969158e-06,
      "loss": 1.2413,
      "step": 149
    },
    {
      "epoch": 0.46547711404189296,
      "grad_norm": 2.4375,
      "learning_rate": 9.706813860549425e-06,
      "loss": 1.2347,
      "step": 150
    },
    {
      "epoch": 0.4685802948021722,
      "grad_norm": 2.359375,
      "learning_rate": 9.702652690821188e-06,
      "loss": 1.2742,
      "step": 151
    },
    {
      "epoch": 0.4716834755624515,
      "grad_norm": 2.21875,
      "learning_rate": 9.698463103929542e-06,
      "loss": 1.2525,
      "step": 152
    },
    {
      "epoch": 0.4747866563227308,
      "grad_norm": 2.796875,
      "learning_rate": 9.6942451251913e-06,
      "loss": 1.2944,
      "step": 153
    },
    {
      "epoch": 0.4778898370830101,
      "grad_norm": 2.421875,
      "learning_rate": 9.689998780094839e-06,
      "loss": 1.2756,
      "step": 154
    },
    {
      "epoch": 0.48099301784328935,
      "grad_norm": 2.390625,
      "learning_rate": 9.68572409429995e-06,
      "loss": 1.2598,
      "step": 155
    },
    {
      "epoch": 0.4840961986035687,
      "grad_norm": 2.21875,
      "learning_rate": 9.681421093637677e-06,
      "loss": 1.2741,
      "step": 156
    },
    {
      "epoch": 0.48719937936384794,
      "grad_norm": 2.46875,
      "learning_rate": 9.677089804110173e-06,
      "loss": 1.2372,
      "step": 157
    },
    {
      "epoch": 0.4903025601241272,
      "grad_norm": 2.40625,
      "learning_rate": 9.67273025189053e-06,
      "loss": 1.2393,
      "step": 158
    },
    {
      "epoch": 0.49340574088440653,
      "grad_norm": 2.453125,
      "learning_rate": 9.668342463322624e-06,
      "loss": 1.2892,
      "step": 159
    },
    {
      "epoch": 0.4965089216446858,
      "grad_norm": 2.34375,
      "learning_rate": 9.663926464920959e-06,
      "loss": 1.2142,
      "step": 160
    },
    {
      "epoch": 0.49961210240496506,
      "grad_norm": 2.546875,
      "learning_rate": 9.659482283370506e-06,
      "loss": 1.2512,
      "step": 161
    },
    {
      "epoch": 0.5027152831652444,
      "grad_norm": 2.453125,
      "learning_rate": 9.655009945526541e-06,
      "loss": 1.2725,
      "step": 162
    },
    {
      "epoch": 0.5058184639255237,
      "grad_norm": 2.375,
      "learning_rate": 9.650509478414483e-06,
      "loss": 1.2361,
      "step": 163
    },
    {
      "epoch": 0.5089216446858029,
      "grad_norm": 2.3125,
      "learning_rate": 9.645980909229727e-06,
      "loss": 1.2686,
      "step": 164
    },
    {
      "epoch": 0.5120248254460822,
      "grad_norm": 2.4375,
      "learning_rate": 9.641424265337485e-06,
      "loss": 1.2444,
      "step": 165
    },
    {
      "epoch": 0.5151280062063616,
      "grad_norm": 2.46875,
      "learning_rate": 9.636839574272623e-06,
      "loss": 1.2604,
      "step": 166
    },
    {
      "epoch": 0.5182311869666408,
      "grad_norm": 2.28125,
      "learning_rate": 9.632226863739485e-06,
      "loss": 1.2956,
      "step": 167
    },
    {
      "epoch": 0.5213343677269201,
      "grad_norm": 2.265625,
      "learning_rate": 9.627586161611732e-06,
      "loss": 1.2143,
      "step": 168
    },
    {
      "epoch": 0.5213343677269201,
      "eval_loss": 2.070341110229492,
      "eval_runtime": 35.8101,
      "eval_samples_per_second": 137.056,
      "eval_steps_per_second": 13.711,
      "step": 168
    },
    {
      "epoch": 0.5244375484871994,
      "grad_norm": 2.65625,
      "learning_rate": 9.622917495932172e-06,
      "loss": 1.2881,
      "step": 169
    },
    {
      "epoch": 0.5275407292474786,
      "grad_norm": 2.453125,
      "learning_rate": 9.618220894912593e-06,
      "loss": 1.2674,
      "step": 170
    },
    {
      "epoch": 0.5306439100077579,
      "grad_norm": 2.5625,
      "learning_rate": 9.613496386933591e-06,
      "loss": 1.3045,
      "step": 171
    },
    {
      "epoch": 0.5337470907680373,
      "grad_norm": 2.5,
      "learning_rate": 9.608744000544392e-06,
      "loss": 1.2742,
      "step": 172
    },
    {
      "epoch": 0.5368502715283165,
      "grad_norm": 2.484375,
      "learning_rate": 9.603963764462695e-06,
      "loss": 1.3176,
      "step": 173
    },
    {
      "epoch": 0.5399534522885958,
      "grad_norm": 2.390625,
      "learning_rate": 9.59915570757448e-06,
      "loss": 1.302,
      "step": 174
    },
    {
      "epoch": 0.5430566330488751,
      "grad_norm": 2.296875,
      "learning_rate": 9.594319858933847e-06,
      "loss": 1.2315,
      "step": 175
    },
    {
      "epoch": 0.5461598138091543,
      "grad_norm": 2.4375,
      "learning_rate": 9.58945624776284e-06,
      "loss": 1.2806,
      "step": 176
    },
    {
      "epoch": 0.5492629945694336,
      "grad_norm": 2.453125,
      "learning_rate": 9.584564903451253e-06,
      "loss": 1.2485,
      "step": 177
    },
    {
      "epoch": 0.552366175329713,
      "grad_norm": 2.21875,
      "learning_rate": 9.579645855556481e-06,
      "loss": 1.2175,
      "step": 178
    },
    {
      "epoch": 0.5554693560899923,
      "grad_norm": 2.421875,
      "learning_rate": 9.574699133803315e-06,
      "loss": 1.2744,
      "step": 179
    },
    {
      "epoch": 0.5585725368502715,
      "grad_norm": 2.46875,
      "learning_rate": 9.569724768083781e-06,
      "loss": 1.2646,
      "step": 180
    },
    {
      "epoch": 0.5616757176105508,
      "grad_norm": 2.390625,
      "learning_rate": 9.564722788456943e-06,
      "loss": 1.2648,
      "step": 181
    },
    {
      "epoch": 0.5647788983708301,
      "grad_norm": 2.3125,
      "learning_rate": 9.559693225148744e-06,
      "loss": 1.2274,
      "step": 182
    },
    {
      "epoch": 0.5678820791311094,
      "grad_norm": 2.375,
      "learning_rate": 9.554636108551792e-06,
      "loss": 1.2932,
      "step": 183
    },
    {
      "epoch": 0.5709852598913887,
      "grad_norm": 2.484375,
      "learning_rate": 9.54955146922521e-06,
      "loss": 1.2684,
      "step": 184
    },
    {
      "epoch": 0.574088440651668,
      "grad_norm": 2.359375,
      "learning_rate": 9.544439337894422e-06,
      "loss": 1.2395,
      "step": 185
    },
    {
      "epoch": 0.5771916214119472,
      "grad_norm": 2.484375,
      "learning_rate": 9.539299745450992e-06,
      "loss": 1.2807,
      "step": 186
    },
    {
      "epoch": 0.5802948021722265,
      "grad_norm": 2.34375,
      "learning_rate": 9.534132722952419e-06,
      "loss": 1.2879,
      "step": 187
    },
    {
      "epoch": 0.5833979829325058,
      "grad_norm": 2.40625,
      "learning_rate": 9.528938301621956e-06,
      "loss": 1.2628,
      "step": 188
    },
    {
      "epoch": 0.5865011636927852,
      "grad_norm": 2.484375,
      "learning_rate": 9.523716512848424e-06,
      "loss": 1.2632,
      "step": 189
    },
    {
      "epoch": 0.5865011636927852,
      "eval_loss": 1.8877675533294678,
      "eval_runtime": 35.7718,
      "eval_samples_per_second": 137.203,
      "eval_steps_per_second": 13.726,
      "step": 189
    },
    {
      "epoch": 0.5896043444530644,
      "grad_norm": 2.375,
      "learning_rate": 9.51846738818602e-06,
      "loss": 1.213,
      "step": 190
    },
    {
      "epoch": 0.5927075252133437,
      "grad_norm": 2.484375,
      "learning_rate": 9.513190959354124e-06,
      "loss": 1.2596,
      "step": 191
    },
    {
      "epoch": 0.595810705973623,
      "grad_norm": 2.484375,
      "learning_rate": 9.507887258237112e-06,
      "loss": 1.206,
      "step": 192
    },
    {
      "epoch": 0.5989138867339022,
      "grad_norm": 2.8125,
      "learning_rate": 9.502556316884157e-06,
      "loss": 1.3051,
      "step": 193
    },
    {
      "epoch": 0.6020170674941815,
      "grad_norm": 2.390625,
      "learning_rate": 9.497198167509045e-06,
      "loss": 1.2638,
      "step": 194
    },
    {
      "epoch": 0.6051202482544609,
      "grad_norm": 2.4375,
      "learning_rate": 9.491812842489966e-06,
      "loss": 1.236,
      "step": 195
    },
    {
      "epoch": 0.6082234290147401,
      "grad_norm": 2.515625,
      "learning_rate": 9.48640037436934e-06,
      "loss": 1.2731,
      "step": 196
    },
    {
      "epoch": 0.6113266097750194,
      "grad_norm": 2.53125,
      "learning_rate": 9.48096079585359e-06,
      "loss": 1.2507,
      "step": 197
    },
    {
      "epoch": 0.6144297905352987,
      "grad_norm": 2.40625,
      "learning_rate": 9.475494139812979e-06,
      "loss": 1.2068,
      "step": 198
    },
    {
      "epoch": 0.6175329712955779,
      "grad_norm": 2.578125,
      "learning_rate": 9.470000439281379e-06,
      "loss": 1.2335,
      "step": 199
    },
    {
      "epoch": 0.6206361520558572,
      "grad_norm": 2.15625,
      "learning_rate": 9.464479727456098e-06,
      "loss": 1.2209,
      "step": 200
    },
    {
      "epoch": 0.6237393328161366,
      "grad_norm": 2.453125,
      "learning_rate": 9.458932037697656e-06,
      "loss": 1.217,
      "step": 201
    },
    {
      "epoch": 0.6268425135764158,
      "grad_norm": 2.21875,
      "learning_rate": 9.453357403529609e-06,
      "loss": 1.2496,
      "step": 202
    },
    {
      "epoch": 0.6299456943366951,
      "grad_norm": 2.46875,
      "learning_rate": 9.447755858638322e-06,
      "loss": 1.227,
      "step": 203
    },
    {
      "epoch": 0.6330488750969744,
      "grad_norm": 2.484375,
      "learning_rate": 9.442127436872778e-06,
      "loss": 1.2088,
      "step": 204
    },
    {
      "epoch": 0.6361520558572537,
      "grad_norm": 2.375,
      "learning_rate": 9.436472172244374e-06,
      "loss": 1.2512,
      "step": 205
    },
    {
      "epoch": 0.6392552366175329,
      "grad_norm": 2.265625,
      "learning_rate": 9.430790098926711e-06,
      "loss": 1.2492,
      "step": 206
    },
    {
      "epoch": 0.6423584173778123,
      "grad_norm": 2.46875,
      "learning_rate": 9.425081251255392e-06,
      "loss": 1.2613,
      "step": 207
    },
    {
      "epoch": 0.6454615981380916,
      "grad_norm": 2.375,
      "learning_rate": 9.419345663727805e-06,
      "loss": 1.1954,
      "step": 208
    },
    {
      "epoch": 0.6485647788983708,
      "grad_norm": 2.40625,
      "learning_rate": 9.413583371002932e-06,
      "loss": 1.2822,
      "step": 209
    },
    {
      "epoch": 0.6516679596586501,
      "grad_norm": 2.40625,
      "learning_rate": 9.40779440790112e-06,
      "loss": 1.2867,
      "step": 210
    },
    {
      "epoch": 0.6516679596586501,
      "eval_loss": 1.755298376083374,
      "eval_runtime": 35.7858,
      "eval_samples_per_second": 137.149,
      "eval_steps_per_second": 13.721,
      "step": 210
    },
    {
      "epoch": 0.6547711404189294,
      "grad_norm": 2.578125,
      "learning_rate": 9.401978809403882e-06,
      "loss": 1.2947,
      "step": 211
    },
    {
      "epoch": 0.6578743211792087,
      "grad_norm": 2.375,
      "learning_rate": 9.396136610653689e-06,
      "loss": 1.2056,
      "step": 212
    },
    {
      "epoch": 0.660977501939488,
      "grad_norm": 2.375,
      "learning_rate": 9.39026784695374e-06,
      "loss": 1.2299,
      "step": 213
    },
    {
      "epoch": 0.6640806826997673,
      "grad_norm": 2.5,
      "learning_rate": 9.38437255376777e-06,
      "loss": 1.2635,
      "step": 214
    },
    {
      "epoch": 0.6671838634600465,
      "grad_norm": 2.515625,
      "learning_rate": 9.37845076671983e-06,
      "loss": 1.2054,
      "step": 215
    },
    {
      "epoch": 0.6702870442203258,
      "grad_norm": 2.46875,
      "learning_rate": 9.372502521594052e-06,
      "loss": 1.2301,
      "step": 216
    },
    {
      "epoch": 0.6733902249806051,
      "grad_norm": 2.546875,
      "learning_rate": 9.366527854334464e-06,
      "loss": 1.2889,
      "step": 217
    },
    {
      "epoch": 0.6764934057408845,
      "grad_norm": 2.375,
      "learning_rate": 9.360526801044751e-06,
      "loss": 1.2453,
      "step": 218
    },
    {
      "epoch": 0.6795965865011637,
      "grad_norm": 2.484375,
      "learning_rate": 9.354499397988045e-06,
      "loss": 1.1997,
      "step": 219
    },
    {
      "epoch": 0.682699767261443,
      "grad_norm": 2.46875,
      "learning_rate": 9.348445681586703e-06,
      "loss": 1.2404,
      "step": 220
    },
    {
      "epoch": 0.6858029480217223,
      "grad_norm": 2.421875,
      "learning_rate": 9.342365688422086e-06,
      "loss": 1.27,
      "step": 221
    },
    {
      "epoch": 0.6889061287820015,
      "grad_norm": 2.4375,
      "learning_rate": 9.336259455234346e-06,
      "loss": 1.252,
      "step": 222
    },
    {
      "epoch": 0.6920093095422808,
      "grad_norm": 2.578125,
      "learning_rate": 9.330127018922195e-06,
      "loss": 1.2425,
      "step": 223
    },
    {
      "epoch": 0.6951124903025602,
      "grad_norm": 2.40625,
      "learning_rate": 9.323968416542682e-06,
      "loss": 1.2549,
      "step": 224
    },
    {
      "epoch": 0.6982156710628394,
      "grad_norm": 2.328125,
      "learning_rate": 9.317783685310977e-06,
      "loss": 1.2956,
      "step": 225
    },
    {
      "epoch": 0.7013188518231187,
      "grad_norm": 2.359375,
      "learning_rate": 9.31157286260014e-06,
      "loss": 1.2372,
      "step": 226
    },
    {
      "epoch": 0.704422032583398,
      "grad_norm": 2.375,
      "learning_rate": 9.305335985940894e-06,
      "loss": 1.3153,
      "step": 227
    },
    {
      "epoch": 0.7075252133436772,
      "grad_norm": 2.3125,
      "learning_rate": 9.299073093021404e-06,
      "loss": 1.2691,
      "step": 228
    },
    {
      "epoch": 0.7106283941039565,
      "grad_norm": 2.265625,
      "learning_rate": 9.292784221687043e-06,
      "loss": 1.2731,
      "step": 229
    },
    {
      "epoch": 0.7137315748642359,
      "grad_norm": 2.46875,
      "learning_rate": 9.286469409940169e-06,
      "loss": 1.2605,
      "step": 230
    },
    {
      "epoch": 0.7168347556245152,
      "grad_norm": 2.296875,
      "learning_rate": 9.280128695939889e-06,
      "loss": 1.2607,
      "step": 231
    },
    {
      "epoch": 0.7168347556245152,
      "eval_loss": 1.8377017974853516,
      "eval_runtime": 35.7725,
      "eval_samples_per_second": 137.2,
      "eval_steps_per_second": 13.726,
      "step": 231
    },
    {
      "epoch": 0.7199379363847944,
      "grad_norm": 2.40625,
      "learning_rate": 9.273762118001837e-06,
      "loss": 1.2574,
      "step": 232
    },
    {
      "epoch": 0.7230411171450737,
      "grad_norm": 2.125,
      "learning_rate": 9.267369714597934e-06,
      "loss": 1.2362,
      "step": 233
    },
    {
      "epoch": 0.726144297905353,
      "grad_norm": 2.21875,
      "learning_rate": 9.260951524356157e-06,
      "loss": 1.1879,
      "step": 234
    },
    {
      "epoch": 0.7292474786656322,
      "grad_norm": 2.4375,
      "learning_rate": 9.25450758606031e-06,
      "loss": 1.2498,
      "step": 235
    },
    {
      "epoch": 0.7323506594259116,
      "grad_norm": 2.296875,
      "learning_rate": 9.248037938649792e-06,
      "loss": 1.2429,
      "step": 236
    },
    {
      "epoch": 0.7354538401861909,
      "grad_norm": 2.53125,
      "learning_rate": 9.24154262121935e-06,
      "loss": 1.2729,
      "step": 237
    },
    {
      "epoch": 0.7385570209464701,
      "grad_norm": 2.328125,
      "learning_rate": 9.235021673018849e-06,
      "loss": 1.269,
      "step": 238
    },
    {
      "epoch": 0.7416602017067494,
      "grad_norm": 2.375,
      "learning_rate": 9.228475133453042e-06,
      "loss": 1.1814,
      "step": 239
    },
    {
      "epoch": 0.7447633824670287,
      "grad_norm": 2.25,
      "learning_rate": 9.221903042081319e-06,
      "loss": 1.2961,
      "step": 240
    },
    {
      "epoch": 0.747866563227308,
      "grad_norm": 2.328125,
      "learning_rate": 9.21530543861748e-06,
      "loss": 1.2619,
      "step": 241
    },
    {
      "epoch": 0.7509697439875873,
      "grad_norm": 2.390625,
      "learning_rate": 9.208682362929484e-06,
      "loss": 1.1953,
      "step": 242
    },
    {
      "epoch": 0.7540729247478666,
      "grad_norm": 2.421875,
      "learning_rate": 9.202033855039216e-06,
      "loss": 1.2219,
      "step": 243
    },
    {
      "epoch": 0.7571761055081458,
      "grad_norm": 2.40625,
      "learning_rate": 9.195359955122244e-06,
      "loss": 1.2114,
      "step": 244
    },
    {
      "epoch": 0.7602792862684251,
      "grad_norm": 2.484375,
      "learning_rate": 9.188660703507571e-06,
      "loss": 1.2276,
      "step": 245
    },
    {
      "epoch": 0.7633824670287044,
      "grad_norm": 2.296875,
      "learning_rate": 9.181936140677398e-06,
      "loss": 1.2294,
      "step": 246
    },
    {
      "epoch": 0.7664856477889838,
      "grad_norm": 2.40625,
      "learning_rate": 9.175186307266877e-06,
      "loss": 1.2669,
      "step": 247
    },
    {
      "epoch": 0.769588828549263,
      "grad_norm": 2.34375,
      "learning_rate": 9.168411244063864e-06,
      "loss": 1.2606,
      "step": 248
    },
    {
      "epoch": 0.7726920093095423,
      "grad_norm": 2.4375,
      "learning_rate": 9.161610992008668e-06,
      "loss": 1.2278,
      "step": 249
    },
    {
      "epoch": 0.7757951900698216,
      "grad_norm": 2.390625,
      "learning_rate": 9.15478559219382e-06,
      "loss": 1.2851,
      "step": 250
    },
    {
      "epoch": 0.7788983708301008,
      "grad_norm": 2.375,
      "learning_rate": 9.147935085863804e-06,
      "loss": 1.2444,
      "step": 251
    },
    {
      "epoch": 0.7820015515903801,
      "grad_norm": 2.28125,
      "learning_rate": 9.141059514414826e-06,
      "loss": 1.2123,
      "step": 252
    },
    {
      "epoch": 0.7820015515903801,
      "eval_loss": 1.836862564086914,
      "eval_runtime": 35.6974,
      "eval_samples_per_second": 137.489,
      "eval_steps_per_second": 13.754,
      "step": 252
    },
    {
      "epoch": 0.7851047323506595,
      "grad_norm": 2.3125,
      "learning_rate": 9.134158919394545e-06,
      "loss": 1.2444,
      "step": 253
    },
    {
      "epoch": 0.7882079131109387,
      "grad_norm": 2.421875,
      "learning_rate": 9.127233342501841e-06,
      "loss": 1.313,
      "step": 254
    },
    {
      "epoch": 0.791311093871218,
      "grad_norm": 2.359375,
      "learning_rate": 9.120282825586551e-06,
      "loss": 1.2486,
      "step": 255
    },
    {
      "epoch": 0.7944142746314973,
      "grad_norm": 2.203125,
      "learning_rate": 9.113307410649222e-06,
      "loss": 1.2544,
      "step": 256
    },
    {
      "epoch": 0.7975174553917765,
      "grad_norm": 2.4375,
      "learning_rate": 9.106307139840849e-06,
      "loss": 1.2811,
      "step": 257
    },
    {
      "epoch": 0.8006206361520558,
      "grad_norm": 2.34375,
      "learning_rate": 9.09928205546263e-06,
      "loss": 1.2686,
      "step": 258
    },
    {
      "epoch": 0.8037238169123352,
      "grad_norm": 2.3125,
      "learning_rate": 9.092232199965707e-06,
      "loss": 1.244,
      "step": 259
    },
    {
      "epoch": 0.8068269976726145,
      "grad_norm": 2.34375,
      "learning_rate": 9.085157615950905e-06,
      "loss": 1.2055,
      "step": 260
    },
    {
      "epoch": 0.8099301784328937,
      "grad_norm": 2.390625,
      "learning_rate": 9.078058346168485e-06,
      "loss": 1.2244,
      "step": 261
    },
    {
      "epoch": 0.813033359193173,
      "grad_norm": 2.625,
      "learning_rate": 9.070934433517872e-06,
      "loss": 1.1869,
      "step": 262
    },
    {
      "epoch": 0.8161365399534523,
      "grad_norm": 2.546875,
      "learning_rate": 9.063785921047406e-06,
      "loss": 1.2589,
      "step": 263
    },
    {
      "epoch": 0.8192397207137315,
      "grad_norm": 2.3125,
      "learning_rate": 9.056612851954077e-06,
      "loss": 1.2352,
      "step": 264
    },
    {
      "epoch": 0.8223429014740109,
      "grad_norm": 2.3125,
      "learning_rate": 9.049415269583268e-06,
      "loss": 1.2561,
      "step": 265
    },
    {
      "epoch": 0.8254460822342902,
      "grad_norm": 2.1875,
      "learning_rate": 9.04219321742849e-06,
      "loss": 1.2023,
      "step": 266
    },
    {
      "epoch": 0.8285492629945694,
      "grad_norm": 2.265625,
      "learning_rate": 9.034946739131119e-06,
      "loss": 1.2348,
      "step": 267
    },
    {
      "epoch": 0.8316524437548487,
      "grad_norm": 2.34375,
      "learning_rate": 9.027675878480131e-06,
      "loss": 1.2443,
      "step": 268
    },
    {
      "epoch": 0.834755624515128,
      "grad_norm": 2.390625,
      "learning_rate": 9.020380679411847e-06,
      "loss": 1.2654,
      "step": 269
    },
    {
      "epoch": 0.8378588052754073,
      "grad_norm": 2.453125,
      "learning_rate": 9.013061186009653e-06,
      "loss": 1.2457,
      "step": 270
    },
    {
      "epoch": 0.8409619860356866,
      "grad_norm": 2.265625,
      "learning_rate": 9.005717442503741e-06,
      "loss": 1.243,
      "step": 271
    },
    {
      "epoch": 0.8440651667959659,
      "grad_norm": 2.3125,
      "learning_rate": 8.998349493270845e-06,
      "loss": 1.2656,
      "step": 272
    },
    {
      "epoch": 0.8471683475562451,
      "grad_norm": 2.515625,
      "learning_rate": 8.990957382833969e-06,
      "loss": 1.2509,
      "step": 273
    },
    {
      "epoch": 0.8471683475562451,
      "eval_loss": 1.8987137079238892,
      "eval_runtime": 250.6516,
      "eval_samples_per_second": 19.581,
      "eval_steps_per_second": 1.959,
      "step": 273
    },
    {
      "epoch": 0.8502715283165244,
      "grad_norm": 2.6875,
      "learning_rate": 8.983541155862114e-06,
      "loss": 1.2908,
      "step": 274
    },
    {
      "epoch": 0.8533747090768037,
      "grad_norm": 2.375,
      "learning_rate": 8.976100857170016e-06,
      "loss": 1.2589,
      "step": 275
    },
    {
      "epoch": 0.8564778898370831,
      "grad_norm": 2.390625,
      "learning_rate": 8.96863653171787e-06,
      "loss": 1.2303,
      "step": 276
    },
    {
      "epoch": 0.8595810705973623,
      "grad_norm": 2.4375,
      "learning_rate": 8.96114822461106e-06,
      "loss": 1.2143,
      "step": 277
    },
    {
      "epoch": 0.8626842513576416,
      "grad_norm": 2.421875,
      "learning_rate": 8.953635981099888e-06,
      "loss": 1.2615,
      "step": 278
    },
    {
      "epoch": 0.8657874321179209,
      "grad_norm": 2.484375,
      "learning_rate": 8.946099846579293e-06,
      "loss": 1.2276,
      "step": 279
    },
    {
      "epoch": 0.8688906128782001,
      "grad_norm": 2.609375,
      "learning_rate": 8.938539866588593e-06,
      "loss": 1.3039,
      "step": 280
    },
    {
      "epoch": 0.8719937936384794,
      "grad_norm": 2.390625,
      "learning_rate": 8.930956086811186e-06,
      "loss": 1.2025,
      "step": 281
    },
    {
      "epoch": 0.8750969743987588,
      "grad_norm": 2.390625,
      "learning_rate": 8.923348553074301e-06,
      "loss": 1.2982,
      "step": 282
    },
    {
      "epoch": 0.878200155159038,
      "grad_norm": 2.578125,
      "learning_rate": 8.915717311348693e-06,
      "loss": 1.1953,
      "step": 283
    },
    {
      "epoch": 0.8813033359193173,
      "grad_norm": 2.484375,
      "learning_rate": 8.908062407748393e-06,
      "loss": 1.2537,
      "step": 284
    },
    {
      "epoch": 0.8844065166795966,
      "grad_norm": 2.421875,
      "learning_rate": 8.900383888530406e-06,
      "loss": 1.2848,
      "step": 285
    },
    {
      "epoch": 0.8875096974398758,
      "grad_norm": 2.359375,
      "learning_rate": 8.892681800094447e-06,
      "loss": 1.2584,
      "step": 286
    },
    {
      "epoch": 0.8906128782001551,
      "grad_norm": 2.359375,
      "learning_rate": 8.884956188982653e-06,
      "loss": 1.2078,
      "step": 287
    },
    {
      "epoch": 0.8937160589604345,
      "grad_norm": 2.59375,
      "learning_rate": 8.877207101879301e-06,
      "loss": 1.2836,
      "step": 288
    },
    {
      "epoch": 0.8968192397207138,
      "grad_norm": 2.703125,
      "learning_rate": 8.869434585610534e-06,
      "loss": 1.2583,
      "step": 289
    },
    {
      "epoch": 0.899922420480993,
      "grad_norm": 2.390625,
      "learning_rate": 8.861638687144071e-06,
      "loss": 1.2266,
      "step": 290
    },
    {
      "epoch": 0.9030256012412723,
      "grad_norm": 2.390625,
      "learning_rate": 8.85381945358892e-06,
      "loss": 1.2314,
      "step": 291
    },
    {
      "epoch": 0.9061287820015516,
      "grad_norm": 2.390625,
      "learning_rate": 8.845976932195104e-06,
      "loss": 1.2327,
      "step": 292
    },
    {
      "epoch": 0.9092319627618308,
      "grad_norm": 2.328125,
      "learning_rate": 8.838111170353367e-06,
      "loss": 1.2145,
      "step": 293
    },
    {
      "epoch": 0.9123351435221102,
      "grad_norm": 2.390625,
      "learning_rate": 8.83022221559489e-06,
      "loss": 1.2117,
      "step": 294
    },
    {
      "epoch": 0.9123351435221102,
      "eval_loss": 1.9623801708221436,
      "eval_runtime": 35.8785,
      "eval_samples_per_second": 136.795,
      "eval_steps_per_second": 13.685,
      "step": 294
    },
    {
      "epoch": 0.9154383242823895,
      "grad_norm": 2.5,
      "learning_rate": 8.822310115591008e-06,
      "loss": 1.2262,
      "step": 295
    },
    {
      "epoch": 0.9185415050426687,
      "grad_norm": 2.21875,
      "learning_rate": 8.81437491815291e-06,
      "loss": 1.1984,
      "step": 296
    },
    {
      "epoch": 0.921644685802948,
      "grad_norm": 2.34375,
      "learning_rate": 8.806416671231365e-06,
      "loss": 1.2082,
      "step": 297
    },
    {
      "epoch": 0.9247478665632273,
      "grad_norm": 2.359375,
      "learning_rate": 8.798435422916425e-06,
      "loss": 1.1831,
      "step": 298
    },
    {
      "epoch": 0.9278510473235065,
      "grad_norm": 2.328125,
      "learning_rate": 8.79043122143713e-06,
      "loss": 1.2115,
      "step": 299
    },
    {
      "epoch": 0.9309542280837859,
      "grad_norm": 2.421875,
      "learning_rate": 8.782404115161226e-06,
      "loss": 1.2618,
      "step": 300
    },
    {
      "epoch": 0.9340574088440652,
      "grad_norm": 2.53125,
      "learning_rate": 8.774354152594862e-06,
      "loss": 1.2367,
      "step": 301
    },
    {
      "epoch": 0.9371605896043445,
      "grad_norm": 2.390625,
      "learning_rate": 8.766281382382312e-06,
      "loss": 1.284,
      "step": 302
    },
    {
      "epoch": 0.9402637703646237,
      "grad_norm": 2.328125,
      "learning_rate": 8.758185853305663e-06,
      "loss": 1.2414,
      "step": 303
    },
    {
      "epoch": 0.943366951124903,
      "grad_norm": 2.421875,
      "learning_rate": 8.750067614284534e-06,
      "loss": 1.2511,
      "step": 304
    },
    {
      "epoch": 0.9464701318851824,
      "grad_norm": 2.28125,
      "learning_rate": 8.741926714375776e-06,
      "loss": 1.2644,
      "step": 305
    },
    {
      "epoch": 0.9495733126454616,
      "grad_norm": 2.53125,
      "learning_rate": 8.733763202773172e-06,
      "loss": 1.2366,
      "step": 306
    },
    {
      "epoch": 0.9526764934057409,
      "grad_norm": 2.40625,
      "learning_rate": 8.725577128807144e-06,
      "loss": 1.187,
      "step": 307
    },
    {
      "epoch": 0.9557796741660202,
      "grad_norm": 2.46875,
      "learning_rate": 8.717368541944453e-06,
      "loss": 1.3023,
      "step": 308
    },
    {
      "epoch": 0.9588828549262994,
      "grad_norm": 2.34375,
      "learning_rate": 8.709137491787904e-06,
      "loss": 1.2771,
      "step": 309
    },
    {
      "epoch": 0.9619860356865787,
      "grad_norm": 2.28125,
      "learning_rate": 8.700884028076042e-06,
      "loss": 1.2141,
      "step": 310
    },
    {
      "epoch": 0.9650892164468581,
      "grad_norm": 2.296875,
      "learning_rate": 8.692608200682848e-06,
      "loss": 1.2747,
      "step": 311
    },
    {
      "epoch": 0.9681923972071373,
      "grad_norm": 2.375,
      "learning_rate": 8.684310059617448e-06,
      "loss": 1.19,
      "step": 312
    },
    {
      "epoch": 0.9712955779674166,
      "grad_norm": 2.25,
      "learning_rate": 8.675989655023802e-06,
      "loss": 1.2019,
      "step": 313
    },
    {
      "epoch": 0.9743987587276959,
      "grad_norm": 2.59375,
      "learning_rate": 8.667647037180403e-06,
      "loss": 1.2308,
      "step": 314
    },
    {
      "epoch": 0.9775019394879751,
      "grad_norm": 2.484375,
      "learning_rate": 8.659282256499975e-06,
      "loss": 1.2354,
      "step": 315
    },
    {
      "epoch": 0.9775019394879751,
      "eval_loss": 1.9195327758789062,
      "eval_runtime": 35.8519,
      "eval_samples_per_second": 136.897,
      "eval_steps_per_second": 13.695,
      "step": 315
    },
    {
      "epoch": 0.9806051202482544,
      "grad_norm": 2.328125,
      "learning_rate": 8.650895363529172e-06,
      "loss": 1.2574,
      "step": 316
    },
    {
      "epoch": 0.9837083010085338,
      "grad_norm": 2.421875,
      "learning_rate": 8.642486408948262e-06,
      "loss": 1.1776,
      "step": 317
    },
    {
      "epoch": 0.9868114817688131,
      "grad_norm": 2.328125,
      "learning_rate": 8.634055443570825e-06,
      "loss": 1.2481,
      "step": 318
    },
    {
      "epoch": 0.9899146625290923,
      "grad_norm": 2.40625,
      "learning_rate": 8.625602518343457e-06,
      "loss": 1.2453,
      "step": 319
    },
    {
      "epoch": 0.9930178432893716,
      "grad_norm": 2.46875,
      "learning_rate": 8.617127684345445e-06,
      "loss": 1.2528,
      "step": 320
    },
    {
      "epoch": 0.9961210240496509,
      "grad_norm": 2.234375,
      "learning_rate": 8.608630992788469e-06,
      "loss": 1.1804,
      "step": 321
    },
    {
      "epoch": 0.9992242048099301,
      "grad_norm": 2.375,
      "learning_rate": 8.600112495016289e-06,
      "loss": 1.2785,
      "step": 322
    },
    {
      "epoch": 1.0,
      "grad_norm": 5.65625,
      "learning_rate": 8.591572242504433e-06,
      "loss": 1.2355,
      "step": 323
    },
    {
      "epoch": 1.0031031807602793,
      "grad_norm": 2.5,
      "learning_rate": 8.583010286859896e-06,
      "loss": 1.1954,
      "step": 324
    },
    {
      "epoch": 1.0062063615205585,
      "grad_norm": 2.453125,
      "learning_rate": 8.574426679820813e-06,
      "loss": 1.1361,
      "step": 325
    },
    {
      "epoch": 1.0093095422808378,
      "grad_norm": 2.390625,
      "learning_rate": 8.565821473256157e-06,
      "loss": 1.1297,
      "step": 326
    },
    {
      "epoch": 1.012412723041117,
      "grad_norm": 2.375,
      "learning_rate": 8.557194719165423e-06,
      "loss": 1.2078,
      "step": 327
    },
    {
      "epoch": 1.0155159038013963,
      "grad_norm": 2.40625,
      "learning_rate": 8.548546469678311e-06,
      "loss": 1.1734,
      "step": 328
    },
    {
      "epoch": 1.0186190845616758,
      "grad_norm": 2.25,
      "learning_rate": 8.539876777054416e-06,
      "loss": 1.1338,
      "step": 329
    },
    {
      "epoch": 1.021722265321955,
      "grad_norm": 2.609375,
      "learning_rate": 8.531185693682909e-06,
      "loss": 1.1638,
      "step": 330
    },
    {
      "epoch": 1.0248254460822344,
      "grad_norm": 2.375,
      "learning_rate": 8.522473272082217e-06,
      "loss": 1.1417,
      "step": 331
    },
    {
      "epoch": 1.0279286268425136,
      "grad_norm": 2.59375,
      "learning_rate": 8.513739564899714e-06,
      "loss": 1.1864,
      "step": 332
    },
    {
      "epoch": 1.031031807602793,
      "grad_norm": 2.265625,
      "learning_rate": 8.504984624911398e-06,
      "loss": 1.1592,
      "step": 333
    },
    {
      "epoch": 1.0341349883630722,
      "grad_norm": 2.40625,
      "learning_rate": 8.496208505021572e-06,
      "loss": 1.1373,
      "step": 334
    },
    {
      "epoch": 1.0372381691233514,
      "grad_norm": 2.46875,
      "learning_rate": 8.487411258262522e-06,
      "loss": 1.1574,
      "step": 335
    },
    {
      "epoch": 1.0403413498836307,
      "grad_norm": 2.40625,
      "learning_rate": 8.478592937794202e-06,
      "loss": 1.196,
      "step": 336
    },
    {
      "epoch": 1.0403413498836307,
      "eval_loss": 1.97748601436615,
      "eval_runtime": 35.8658,
      "eval_samples_per_second": 136.844,
      "eval_steps_per_second": 13.69,
      "step": 336
    },
    {
      "epoch": 1.04344453064391,
      "grad_norm": 2.375,
      "learning_rate": 8.469753596903909e-06,
      "loss": 1.1376,
      "step": 337
    },
    {
      "epoch": 1.0465477114041892,
      "grad_norm": 2.40625,
      "learning_rate": 8.460893289005964e-06,
      "loss": 1.1185,
      "step": 338
    },
    {
      "epoch": 1.0496508921644685,
      "grad_norm": 2.5,
      "learning_rate": 8.452012067641389e-06,
      "loss": 1.137,
      "step": 339
    },
    {
      "epoch": 1.0527540729247478,
      "grad_norm": 2.609375,
      "learning_rate": 8.443109986477574e-06,
      "loss": 1.1651,
      "step": 340
    },
    {
      "epoch": 1.0558572536850273,
      "grad_norm": 2.453125,
      "learning_rate": 8.434187099307966e-06,
      "loss": 1.1089,
      "step": 341
    },
    {
      "epoch": 1.0589604344453065,
      "grad_norm": 2.265625,
      "learning_rate": 8.42524346005174e-06,
      "loss": 1.1332,
      "step": 342
    },
    {
      "epoch": 1.0620636152055858,
      "grad_norm": 2.25,
      "learning_rate": 8.416279122753468e-06,
      "loss": 1.1763,
      "step": 343
    },
    {
      "epoch": 1.065166795965865,
      "grad_norm": 2.4375,
      "learning_rate": 8.407294141582797e-06,
      "loss": 1.2123,
      "step": 344
    },
    {
      "epoch": 1.0682699767261443,
      "grad_norm": 2.328125,
      "learning_rate": 8.398288570834124e-06,
      "loss": 1.1491,
      "step": 345
    },
    {
      "epoch": 1.0713731574864236,
      "grad_norm": 2.3125,
      "learning_rate": 8.389262464926256e-06,
      "loss": 1.1794,
      "step": 346
    },
    {
      "epoch": 1.0744763382467029,
      "grad_norm": 2.515625,
      "learning_rate": 8.380215878402102e-06,
      "loss": 1.1142,
      "step": 347
    },
    {
      "epoch": 1.0775795190069821,
      "grad_norm": 2.484375,
      "learning_rate": 8.37114886592832e-06,
      "loss": 1.1929,
      "step": 348
    },
    {
      "epoch": 1.0806826997672614,
      "grad_norm": 2.359375,
      "learning_rate": 8.362061482295001e-06,
      "loss": 1.1814,
      "step": 349
    },
    {
      "epoch": 1.0837858805275407,
      "grad_norm": 2.4375,
      "learning_rate": 8.35295378241534e-06,
      "loss": 1.1803,
      "step": 350
    },
    {
      "epoch": 1.08688906128782,
      "grad_norm": 2.390625,
      "learning_rate": 8.343825821325294e-06,
      "loss": 1.1848,
      "step": 351
    },
    {
      "epoch": 1.0899922420480994,
      "grad_norm": 2.484375,
      "learning_rate": 8.334677654183254e-06,
      "loss": 1.2559,
      "step": 352
    },
    {
      "epoch": 1.0930954228083787,
      "grad_norm": 2.5,
      "learning_rate": 8.325509336269715e-06,
      "loss": 1.135,
      "step": 353
    },
    {
      "epoch": 1.096198603568658,
      "grad_norm": 2.28125,
      "learning_rate": 8.316320922986935e-06,
      "loss": 1.161,
      "step": 354
    },
    {
      "epoch": 1.0993017843289372,
      "grad_norm": 2.4375,
      "learning_rate": 8.307112469858608e-06,
      "loss": 1.1508,
      "step": 355
    },
    {
      "epoch": 1.1024049650892165,
      "grad_norm": 2.390625,
      "learning_rate": 8.297884032529525e-06,
      "loss": 1.2198,
      "step": 356
    },
    {
      "epoch": 1.1055081458494957,
      "grad_norm": 2.328125,
      "learning_rate": 8.288635666765232e-06,
      "loss": 1.1979,
      "step": 357
    },
    {
      "epoch": 1.1055081458494957,
      "eval_loss": 2.036484718322754,
      "eval_runtime": 36.1857,
      "eval_samples_per_second": 135.634,
      "eval_steps_per_second": 13.569,
      "step": 357
    },
    {
      "epoch": 1.108611326609775,
      "grad_norm": 2.375,
      "learning_rate": 8.279367428451703e-06,
      "loss": 1.1964,
      "step": 358
    },
    {
      "epoch": 1.1117145073700543,
      "grad_norm": 2.359375,
      "learning_rate": 8.270079373594996e-06,
      "loss": 1.1771,
      "step": 359
    },
    {
      "epoch": 1.1148176881303336,
      "grad_norm": 2.5,
      "learning_rate": 8.26077155832092e-06,
      "loss": 1.1847,
      "step": 360
    },
    {
      "epoch": 1.1179208688906128,
      "grad_norm": 2.328125,
      "learning_rate": 8.251444038874685e-06,
      "loss": 1.1378,
      "step": 361
    },
    {
      "epoch": 1.121024049650892,
      "grad_norm": 2.546875,
      "learning_rate": 8.242096871620577e-06,
      "loss": 1.213,
      "step": 362
    },
    {
      "epoch": 1.1241272304111714,
      "grad_norm": 2.40625,
      "learning_rate": 8.232730113041606e-06,
      "loss": 1.1765,
      "step": 363
    },
    {
      "epoch": 1.1272304111714506,
      "grad_norm": 2.4375,
      "learning_rate": 8.223343819739164e-06,
      "loss": 1.1371,
      "step": 364
    },
    {
      "epoch": 1.13033359193173,
      "grad_norm": 2.53125,
      "learning_rate": 8.213938048432697e-06,
      "loss": 1.176,
      "step": 365
    },
    {
      "epoch": 1.1334367726920094,
      "grad_norm": 2.375,
      "learning_rate": 8.204512855959345e-06,
      "loss": 1.1724,
      "step": 366
    },
    {
      "epoch": 1.1365399534522886,
      "grad_norm": 2.609375,
      "learning_rate": 8.19506829927361e-06,
      "loss": 1.1465,
      "step": 367
    },
    {
      "epoch": 1.139643134212568,
      "grad_norm": 2.53125,
      "learning_rate": 8.185604435447001e-06,
      "loss": 1.1733,
      "step": 368
    },
    {
      "epoch": 1.1427463149728472,
      "grad_norm": 2.25,
      "learning_rate": 8.176121321667708e-06,
      "loss": 1.1048,
      "step": 369
    },
    {
      "epoch": 1.1458494957331264,
      "grad_norm": 2.40625,
      "learning_rate": 8.166619015240236e-06,
      "loss": 1.1502,
      "step": 370
    },
    {
      "epoch": 1.1489526764934057,
      "grad_norm": 2.453125,
      "learning_rate": 8.157097573585066e-06,
      "loss": 1.1734,
      "step": 371
    },
    {
      "epoch": 1.152055857253685,
      "grad_norm": 2.40625,
      "learning_rate": 8.147557054238317e-06,
      "loss": 1.1346,
      "step": 372
    },
    {
      "epoch": 1.1551590380139642,
      "grad_norm": 2.40625,
      "learning_rate": 8.137997514851385e-06,
      "loss": 1.1649,
      "step": 373
    },
    {
      "epoch": 1.1582622187742435,
      "grad_norm": 2.59375,
      "learning_rate": 8.128419013190599e-06,
      "loss": 1.1594,
      "step": 374
    },
    {
      "epoch": 1.1613653995345228,
      "grad_norm": 2.5,
      "learning_rate": 8.118821607136876e-06,
      "loss": 1.1648,
      "step": 375
    },
    {
      "epoch": 1.1644685802948023,
      "grad_norm": 2.703125,
      "learning_rate": 8.109205354685367e-06,
      "loss": 1.1687,
      "step": 376
    },
    {
      "epoch": 1.1675717610550815,
      "grad_norm": 2.453125,
      "learning_rate": 8.099570313945114e-06,
      "loss": 1.179,
      "step": 377
    },
    {
      "epoch": 1.1706749418153608,
      "grad_norm": 2.34375,
      "learning_rate": 8.08991654313868e-06,
      "loss": 1.1228,
      "step": 378
    },
    {
      "epoch": 1.1706749418153608,
      "eval_loss": 1.9693981409072876,
      "eval_runtime": 35.7658,
      "eval_samples_per_second": 137.226,
      "eval_steps_per_second": 13.728,
      "step": 378
    },
    {
      "epoch": 1.17377812257564,
      "grad_norm": 2.40625,
      "learning_rate": 8.080244100601822e-06,
      "loss": 1.1577,
      "step": 379
    },
    {
      "epoch": 1.1768813033359193,
      "grad_norm": 2.390625,
      "learning_rate": 8.07055304478312e-06,
      "loss": 1.1484,
      "step": 380
    },
    {
      "epoch": 1.1799844840961986,
      "grad_norm": 2.359375,
      "learning_rate": 8.060843434243631e-06,
      "loss": 1.1235,
      "step": 381
    },
    {
      "epoch": 1.1830876648564779,
      "grad_norm": 2.359375,
      "learning_rate": 8.051115327656538e-06,
      "loss": 1.1327,
      "step": 382
    },
    {
      "epoch": 1.1861908456167571,
      "grad_norm": 2.453125,
      "learning_rate": 8.041368783806786e-06,
      "loss": 1.0883,
      "step": 383
    },
    {
      "epoch": 1.1892940263770364,
      "grad_norm": 2.328125,
      "learning_rate": 8.031603861590738e-06,
      "loss": 1.1616,
      "step": 384
    },
    {
      "epoch": 1.1923972071373157,
      "grad_norm": 2.40625,
      "learning_rate": 8.021820620015811e-06,
      "loss": 1.0995,
      "step": 385
    },
    {
      "epoch": 1.195500387897595,
      "grad_norm": 2.5,
      "learning_rate": 8.012019118200123e-06,
      "loss": 1.1695,
      "step": 386
    },
    {
      "epoch": 1.1986035686578744,
      "grad_norm": 2.53125,
      "learning_rate": 8.002199415372133e-06,
      "loss": 1.1907,
      "step": 387
    },
    {
      "epoch": 1.2017067494181537,
      "grad_norm": 2.5625,
      "learning_rate": 7.992361570870289e-06,
      "loss": 1.1828,
      "step": 388
    },
    {
      "epoch": 1.204809930178433,
      "grad_norm": 2.46875,
      "learning_rate": 7.982505644142658e-06,
      "loss": 1.1774,
      "step": 389
    },
    {
      "epoch": 1.2079131109387122,
      "grad_norm": 2.59375,
      "learning_rate": 7.972631694746583e-06,
      "loss": 1.1403,
      "step": 390
    },
    {
      "epoch": 1.2110162916989915,
      "grad_norm": 2.640625,
      "learning_rate": 7.96273978234831e-06,
      "loss": 1.1674,
      "step": 391
    },
    {
      "epoch": 1.2141194724592708,
      "grad_norm": 2.4375,
      "learning_rate": 7.95282996672263e-06,
      "loss": 1.146,
      "step": 392
    },
    {
      "epoch": 1.21722265321955,
      "grad_norm": 2.40625,
      "learning_rate": 7.942902307752523e-06,
      "loss": 1.1752,
      "step": 393
    },
    {
      "epoch": 1.2203258339798293,
      "grad_norm": 2.5,
      "learning_rate": 7.932956865428792e-06,
      "loss": 1.1953,
      "step": 394
    },
    {
      "epoch": 1.2234290147401086,
      "grad_norm": 2.453125,
      "learning_rate": 7.922993699849698e-06,
      "loss": 1.1515,
      "step": 395
    },
    {
      "epoch": 1.2265321955003878,
      "grad_norm": 2.65625,
      "learning_rate": 7.913012871220605e-06,
      "loss": 1.1567,
      "step": 396
    },
    {
      "epoch": 1.229635376260667,
      "grad_norm": 2.578125,
      "learning_rate": 7.903014439853605e-06,
      "loss": 1.1189,
      "step": 397
    },
    {
      "epoch": 1.2327385570209466,
      "grad_norm": 2.46875,
      "learning_rate": 7.892998466167166e-06,
      "loss": 1.2005,
      "step": 398
    },
    {
      "epoch": 1.2358417377812256,
      "grad_norm": 2.234375,
      "learning_rate": 7.882965010685756e-06,
      "loss": 1.172,
      "step": 399
    },
    {
      "epoch": 1.2358417377812256,
      "eval_loss": 1.978397011756897,
      "eval_runtime": 35.7917,
      "eval_samples_per_second": 137.127,
      "eval_steps_per_second": 13.718,
      "step": 399
    },
    {
      "epoch": 1.2389449185415051,
      "grad_norm": 2.625,
      "learning_rate": 7.872914134039485e-06,
      "loss": 1.214,
      "step": 400
    },
    {
      "epoch": 1.2420480993017844,
      "grad_norm": 2.53125,
      "learning_rate": 7.862845896963735e-06,
      "loss": 1.1649,
      "step": 401
    },
    {
      "epoch": 1.2451512800620637,
      "grad_norm": 2.421875,
      "learning_rate": 7.85276036029879e-06,
      "loss": 1.1688,
      "step": 402
    },
    {
      "epoch": 1.248254460822343,
      "grad_norm": 2.359375,
      "learning_rate": 7.842657584989477e-06,
      "loss": 1.091,
      "step": 403
    },
    {
      "epoch": 1.2513576415826222,
      "grad_norm": 2.40625,
      "learning_rate": 7.83253763208479e-06,
      "loss": 1.1585,
      "step": 404
    },
    {
      "epoch": 1.2544608223429015,
      "grad_norm": 2.484375,
      "learning_rate": 7.822400562737524e-06,
      "loss": 1.1149,
      "step": 405
    },
    {
      "epoch": 1.2575640031031807,
      "grad_norm": 2.34375,
      "learning_rate": 7.812246438203905e-06,
      "loss": 1.2065,
      "step": 406
    },
    {
      "epoch": 1.26066718386346,
      "grad_norm": 2.34375,
      "learning_rate": 7.802075319843218e-06,
      "loss": 1.2129,
      "step": 407
    },
    {
      "epoch": 1.2637703646237393,
      "grad_norm": 2.484375,
      "learning_rate": 7.791887269117441e-06,
      "loss": 1.1309,
      "step": 408
    },
    {
      "epoch": 1.2668735453840188,
      "grad_norm": 2.421875,
      "learning_rate": 7.78168234759087e-06,
      "loss": 1.2209,
      "step": 409
    },
    {
      "epoch": 1.2699767261442978,
      "grad_norm": 2.390625,
      "learning_rate": 7.771460616929751e-06,
      "loss": 1.1344,
      "step": 410
    },
    {
      "epoch": 1.2730799069045773,
      "grad_norm": 2.40625,
      "learning_rate": 7.761222138901899e-06,
      "loss": 1.2226,
      "step": 411
    },
    {
      "epoch": 1.2761830876648566,
      "grad_norm": 2.375,
      "learning_rate": 7.750966975376328e-06,
      "loss": 1.2185,
      "step": 412
    },
    {
      "epoch": 1.2792862684251358,
      "grad_norm": 2.59375,
      "learning_rate": 7.740695188322888e-06,
      "loss": 1.1982,
      "step": 413
    },
    {
      "epoch": 1.282389449185415,
      "grad_norm": 2.453125,
      "learning_rate": 7.730406839811874e-06,
      "loss": 1.1562,
      "step": 414
    },
    {
      "epoch": 1.2854926299456944,
      "grad_norm": 2.421875,
      "learning_rate": 7.720101992013661e-06,
      "loss": 1.1604,
      "step": 415
    },
    {
      "epoch": 1.2885958107059736,
      "grad_norm": 2.78125,
      "learning_rate": 7.709780707198328e-06,
      "loss": 1.1235,
      "step": 416
    },
    {
      "epoch": 1.2916989914662529,
      "grad_norm": 2.453125,
      "learning_rate": 7.699443047735274e-06,
      "loss": 1.1747,
      "step": 417
    },
    {
      "epoch": 1.2948021722265322,
      "grad_norm": 2.484375,
      "learning_rate": 7.689089076092851e-06,
      "loss": 1.1309,
      "step": 418
    },
    {
      "epoch": 1.2979053529868114,
      "grad_norm": 2.390625,
      "learning_rate": 7.678718854837982e-06,
      "loss": 1.1509,
      "step": 419
    },
    {
      "epoch": 1.3010085337470907,
      "grad_norm": 2.640625,
      "learning_rate": 7.66833244663578e-06,
      "loss": 1.1678,
      "step": 420
    },
    {
      "epoch": 1.3010085337470907,
      "eval_loss": 1.9534467458724976,
      "eval_runtime": 35.7845,
      "eval_samples_per_second": 137.154,
      "eval_steps_per_second": 13.721,
      "step": 420
    },
    {
      "epoch": 1.30411171450737,
      "grad_norm": 2.671875,
      "learning_rate": 7.657929914249177e-06,
      "loss": 1.2053,
      "step": 421
    },
    {
      "epoch": 1.3072148952676494,
      "grad_norm": 2.515625,
      "learning_rate": 7.647511320538537e-06,
      "loss": 1.1725,
      "step": 422
    },
    {
      "epoch": 1.3103180760279285,
      "grad_norm": 2.421875,
      "learning_rate": 7.637076728461277e-06,
      "loss": 1.1406,
      "step": 423
    },
    {
      "epoch": 1.313421256788208,
      "grad_norm": 2.40625,
      "learning_rate": 7.626626201071494e-06,
      "loss": 1.1241,
      "step": 424
    },
    {
      "epoch": 1.3165244375484872,
      "grad_norm": 2.375,
      "learning_rate": 7.616159801519578e-06,
      "loss": 1.1318,
      "step": 425
    },
    {
      "epoch": 1.3196276183087665,
      "grad_norm": 2.296875,
      "learning_rate": 7.605677593051828e-06,
      "loss": 1.1945,
      "step": 426
    },
    {
      "epoch": 1.3227307990690458,
      "grad_norm": 2.59375,
      "learning_rate": 7.5951796390100755e-06,
      "loss": 1.1873,
      "step": 427
    },
    {
      "epoch": 1.325833979829325,
      "grad_norm": 2.609375,
      "learning_rate": 7.584666002831296e-06,
      "loss": 1.1874,
      "step": 428
    },
    {
      "epoch": 1.3289371605896043,
      "grad_norm": 2.671875,
      "learning_rate": 7.574136748047234e-06,
      "loss": 1.1678,
      "step": 429
    },
    {
      "epoch": 1.3320403413498836,
      "grad_norm": 2.40625,
      "learning_rate": 7.563591938284012e-06,
      "loss": 1.1927,
      "step": 430
    },
    {
      "epoch": 1.3351435221101629,
      "grad_norm": 2.40625,
      "learning_rate": 7.5530316372617466e-06,
      "loss": 1.1717,
      "step": 431
    },
    {
      "epoch": 1.3382467028704421,
      "grad_norm": 2.421875,
      "learning_rate": 7.542455908794164e-06,
      "loss": 1.184,
      "step": 432
    },
    {
      "epoch": 1.3413498836307216,
      "grad_norm": 2.6875,
      "learning_rate": 7.53186481678822e-06,
      "loss": 1.167,
      "step": 433
    },
    {
      "epoch": 1.3444530643910007,
      "grad_norm": 2.484375,
      "learning_rate": 7.521258425243705e-06,
      "loss": 1.1959,
      "step": 434
    },
    {
      "epoch": 1.3475562451512801,
      "grad_norm": 2.609375,
      "learning_rate": 7.510636798252863e-06,
      "loss": 1.1461,
      "step": 435
    },
    {
      "epoch": 1.3506594259115594,
      "grad_norm": 2.296875,
      "learning_rate": 7.500000000000001e-06,
      "loss": 1.1515,
      "step": 436
    },
    {
      "epoch": 1.3537626066718387,
      "grad_norm": 2.375,
      "learning_rate": 7.489348094761104e-06,
      "loss": 1.1457,
      "step": 437
    },
    {
      "epoch": 1.356865787432118,
      "grad_norm": 2.46875,
      "learning_rate": 7.478681146903448e-06,
      "loss": 1.1341,
      "step": 438
    },
    {
      "epoch": 1.3599689681923972,
      "grad_norm": 2.28125,
      "learning_rate": 7.467999220885204e-06,
      "loss": 1.1546,
      "step": 439
    },
    {
      "epoch": 1.3630721489526765,
      "grad_norm": 2.484375,
      "learning_rate": 7.457302381255055e-06,
      "loss": 1.1084,
      "step": 440
    },
    {
      "epoch": 1.3661753297129557,
      "grad_norm": 2.640625,
      "learning_rate": 7.446590692651805e-06,
      "loss": 1.1563,
      "step": 441
    },
    {
      "epoch": 1.3661753297129557,
      "eval_loss": 2.031402111053467,
      "eval_runtime": 35.8545,
      "eval_samples_per_second": 136.886,
      "eval_steps_per_second": 13.694,
      "step": 441
    },
    {
      "epoch": 1.369278510473235,
      "grad_norm": 2.46875,
      "learning_rate": 7.4358642198039835e-06,
      "loss": 1.1775,
      "step": 442
    },
    {
      "epoch": 1.3723816912335143,
      "grad_norm": 2.296875,
      "learning_rate": 7.425123027529462e-06,
      "loss": 1.158,
      "step": 443
    },
    {
      "epoch": 1.3754848719937938,
      "grad_norm": 2.515625,
      "learning_rate": 7.414367180735058e-06,
      "loss": 1.1896,
      "step": 444
    },
    {
      "epoch": 1.3785880527540728,
      "grad_norm": 2.46875,
      "learning_rate": 7.403596744416142e-06,
      "loss": 1.1765,
      "step": 445
    },
    {
      "epoch": 1.3816912335143523,
      "grad_norm": 2.5625,
      "learning_rate": 7.392811783656246e-06,
      "loss": 1.1446,
      "step": 446
    },
    {
      "epoch": 1.3847944142746316,
      "grad_norm": 2.4375,
      "learning_rate": 7.382012363626671e-06,
      "loss": 1.1335,
      "step": 447
    },
    {
      "epoch": 1.3878975950349108,
      "grad_norm": 2.546875,
      "learning_rate": 7.371198549586091e-06,
      "loss": 1.1915,
      "step": 448
    },
    {
      "epoch": 1.39100077579519,
      "grad_norm": 2.359375,
      "learning_rate": 7.360370406880163e-06,
      "loss": 1.0985,
      "step": 449
    },
    {
      "epoch": 1.3941039565554694,
      "grad_norm": 2.546875,
      "learning_rate": 7.349528000941126e-06,
      "loss": 1.1282,
      "step": 450
    },
    {
      "epoch": 1.3972071373157486,
      "grad_norm": 2.390625,
      "learning_rate": 7.338671397287409e-06,
      "loss": 1.143,
      "step": 451
    },
    {
      "epoch": 1.400310318076028,
      "grad_norm": 2.421875,
      "learning_rate": 7.327800661523239e-06,
      "loss": 1.1618,
      "step": 452
    },
    {
      "epoch": 1.4034134988363072,
      "grad_norm": 2.4375,
      "learning_rate": 7.316915859338234e-06,
      "loss": 1.1189,
      "step": 453
    },
    {
      "epoch": 1.4065166795965864,
      "grad_norm": 2.453125,
      "learning_rate": 7.306017056507018e-06,
      "loss": 1.1848,
      "step": 454
    },
    {
      "epoch": 1.4096198603568657,
      "grad_norm": 2.46875,
      "learning_rate": 7.295104318888816e-06,
      "loss": 1.1913,
      "step": 455
    },
    {
      "epoch": 1.412723041117145,
      "grad_norm": 2.546875,
      "learning_rate": 7.284177712427056e-06,
      "loss": 1.1528,
      "step": 456
    },
    {
      "epoch": 1.4158262218774245,
      "grad_norm": 2.40625,
      "learning_rate": 7.273237303148976e-06,
      "loss": 1.1662,
      "step": 457
    },
    {
      "epoch": 1.4189294026377037,
      "grad_norm": 2.421875,
      "learning_rate": 7.262283157165219e-06,
      "loss": 1.1411,
      "step": 458
    },
    {
      "epoch": 1.422032583397983,
      "grad_norm": 2.234375,
      "learning_rate": 7.251315340669438e-06,
      "loss": 1.1499,
      "step": 459
    },
    {
      "epoch": 1.4251357641582623,
      "grad_norm": 2.4375,
      "learning_rate": 7.240333919937893e-06,
      "loss": 1.1134,
      "step": 460
    },
    {
      "epoch": 1.4282389449185415,
      "grad_norm": 2.453125,
      "learning_rate": 7.229338961329053e-06,
      "loss": 1.1471,
      "step": 461
    },
    {
      "epoch": 1.4313421256788208,
      "grad_norm": 2.484375,
      "learning_rate": 7.21833053128319e-06,
      "loss": 1.168,
      "step": 462
    },
    {
      "epoch": 1.4313421256788208,
      "eval_loss": 1.9588382244110107,
      "eval_runtime": 35.8415,
      "eval_samples_per_second": 136.936,
      "eval_steps_per_second": 13.699,
      "step": 462
    },
    {
      "epoch": 1.4344453064391,
      "grad_norm": 2.6875,
      "learning_rate": 7.207308696321985e-06,
      "loss": 1.1538,
      "step": 463
    },
    {
      "epoch": 1.4375484871993793,
      "grad_norm": 2.453125,
      "learning_rate": 7.196273523048121e-06,
      "loss": 1.2015,
      "step": 464
    },
    {
      "epoch": 1.4406516679596586,
      "grad_norm": 2.34375,
      "learning_rate": 7.1852250781448816e-06,
      "loss": 1.1491,
      "step": 465
    },
    {
      "epoch": 1.4437548487199379,
      "grad_norm": 2.46875,
      "learning_rate": 7.174163428375748e-06,
      "loss": 1.1466,
      "step": 466
    },
    {
      "epoch": 1.4468580294802171,
      "grad_norm": 2.375,
      "learning_rate": 7.163088640583993e-06,
      "loss": 1.1376,
      "step": 467
    },
    {
      "epoch": 1.4499612102404966,
      "grad_norm": 2.484375,
      "learning_rate": 7.1520007816922864e-06,
      "loss": 1.1642,
      "step": 468
    },
    {
      "epoch": 1.4530643910007757,
      "grad_norm": 2.53125,
      "learning_rate": 7.140899918702276e-06,
      "loss": 1.1521,
      "step": 469
    },
    {
      "epoch": 1.4561675717610552,
      "grad_norm": 2.390625,
      "learning_rate": 7.1297861186942e-06,
      "loss": 1.1807,
      "step": 470
    },
    {
      "epoch": 1.4592707525213344,
      "grad_norm": 2.46875,
      "learning_rate": 7.118659448826461e-06,
      "loss": 1.2098,
      "step": 471
    },
    {
      "epoch": 1.4623739332816137,
      "grad_norm": 2.515625,
      "learning_rate": 7.107519976335241e-06,
      "loss": 1.203,
      "step": 472
    },
    {
      "epoch": 1.465477114041893,
      "grad_norm": 2.375,
      "learning_rate": 7.09636776853408e-06,
      "loss": 1.1532,
      "step": 473
    },
    {
      "epoch": 1.4685802948021722,
      "grad_norm": 2.609375,
      "learning_rate": 7.08520289281348e-06,
      "loss": 1.1522,
      "step": 474
    },
    {
      "epoch": 1.4716834755624515,
      "grad_norm": 2.375,
      "learning_rate": 7.0740254166404884e-06,
      "loss": 1.1642,
      "step": 475
    },
    {
      "epoch": 1.4747866563227308,
      "grad_norm": 2.359375,
      "learning_rate": 7.062835407558295e-06,
      "loss": 1.1297,
      "step": 476
    },
    {
      "epoch": 1.47788983708301,
      "grad_norm": 2.421875,
      "learning_rate": 7.051632933185823e-06,
      "loss": 1.1004,
      "step": 477
    },
    {
      "epoch": 1.4809930178432893,
      "grad_norm": 2.390625,
      "learning_rate": 7.040418061217325e-06,
      "loss": 1.157,
      "step": 478
    },
    {
      "epoch": 1.4840961986035688,
      "grad_norm": 2.484375,
      "learning_rate": 7.029190859421965e-06,
      "loss": 1.2079,
      "step": 479
    },
    {
      "epoch": 1.4871993793638478,
      "grad_norm": 2.5,
      "learning_rate": 7.017951395643417e-06,
      "loss": 1.1726,
      "step": 480
    },
    {
      "epoch": 1.4903025601241273,
      "grad_norm": 2.3125,
      "learning_rate": 7.006699737799445e-06,
      "loss": 1.1703,
      "step": 481
    },
    {
      "epoch": 1.4934057408844066,
      "grad_norm": 2.484375,
      "learning_rate": 6.99543595388151e-06,
      "loss": 1.1953,
      "step": 482
    },
    {
      "epoch": 1.4965089216446859,
      "grad_norm": 2.359375,
      "learning_rate": 6.984160111954339e-06,
      "loss": 1.1265,
      "step": 483
    },
    {
      "epoch": 1.4965089216446859,
      "eval_loss": 2.08042573928833,
      "eval_runtime": 35.6965,
      "eval_samples_per_second": 137.492,
      "eval_steps_per_second": 13.755,
      "step": 483
    },
    {
      "epoch": 1.4996121024049651,
      "grad_norm": 2.390625,
      "learning_rate": 6.972872280155528e-06,
      "loss": 1.094,
      "step": 484
    },
    {
      "epoch": 1.5027152831652444,
      "grad_norm": 2.53125,
      "learning_rate": 6.9615725266951236e-06,
      "loss": 1.1665,
      "step": 485
    },
    {
      "epoch": 1.5058184639255237,
      "grad_norm": 2.53125,
      "learning_rate": 6.950260919855214e-06,
      "loss": 1.1463,
      "step": 486
    },
    {
      "epoch": 1.508921644685803,
      "grad_norm": 2.578125,
      "learning_rate": 6.938937527989511e-06,
      "loss": 1.1631,
      "step": 487
    },
    {
      "epoch": 1.5120248254460822,
      "grad_norm": 2.140625,
      "learning_rate": 6.927602419522947e-06,
      "loss": 1.1669,
      "step": 488
    },
    {
      "epoch": 1.5151280062063615,
      "grad_norm": 2.453125,
      "learning_rate": 6.916255662951249e-06,
      "loss": 1.1507,
      "step": 489
    },
    {
      "epoch": 1.518231186966641,
      "grad_norm": 2.4375,
      "learning_rate": 6.9048973268405375e-06,
      "loss": 1.1594,
      "step": 490
    },
    {
      "epoch": 1.52133436772692,
      "grad_norm": 2.390625,
      "learning_rate": 6.8935274798269e-06,
      "loss": 1.1285,
      "step": 491
    },
    {
      "epoch": 1.5244375484871995,
      "grad_norm": 2.53125,
      "learning_rate": 6.882146190615985e-06,
      "loss": 1.1226,
      "step": 492
    },
    {
      "epoch": 1.5275407292474785,
      "grad_norm": 2.40625,
      "learning_rate": 6.8707535279825845e-06,
      "loss": 1.1681,
      "step": 493
    },
    {
      "epoch": 1.530643910007758,
      "grad_norm": 2.484375,
      "learning_rate": 6.859349560770215e-06,
      "loss": 1.1341,
      "step": 494
    },
    {
      "epoch": 1.5337470907680373,
      "grad_norm": 2.59375,
      "learning_rate": 6.847934357890708e-06,
      "loss": 1.1456,
      "step": 495
    },
    {
      "epoch": 1.5368502715283165,
      "grad_norm": 2.359375,
      "learning_rate": 6.836507988323785e-06,
      "loss": 1.1539,
      "step": 496
    },
    {
      "epoch": 1.5399534522885958,
      "grad_norm": 2.453125,
      "learning_rate": 6.825070521116649e-06,
      "loss": 1.1387,
      "step": 497
    },
    {
      "epoch": 1.543056633048875,
      "grad_norm": 2.390625,
      "learning_rate": 6.813622025383565e-06,
      "loss": 1.1544,
      "step": 498
    },
    {
      "epoch": 1.5461598138091543,
      "grad_norm": 2.46875,
      "learning_rate": 6.802162570305436e-06,
      "loss": 1.1427,
      "step": 499
    },
    {
      "epoch": 1.5492629945694336,
      "grad_norm": 2.328125,
      "learning_rate": 6.790692225129398e-06,
      "loss": 1.2065,
      "step": 500
    },
    {
      "epoch": 1.552366175329713,
      "grad_norm": 2.28125,
      "learning_rate": 6.779211059168382e-06,
      "loss": 1.1164,
      "step": 501
    },
    {
      "epoch": 1.5554693560899922,
      "grad_norm": 2.59375,
      "learning_rate": 6.767719141800718e-06,
      "loss": 1.2022,
      "step": 502
    },
    {
      "epoch": 1.5585725368502716,
      "grad_norm": 2.453125,
      "learning_rate": 6.756216542469699e-06,
      "loss": 1.1622,
      "step": 503
    },
    {
      "epoch": 1.5616757176105507,
      "grad_norm": 2.359375,
      "learning_rate": 6.7447033306831685e-06,
      "loss": 1.1641,
      "step": 504
    },
    {
      "epoch": 1.5616757176105507,
      "eval_loss": 2.0795576572418213,
      "eval_runtime": 35.7223,
      "eval_samples_per_second": 137.393,
      "eval_steps_per_second": 13.745,
      "step": 504
    },
    {
      "epoch": 1.5647788983708302,
      "grad_norm": 2.5,
      "learning_rate": 6.733179576013098e-06,
      "loss": 1.2067,
      "step": 505
    },
    {
      "epoch": 1.5678820791311094,
      "grad_norm": 2.46875,
      "learning_rate": 6.7216453480951674e-06,
      "loss": 1.162,
      "step": 506
    },
    {
      "epoch": 1.5709852598913887,
      "grad_norm": 2.46875,
      "learning_rate": 6.710100716628345e-06,
      "loss": 1.1947,
      "step": 507
    },
    {
      "epoch": 1.574088440651668,
      "grad_norm": 2.546875,
      "learning_rate": 6.698545751374465e-06,
      "loss": 1.1546,
      "step": 508
    },
    {
      "epoch": 1.5771916214119472,
      "grad_norm": 2.34375,
      "learning_rate": 6.686980522157807e-06,
      "loss": 1.2119,
      "step": 509
    },
    {
      "epoch": 1.5802948021722265,
      "grad_norm": 2.4375,
      "learning_rate": 6.675405098864676e-06,
      "loss": 1.1596,
      "step": 510
    },
    {
      "epoch": 1.5833979829325058,
      "grad_norm": 2.359375,
      "learning_rate": 6.663819551442969e-06,
      "loss": 1.1021,
      "step": 511
    },
    {
      "epoch": 1.5865011636927853,
      "grad_norm": 2.46875,
      "learning_rate": 6.652223949901775e-06,
      "loss": 1.1899,
      "step": 512
    },
    {
      "epoch": 1.5896043444530643,
      "grad_norm": 2.546875,
      "learning_rate": 6.640618364310924e-06,
      "loss": 1.2043,
      "step": 513
    },
    {
      "epoch": 1.5927075252133438,
      "grad_norm": 2.40625,
      "learning_rate": 6.629002864800589e-06,
      "loss": 1.1882,
      "step": 514
    },
    {
      "epoch": 1.5958107059736228,
      "grad_norm": 2.5625,
      "learning_rate": 6.6173775215608405e-06,
      "loss": 1.2291,
      "step": 515
    },
    {
      "epoch": 1.5989138867339023,
      "grad_norm": 2.359375,
      "learning_rate": 6.605742404841241e-06,
      "loss": 1.0876,
      "step": 516
    },
    {
      "epoch": 1.6020170674941814,
      "grad_norm": 2.265625,
      "learning_rate": 6.594097584950406e-06,
      "loss": 1.1539,
      "step": 517
    },
    {
      "epoch": 1.6051202482544609,
      "grad_norm": 2.265625,
      "learning_rate": 6.582443132255591e-06,
      "loss": 1.1623,
      "step": 518
    },
    {
      "epoch": 1.6082234290147401,
      "grad_norm": 2.59375,
      "learning_rate": 6.570779117182258e-06,
      "loss": 1.1727,
      "step": 519
    },
    {
      "epoch": 1.6113266097750194,
      "grad_norm": 2.40625,
      "learning_rate": 6.55910561021365e-06,
      "loss": 1.1476,
      "step": 520
    },
    {
      "epoch": 1.6144297905352987,
      "grad_norm": 2.375,
      "learning_rate": 6.5474226818903695e-06,
      "loss": 1.1748,
      "step": 521
    },
    {
      "epoch": 1.617532971295578,
      "grad_norm": 2.359375,
      "learning_rate": 6.53573040280995e-06,
      "loss": 1.1262,
      "step": 522
    },
    {
      "epoch": 1.6206361520558572,
      "grad_norm": 2.4375,
      "learning_rate": 6.524028843626433e-06,
      "loss": 1.1376,
      "step": 523
    },
    {
      "epoch": 1.6237393328161365,
      "grad_norm": 2.34375,
      "learning_rate": 6.512318075049935e-06,
      "loss": 1.1946,
      "step": 524
    },
    {
      "epoch": 1.626842513576416,
      "grad_norm": 2.28125,
      "learning_rate": 6.50059816784622e-06,
      "loss": 1.1605,
      "step": 525
    },
    {
      "epoch": 1.626842513576416,
      "eval_loss": 1.952155351638794,
      "eval_runtime": 251.6428,
      "eval_samples_per_second": 19.504,
      "eval_steps_per_second": 1.951,
      "step": 525
    },
    {
      "epoch": 1.629945694336695,
      "grad_norm": 2.390625,
      "learning_rate": 6.488869192836279e-06,
      "loss": 1.2135,
      "step": 526
    },
    {
      "epoch": 1.6330488750969745,
      "grad_norm": 2.46875,
      "learning_rate": 6.477131220895896e-06,
      "loss": 1.2096,
      "step": 527
    },
    {
      "epoch": 1.6361520558572535,
      "grad_norm": 2.5,
      "learning_rate": 6.465384322955225e-06,
      "loss": 1.2101,
      "step": 528
    },
    {
      "epoch": 1.639255236617533,
      "grad_norm": 2.53125,
      "learning_rate": 6.453628569998352e-06,
      "loss": 1.1372,
      "step": 529
    },
    {
      "epoch": 1.6423584173778123,
      "grad_norm": 2.453125,
      "learning_rate": 6.441864033062879e-06,
      "loss": 1.126,
      "step": 530
    },
    {
      "epoch": 1.6454615981380916,
      "grad_norm": 2.390625,
      "learning_rate": 6.430090783239479e-06,
      "loss": 1.2024,
      "step": 531
    },
    {
      "epoch": 1.6485647788983708,
      "grad_norm": 2.421875,
      "learning_rate": 6.418308891671484e-06,
      "loss": 1.1499,
      "step": 532
    },
    {
      "epoch": 1.65166795965865,
      "grad_norm": 2.265625,
      "learning_rate": 6.406518429554442e-06,
      "loss": 1.1552,
      "step": 533
    },
    {
      "epoch": 1.6547711404189294,
      "grad_norm": 2.359375,
      "learning_rate": 6.39471946813569e-06,
      "loss": 1.1941,
      "step": 534
    },
    {
      "epoch": 1.6578743211792086,
      "grad_norm": 2.234375,
      "learning_rate": 6.38291207871393e-06,
      "loss": 1.194,
      "step": 535
    },
    {
      "epoch": 1.6609775019394881,
      "grad_norm": 2.25,
      "learning_rate": 6.3710963326387845e-06,
      "loss": 1.1875,
      "step": 536
    },
    {
      "epoch": 1.6640806826997672,
      "grad_norm": 2.390625,
      "learning_rate": 6.35927230131038e-06,
      "loss": 1.2164,
      "step": 537
    },
    {
      "epoch": 1.6671838634600467,
      "grad_norm": 2.328125,
      "learning_rate": 6.347440056178904e-06,
      "loss": 1.1272,
      "step": 538
    },
    {
      "epoch": 1.6702870442203257,
      "grad_norm": 2.34375,
      "learning_rate": 6.335599668744187e-06,
      "loss": 1.1435,
      "step": 539
    },
    {
      "epoch": 1.6733902249806052,
      "grad_norm": 2.40625,
      "learning_rate": 6.323751210555252e-06,
      "loss": 1.1615,
      "step": 540
    },
    {
      "epoch": 1.6764934057408845,
      "grad_norm": 2.390625,
      "learning_rate": 6.311894753209896e-06,
      "loss": 1.1166,
      "step": 541
    },
    {
      "epoch": 1.6795965865011637,
      "grad_norm": 2.40625,
      "learning_rate": 6.300030368354255e-06,
      "loss": 1.1649,
      "step": 542
    },
    {
      "epoch": 1.682699767261443,
      "grad_norm": 2.4375,
      "learning_rate": 6.288158127682366e-06,
      "loss": 1.1628,
      "step": 543
    },
    {
      "epoch": 1.6858029480217223,
      "grad_norm": 2.375,
      "learning_rate": 6.27627810293574e-06,
      "loss": 1.1455,
      "step": 544
    },
    {
      "epoch": 1.6889061287820015,
      "grad_norm": 2.25,
      "learning_rate": 6.264390365902921e-06,
      "loss": 1.1832,
      "step": 545
    },
    {
      "epoch": 1.6920093095422808,
      "grad_norm": 2.328125,
      "learning_rate": 6.252494988419065e-06,
      "loss": 1.1646,
      "step": 546
    },
    {
      "epoch": 1.6920093095422808,
      "eval_loss": 1.9887759685516357,
      "eval_runtime": 35.8685,
      "eval_samples_per_second": 136.833,
      "eval_steps_per_second": 13.689,
      "step": 546
    },
    {
      "epoch": 1.6951124903025603,
      "grad_norm": 2.421875,
      "learning_rate": 6.2405920423654885e-06,
      "loss": 1.1503,
      "step": 547
    },
    {
      "epoch": 1.6982156710628393,
      "grad_norm": 2.4375,
      "learning_rate": 6.228681599669248e-06,
      "loss": 1.2077,
      "step": 548
    },
    {
      "epoch": 1.7013188518231188,
      "grad_norm": 2.421875,
      "learning_rate": 6.216763732302704e-06,
      "loss": 1.159,
      "step": 549
    },
    {
      "epoch": 1.7044220325833979,
      "grad_norm": 2.46875,
      "learning_rate": 6.204838512283073e-06,
      "loss": 1.1653,
      "step": 550
    },
    {
      "epoch": 1.7075252133436774,
      "grad_norm": 2.515625,
      "learning_rate": 6.192906011672011e-06,
      "loss": 1.1851,
      "step": 551
    },
    {
      "epoch": 1.7106283941039564,
      "grad_norm": 2.578125,
      "learning_rate": 6.180966302575166e-06,
      "loss": 1.1317,
      "step": 552
    },
    {
      "epoch": 1.7137315748642359,
      "grad_norm": 2.421875,
      "learning_rate": 6.169019457141745e-06,
      "loss": 1.185,
      "step": 553
    },
    {
      "epoch": 1.7168347556245152,
      "grad_norm": 2.328125,
      "learning_rate": 6.157065547564079e-06,
      "loss": 1.1506,
      "step": 554
    },
    {
      "epoch": 1.7199379363847944,
      "grad_norm": 2.359375,
      "learning_rate": 6.145104646077186e-06,
      "loss": 1.1666,
      "step": 555
    },
    {
      "epoch": 1.7230411171450737,
      "grad_norm": 2.390625,
      "learning_rate": 6.133136824958334e-06,
      "loss": 1.2083,
      "step": 556
    },
    {
      "epoch": 1.726144297905353,
      "grad_norm": 2.421875,
      "learning_rate": 6.121162156526607e-06,
      "loss": 1.1284,
      "step": 557
    },
    {
      "epoch": 1.7292474786656322,
      "grad_norm": 2.328125,
      "learning_rate": 6.109180713142464e-06,
      "loss": 1.1647,
      "step": 558
    },
    {
      "epoch": 1.7323506594259115,
      "grad_norm": 2.328125,
      "learning_rate": 6.097192567207304e-06,
      "loss": 1.1642,
      "step": 559
    },
    {
      "epoch": 1.735453840186191,
      "grad_norm": 2.4375,
      "learning_rate": 6.0851977911630275e-06,
      "loss": 1.0983,
      "step": 560
    },
    {
      "epoch": 1.73855702094647,
      "grad_norm": 2.375,
      "learning_rate": 6.073196457491602e-06,
      "loss": 1.1573,
      "step": 561
    },
    {
      "epoch": 1.7416602017067495,
      "grad_norm": 2.4375,
      "learning_rate": 6.061188638714616e-06,
      "loss": 1.1853,
      "step": 562
    },
    {
      "epoch": 1.7447633824670286,
      "grad_norm": 2.28125,
      "learning_rate": 6.049174407392855e-06,
      "loss": 1.1374,
      "step": 563
    },
    {
      "epoch": 1.747866563227308,
      "grad_norm": 2.375,
      "learning_rate": 6.037153836125844e-06,
      "loss": 1.1817,
      "step": 564
    },
    {
      "epoch": 1.7509697439875873,
      "grad_norm": 2.609375,
      "learning_rate": 6.0251269975514275e-06,
      "loss": 1.1961,
      "step": 565
    },
    {
      "epoch": 1.7540729247478666,
      "grad_norm": 2.46875,
      "learning_rate": 6.013093964345315e-06,
      "loss": 1.1483,
      "step": 566
    },
    {
      "epoch": 1.7571761055081458,
      "grad_norm": 2.546875,
      "learning_rate": 6.001054809220655e-06,
      "loss": 1.1672,
      "step": 567
    },
    {
      "epoch": 1.7571761055081458,
      "eval_loss": 2.0044093132019043,
      "eval_runtime": 37.4792,
      "eval_samples_per_second": 130.953,
      "eval_steps_per_second": 13.101,
      "step": 567
    },
    {
      "epoch": 1.7602792862684251,
      "grad_norm": 2.28125,
      "learning_rate": 5.989009604927587e-06,
      "loss": 1.1143,
      "step": 568
    },
    {
      "epoch": 1.7633824670287044,
      "grad_norm": 2.34375,
      "learning_rate": 5.9769584242528026e-06,
      "loss": 1.181,
      "step": 569
    },
    {
      "epoch": 1.7664856477889836,
      "grad_norm": 2.265625,
      "learning_rate": 5.96490134001911e-06,
      "loss": 1.1256,
      "step": 570
    },
    {
      "epoch": 1.7695888285492631,
      "grad_norm": 2.375,
      "learning_rate": 5.95283842508499e-06,
      "loss": 1.1619,
      "step": 571
    },
    {
      "epoch": 1.7726920093095422,
      "grad_norm": 2.328125,
      "learning_rate": 5.9407697523441574e-06,
      "loss": 1.157,
      "step": 572
    },
    {
      "epoch": 1.7757951900698217,
      "grad_norm": 2.296875,
      "learning_rate": 5.9286953947251226e-06,
      "loss": 1.1191,
      "step": 573
    },
    {
      "epoch": 1.7788983708301007,
      "grad_norm": 2.4375,
      "learning_rate": 5.916615425190744e-06,
      "loss": 1.2208,
      "step": 574
    },
    {
      "epoch": 1.7820015515903802,
      "grad_norm": 2.375,
      "learning_rate": 5.904529916737793e-06,
      "loss": 1.1728,
      "step": 575
    },
    {
      "epoch": 1.7851047323506595,
      "grad_norm": 2.359375,
      "learning_rate": 5.892438942396515e-06,
      "loss": 1.1498,
      "step": 576
    },
    {
      "epoch": 1.7882079131109387,
      "grad_norm": 2.265625,
      "learning_rate": 5.8803425752301814e-06,
      "loss": 1.1302,
      "step": 577
    },
    {
      "epoch": 1.791311093871218,
      "grad_norm": 2.3125,
      "learning_rate": 5.8682408883346535e-06,
      "loss": 1.2117,
      "step": 578
    },
    {
      "epoch": 1.7944142746314973,
      "grad_norm": 2.421875,
      "learning_rate": 5.856133954837934e-06,
      "loss": 1.1637,
      "step": 579
    },
    {
      "epoch": 1.7975174553917765,
      "grad_norm": 2.34375,
      "learning_rate": 5.844021847899735e-06,
      "loss": 1.1757,
      "step": 580
    },
    {
      "epoch": 1.8006206361520558,
      "grad_norm": 2.53125,
      "learning_rate": 5.831904640711029e-06,
      "loss": 1.1473,
      "step": 581
    },
    {
      "epoch": 1.8037238169123353,
      "grad_norm": 2.5,
      "learning_rate": 5.819782406493607e-06,
      "loss": 1.1207,
      "step": 582
    },
    {
      "epoch": 1.8068269976726143,
      "grad_norm": 2.34375,
      "learning_rate": 5.80765521849964e-06,
      "loss": 1.1871,
      "step": 583
    },
    {
      "epoch": 1.8099301784328938,
      "grad_norm": 2.546875,
      "learning_rate": 5.795523150011229e-06,
      "loss": 1.1812,
      "step": 584
    },
    {
      "epoch": 1.8130333591931729,
      "grad_norm": 2.421875,
      "learning_rate": 5.7833862743399714e-06,
      "loss": 1.1276,
      "step": 585
    },
    {
      "epoch": 1.8161365399534524,
      "grad_norm": 2.4375,
      "learning_rate": 5.771244664826512e-06,
      "loss": 1.0886,
      "step": 586
    },
    {
      "epoch": 1.8192397207137314,
      "grad_norm": 2.390625,
      "learning_rate": 5.7590983948401e-06,
      "loss": 1.2179,
      "step": 587
    },
    {
      "epoch": 1.822342901474011,
      "grad_norm": 2.515625,
      "learning_rate": 5.7469475377781505e-06,
      "loss": 1.2376,
      "step": 588
    },
    {
      "epoch": 1.822342901474011,
      "eval_loss": 2.0171802043914795,
      "eval_runtime": 35.8207,
      "eval_samples_per_second": 137.016,
      "eval_steps_per_second": 13.707,
      "step": 588
    },
    {
      "epoch": 1.8254460822342902,
      "grad_norm": 2.625,
      "learning_rate": 5.734792167065789e-06,
      "loss": 1.1783,
      "step": 589
    },
    {
      "epoch": 1.8285492629945694,
      "grad_norm": 2.40625,
      "learning_rate": 5.722632356155427e-06,
      "loss": 1.0943,
      "step": 590
    },
    {
      "epoch": 1.8316524437548487,
      "grad_norm": 2.5,
      "learning_rate": 5.710468178526299e-06,
      "loss": 1.1657,
      "step": 591
    },
    {
      "epoch": 1.834755624515128,
      "grad_norm": 2.5,
      "learning_rate": 5.698299707684031e-06,
      "loss": 1.1388,
      "step": 592
    },
    {
      "epoch": 1.8378588052754075,
      "grad_norm": 2.453125,
      "learning_rate": 5.686127017160192e-06,
      "loss": 1.1095,
      "step": 593
    },
    {
      "epoch": 1.8409619860356865,
      "grad_norm": 2.4375,
      "learning_rate": 5.6739501805118446e-06,
      "loss": 1.1695,
      "step": 594
    },
    {
      "epoch": 1.844065166795966,
      "grad_norm": 2.4375,
      "learning_rate": 5.661769271321113e-06,
      "loss": 1.1613,
      "step": 595
    },
    {
      "epoch": 1.847168347556245,
      "grad_norm": 2.484375,
      "learning_rate": 5.649584363194725e-06,
      "loss": 1.1707,
      "step": 596
    },
    {
      "epoch": 1.8502715283165245,
      "grad_norm": 2.5,
      "learning_rate": 5.637395529763575e-06,
      "loss": 1.2131,
      "step": 597
    },
    {
      "epoch": 1.8533747090768036,
      "grad_norm": 2.40625,
      "learning_rate": 5.6252028446822805e-06,
      "loss": 1.0979,
      "step": 598
    },
    {
      "epoch": 1.856477889837083,
      "grad_norm": 2.359375,
      "learning_rate": 5.613006381628726e-06,
      "loss": 1.1375,
      "step": 599
    },
    {
      "epoch": 1.8595810705973623,
      "grad_norm": 2.421875,
      "learning_rate": 5.600806214303635e-06,
      "loss": 1.1986,
      "step": 600
    },
    {
      "epoch": 1.8626842513576416,
      "grad_norm": 2.515625,
      "learning_rate": 5.588602416430107e-06,
      "loss": 1.1609,
      "step": 601
    },
    {
      "epoch": 1.8657874321179209,
      "grad_norm": 2.53125,
      "learning_rate": 5.576395061753187e-06,
      "loss": 1.1431,
      "step": 602
    },
    {
      "epoch": 1.8688906128782001,
      "grad_norm": 2.359375,
      "learning_rate": 5.564184224039407e-06,
      "loss": 1.1419,
      "step": 603
    },
    {
      "epoch": 1.8719937936384794,
      "grad_norm": 2.421875,
      "learning_rate": 5.55196997707635e-06,
      "loss": 1.184,
      "step": 604
    },
    {
      "epoch": 1.8750969743987587,
      "grad_norm": 2.4375,
      "learning_rate": 5.5397523946721975e-06,
      "loss": 1.2089,
      "step": 605
    },
    {
      "epoch": 1.8782001551590382,
      "grad_norm": 2.421875,
      "learning_rate": 5.5275315506552904e-06,
      "loss": 1.1781,
      "step": 606
    },
    {
      "epoch": 1.8813033359193172,
      "grad_norm": 2.359375,
      "learning_rate": 5.515307518873677e-06,
      "loss": 1.1243,
      "step": 607
    },
    {
      "epoch": 1.8844065166795967,
      "grad_norm": 2.3125,
      "learning_rate": 5.5030803731946665e-06,
      "loss": 1.1906,
      "step": 608
    },
    {
      "epoch": 1.8875096974398757,
      "grad_norm": 2.296875,
      "learning_rate": 5.490850187504389e-06,
      "loss": 1.1463,
      "step": 609
    },
    {
      "epoch": 1.8875096974398757,
      "eval_loss": 1.9254013299942017,
      "eval_runtime": 35.728,
      "eval_samples_per_second": 137.371,
      "eval_steps_per_second": 13.743,
      "step": 609
    },
    {
      "epoch": 1.8906128782001552,
      "grad_norm": 2.28125,
      "learning_rate": 5.478617035707337e-06,
      "loss": 1.1753,
      "step": 610
    },
    {
      "epoch": 1.8937160589604345,
      "grad_norm": 2.390625,
      "learning_rate": 5.4663809917259395e-06,
      "loss": 1.1966,
      "step": 611
    },
    {
      "epoch": 1.8968192397207138,
      "grad_norm": 2.40625,
      "learning_rate": 5.454142129500086e-06,
      "loss": 1.1586,
      "step": 612
    },
    {
      "epoch": 1.899922420480993,
      "grad_norm": 2.421875,
      "learning_rate": 5.441900522986712e-06,
      "loss": 1.1549,
      "step": 613
    },
    {
      "epoch": 1.9030256012412723,
      "grad_norm": 2.34375,
      "learning_rate": 5.4296562461593275e-06,
      "loss": 1.1511,
      "step": 614
    },
    {
      "epoch": 1.9061287820015516,
      "grad_norm": 2.40625,
      "learning_rate": 5.4174093730075776e-06,
      "loss": 1.134,
      "step": 615
    },
    {
      "epoch": 1.9092319627618308,
      "grad_norm": 2.40625,
      "learning_rate": 5.4051599775368e-06,
      "loss": 1.1616,
      "step": 616
    },
    {
      "epoch": 1.9123351435221103,
      "grad_norm": 2.578125,
      "learning_rate": 5.392908133767573e-06,
      "loss": 1.2098,
      "step": 617
    },
    {
      "epoch": 1.9154383242823894,
      "grad_norm": 2.40625,
      "learning_rate": 5.380653915735272e-06,
      "loss": 1.1173,
      "step": 618
    },
    {
      "epoch": 1.9185415050426688,
      "grad_norm": 2.484375,
      "learning_rate": 5.368397397489615e-06,
      "loss": 1.2081,
      "step": 619
    },
    {
      "epoch": 1.921644685802948,
      "grad_norm": 2.390625,
      "learning_rate": 5.3561386530942195e-06,
      "loss": 1.1217,
      "step": 620
    },
    {
      "epoch": 1.9247478665632274,
      "grad_norm": 2.390625,
      "learning_rate": 5.343877756626163e-06,
      "loss": 1.1726,
      "step": 621
    },
    {
      "epoch": 1.9278510473235064,
      "grad_norm": 2.40625,
      "learning_rate": 5.33161478217552e-06,
      "loss": 1.1427,
      "step": 622
    },
    {
      "epoch": 1.930954228083786,
      "grad_norm": 2.34375,
      "learning_rate": 5.319349803844924e-06,
      "loss": 1.1603,
      "step": 623
    },
    {
      "epoch": 1.9340574088440652,
      "grad_norm": 2.375,
      "learning_rate": 5.307082895749116e-06,
      "loss": 1.1571,
      "step": 624
    },
    {
      "epoch": 1.9371605896043445,
      "grad_norm": 2.375,
      "learning_rate": 5.294814132014503e-06,
      "loss": 1.1626,
      "step": 625
    },
    {
      "epoch": 1.9402637703646237,
      "grad_norm": 2.28125,
      "learning_rate": 5.282543586778698e-06,
      "loss": 1.1779,
      "step": 626
    },
    {
      "epoch": 1.943366951124903,
      "grad_norm": 2.328125,
      "learning_rate": 5.270271334190087e-06,
      "loss": 1.1771,
      "step": 627
    },
    {
      "epoch": 1.9464701318851825,
      "grad_norm": 2.3125,
      "learning_rate": 5.257997448407366e-06,
      "loss": 1.1565,
      "step": 628
    },
    {
      "epoch": 1.9495733126454615,
      "grad_norm": 2.46875,
      "learning_rate": 5.2457220035991055e-06,
      "loss": 1.185,
      "step": 629
    },
    {
      "epoch": 1.952676493405741,
      "grad_norm": 2.5,
      "learning_rate": 5.233445073943295e-06,
      "loss": 1.101,
      "step": 630
    },
    {
      "epoch": 1.952676493405741,
      "eval_loss": 1.9600006341934204,
      "eval_runtime": 36.2147,
      "eval_samples_per_second": 135.525,
      "eval_steps_per_second": 13.558,
      "step": 630
    },
    {
      "epoch": 1.95577967416602,
      "grad_norm": 2.625,
      "learning_rate": 5.221166733626895e-06,
      "loss": 1.1834,
      "step": 631
    },
    {
      "epoch": 1.9588828549262995,
      "grad_norm": 2.53125,
      "learning_rate": 5.208887056845394e-06,
      "loss": 1.2472,
      "step": 632
    },
    {
      "epoch": 1.9619860356865786,
      "grad_norm": 2.34375,
      "learning_rate": 5.196606117802351e-06,
      "loss": 1.081,
      "step": 633
    },
    {
      "epoch": 1.965089216446858,
      "grad_norm": 2.5,
      "learning_rate": 5.184323990708959e-06,
      "loss": 1.1582,
      "step": 634
    },
    {
      "epoch": 1.9681923972071373,
      "grad_norm": 2.34375,
      "learning_rate": 5.1720407497835835e-06,
      "loss": 1.1906,
      "step": 635
    },
    {
      "epoch": 1.9712955779674166,
      "grad_norm": 2.375,
      "learning_rate": 5.159756469251327e-06,
      "loss": 1.1679,
      "step": 636
    },
    {
      "epoch": 1.9743987587276959,
      "grad_norm": 2.421875,
      "learning_rate": 5.147471223343573e-06,
      "loss": 1.1236,
      "step": 637
    },
    {
      "epoch": 1.9775019394879751,
      "grad_norm": 2.671875,
      "learning_rate": 5.135185086297532e-06,
      "loss": 1.1412,
      "step": 638
    },
    {
      "epoch": 1.9806051202482544,
      "grad_norm": 2.421875,
      "learning_rate": 5.122898132355808e-06,
      "loss": 1.1555,
      "step": 639
    },
    {
      "epoch": 1.9837083010085337,
      "grad_norm": 2.46875,
      "learning_rate": 5.110610435765935e-06,
      "loss": 1.1121,
      "step": 640
    },
    {
      "epoch": 1.9868114817688132,
      "grad_norm": 2.359375,
      "learning_rate": 5.09832207077994e-06,
      "loss": 1.1782,
      "step": 641
    },
    {
      "epoch": 1.9899146625290922,
      "grad_norm": 2.421875,
      "learning_rate": 5.0860331116538845e-06,
      "loss": 1.1413,
      "step": 642
    },
    {
      "epoch": 1.9930178432893717,
      "grad_norm": 2.625,
      "learning_rate": 5.073743632647422e-06,
      "loss": 1.1401,
      "step": 643
    },
    {
      "epoch": 1.9961210240496507,
      "grad_norm": 2.390625,
      "learning_rate": 5.061453708023348e-06,
      "loss": 1.1626,
      "step": 644
    },
    {
      "epoch": 1.9992242048099302,
      "grad_norm": 2.640625,
      "learning_rate": 5.0491634120471515e-06,
      "loss": 1.181,
      "step": 645
    },
    {
      "epoch": 2.0,
      "grad_norm": 4.75,
      "learning_rate": 5.0368728189865624e-06,
      "loss": 1.1618,
      "step": 646
    },
    {
      "epoch": 2.0031031807602795,
      "grad_norm": 2.484375,
      "learning_rate": 5.024582003111109e-06,
      "loss": 1.1656,
      "step": 647
    },
    {
      "epoch": 2.0062063615205585,
      "grad_norm": 2.28125,
      "learning_rate": 5.012291038691665e-06,
      "loss": 1.1202,
      "step": 648
    },
    {
      "epoch": 2.009309542280838,
      "grad_norm": 2.5,
      "learning_rate": 5e-06,
      "loss": 1.1216,
      "step": 649
    },
    {
      "epoch": 2.012412723041117,
      "grad_norm": 2.3125,
      "learning_rate": 4.9877089613083356e-06,
      "loss": 1.1373,
      "step": 650
    },
    {
      "epoch": 2.0155159038013966,
      "grad_norm": 2.359375,
      "learning_rate": 4.975417996888892e-06,
      "loss": 1.1448,
      "step": 651
    },
    {
      "epoch": 2.0155159038013966,
      "eval_loss": 2.000762462615967,
      "eval_runtime": 35.8142,
      "eval_samples_per_second": 137.04,
      "eval_steps_per_second": 13.71,
      "step": 651
    },
    {
      "epoch": 2.0186190845616756,
      "grad_norm": 2.203125,
      "learning_rate": 4.9631271810134375e-06,
      "loss": 1.1132,
      "step": 652
    },
    {
      "epoch": 2.021722265321955,
      "grad_norm": 2.40625,
      "learning_rate": 4.950836587952849e-06,
      "loss": 1.0976,
      "step": 653
    },
    {
      "epoch": 2.024825446082234,
      "grad_norm": 2.546875,
      "learning_rate": 4.938546291976654e-06,
      "loss": 1.1023,
      "step": 654
    },
    {
      "epoch": 2.0279286268425136,
      "grad_norm": 2.296875,
      "learning_rate": 4.9262563673525795e-06,
      "loss": 1.1135,
      "step": 655
    },
    {
      "epoch": 2.0310318076027927,
      "grad_norm": 2.453125,
      "learning_rate": 4.913966888346118e-06,
      "loss": 1.122,
      "step": 656
    },
    {
      "epoch": 2.034134988363072,
      "grad_norm": 2.484375,
      "learning_rate": 4.901677929220062e-06,
      "loss": 1.1048,
      "step": 657
    },
    {
      "epoch": 2.0372381691233516,
      "grad_norm": 2.421875,
      "learning_rate": 4.8893895642340665e-06,
      "loss": 1.0579,
      "step": 658
    },
    {
      "epoch": 2.0403413498836307,
      "grad_norm": 2.75,
      "learning_rate": 4.8771018676441935e-06,
      "loss": 1.1112,
      "step": 659
    },
    {
      "epoch": 2.04344453064391,
      "grad_norm": 2.453125,
      "learning_rate": 4.864814913702469e-06,
      "loss": 1.102,
      "step": 660
    },
    {
      "epoch": 2.0465477114041892,
      "grad_norm": 2.703125,
      "learning_rate": 4.852528776656428e-06,
      "loss": 1.1274,
      "step": 661
    },
    {
      "epoch": 2.0496508921644687,
      "grad_norm": 2.53125,
      "learning_rate": 4.840243530748674e-06,
      "loss": 1.095,
      "step": 662
    },
    {
      "epoch": 2.0527540729247478,
      "grad_norm": 2.5625,
      "learning_rate": 4.8279592502164165e-06,
      "loss": 1.1688,
      "step": 663
    },
    {
      "epoch": 2.0558572536850273,
      "grad_norm": 2.359375,
      "learning_rate": 4.815676009291044e-06,
      "loss": 1.1001,
      "step": 664
    },
    {
      "epoch": 2.0589604344453063,
      "grad_norm": 2.484375,
      "learning_rate": 4.803393882197652e-06,
      "loss": 1.106,
      "step": 665
    },
    {
      "epoch": 2.062063615205586,
      "grad_norm": 2.390625,
      "learning_rate": 4.791112943154609e-06,
      "loss": 1.0511,
      "step": 666
    },
    {
      "epoch": 2.065166795965865,
      "grad_norm": 2.484375,
      "learning_rate": 4.778833266373107e-06,
      "loss": 1.0684,
      "step": 667
    },
    {
      "epoch": 2.0682699767261443,
      "grad_norm": 2.328125,
      "learning_rate": 4.766554926056707e-06,
      "loss": 1.1014,
      "step": 668
    },
    {
      "epoch": 2.0713731574864234,
      "grad_norm": 2.515625,
      "learning_rate": 4.754277996400896e-06,
      "loss": 1.084,
      "step": 669
    },
    {
      "epoch": 2.074476338246703,
      "grad_norm": 2.5625,
      "learning_rate": 4.742002551592635e-06,
      "loss": 1.0875,
      "step": 670
    },
    {
      "epoch": 2.0775795190069823,
      "grad_norm": 2.46875,
      "learning_rate": 4.729728665809915e-06,
      "loss": 1.1406,
      "step": 671
    },
    {
      "epoch": 2.0806826997672614,
      "grad_norm": 2.5625,
      "learning_rate": 4.717456413221302e-06,
      "loss": 1.1309,
      "step": 672
    },
    {
      "epoch": 2.0806826997672614,
      "eval_loss": 1.9426084756851196,
      "eval_runtime": 35.8524,
      "eval_samples_per_second": 136.895,
      "eval_steps_per_second": 13.695,
      "step": 672
    },
    {
      "epoch": 2.083785880527541,
      "grad_norm": 2.578125,
      "learning_rate": 4.705185867985498e-06,
      "loss": 1.1033,
      "step": 673
    },
    {
      "epoch": 2.08688906128782,
      "grad_norm": 2.3125,
      "learning_rate": 4.692917104250886e-06,
      "loss": 1.1606,
      "step": 674
    },
    {
      "epoch": 2.0899922420480994,
      "grad_norm": 2.390625,
      "learning_rate": 4.680650196155078e-06,
      "loss": 1.067,
      "step": 675
    },
    {
      "epoch": 2.0930954228083785,
      "grad_norm": 2.375,
      "learning_rate": 4.668385217824482e-06,
      "loss": 1.1679,
      "step": 676
    },
    {
      "epoch": 2.096198603568658,
      "grad_norm": 2.4375,
      "learning_rate": 4.656122243373838e-06,
      "loss": 1.0702,
      "step": 677
    },
    {
      "epoch": 2.099301784328937,
      "grad_norm": 2.484375,
      "learning_rate": 4.643861346905781e-06,
      "loss": 1.1291,
      "step": 678
    },
    {
      "epoch": 2.1024049650892165,
      "grad_norm": 2.484375,
      "learning_rate": 4.631602602510388e-06,
      "loss": 1.1655,
      "step": 679
    },
    {
      "epoch": 2.1055081458494955,
      "grad_norm": 2.296875,
      "learning_rate": 4.619346084264729e-06,
      "loss": 1.1048,
      "step": 680
    },
    {
      "epoch": 2.108611326609775,
      "grad_norm": 2.484375,
      "learning_rate": 4.607091866232428e-06,
      "loss": 1.0637,
      "step": 681
    },
    {
      "epoch": 2.1117145073700545,
      "grad_norm": 2.453125,
      "learning_rate": 4.594840022463201e-06,
      "loss": 1.0853,
      "step": 682
    },
    {
      "epoch": 2.1148176881303336,
      "grad_norm": 2.375,
      "learning_rate": 4.582590626992423e-06,
      "loss": 1.1207,
      "step": 683
    },
    {
      "epoch": 2.117920868890613,
      "grad_norm": 2.5625,
      "learning_rate": 4.570343753840675e-06,
      "loss": 1.1075,
      "step": 684
    },
    {
      "epoch": 2.121024049650892,
      "grad_norm": 2.515625,
      "learning_rate": 4.558099477013288e-06,
      "loss": 1.1045,
      "step": 685
    },
    {
      "epoch": 2.1241272304111716,
      "grad_norm": 2.421875,
      "learning_rate": 4.5458578704999144e-06,
      "loss": 1.0826,
      "step": 686
    },
    {
      "epoch": 2.1272304111714506,
      "grad_norm": 2.359375,
      "learning_rate": 4.533619008274063e-06,
      "loss": 1.1227,
      "step": 687
    },
    {
      "epoch": 2.13033359193173,
      "grad_norm": 2.4375,
      "learning_rate": 4.5213829642926635e-06,
      "loss": 1.1167,
      "step": 688
    },
    {
      "epoch": 2.133436772692009,
      "grad_norm": 2.59375,
      "learning_rate": 4.509149812495613e-06,
      "loss": 1.1303,
      "step": 689
    },
    {
      "epoch": 2.1365399534522886,
      "grad_norm": 2.5,
      "learning_rate": 4.496919626805334e-06,
      "loss": 1.0859,
      "step": 690
    },
    {
      "epoch": 2.1396431342125677,
      "grad_norm": 2.390625,
      "learning_rate": 4.484692481126323e-06,
      "loss": 1.1335,
      "step": 691
    },
    {
      "epoch": 2.142746314972847,
      "grad_norm": 2.53125,
      "learning_rate": 4.4724684493447095e-06,
      "loss": 1.086,
      "step": 692
    },
    {
      "epoch": 2.1458494957331267,
      "grad_norm": 2.5,
      "learning_rate": 4.4602476053278025e-06,
      "loss": 1.13,
      "step": 693
    },
    {
      "epoch": 2.1458494957331267,
      "eval_loss": 1.9602453708648682,
      "eval_runtime": 35.7677,
      "eval_samples_per_second": 137.219,
      "eval_steps_per_second": 13.727,
      "step": 693
    },
    {
      "epoch": 2.1489526764934057,
      "grad_norm": 2.515625,
      "learning_rate": 4.4480300229236525e-06,
      "loss": 1.0437,
      "step": 694
    },
    {
      "epoch": 2.152055857253685,
      "grad_norm": 2.328125,
      "learning_rate": 4.435815775960596e-06,
      "loss": 1.1217,
      "step": 695
    },
    {
      "epoch": 2.1551590380139642,
      "grad_norm": 2.359375,
      "learning_rate": 4.423604938246815e-06,
      "loss": 1.1015,
      "step": 696
    },
    {
      "epoch": 2.1582622187742437,
      "grad_norm": 2.328125,
      "learning_rate": 4.411397583569894e-06,
      "loss": 1.1722,
      "step": 697
    },
    {
      "epoch": 2.161365399534523,
      "grad_norm": 2.515625,
      "learning_rate": 4.399193785696366e-06,
      "loss": 1.145,
      "step": 698
    },
    {
      "epoch": 2.1644685802948023,
      "grad_norm": 2.703125,
      "learning_rate": 4.386993618371275e-06,
      "loss": 1.0876,
      "step": 699
    },
    {
      "epoch": 2.1675717610550813,
      "grad_norm": 2.40625,
      "learning_rate": 4.374797155317721e-06,
      "loss": 1.1115,
      "step": 700
    },
    {
      "epoch": 2.170674941815361,
      "grad_norm": 2.5,
      "learning_rate": 4.362604470236426e-06,
      "loss": 1.0929,
      "step": 701
    },
    {
      "epoch": 2.17377812257564,
      "grad_norm": 2.34375,
      "learning_rate": 4.350415636805276e-06,
      "loss": 1.0232,
      "step": 702
    },
    {
      "epoch": 2.1768813033359193,
      "grad_norm": 2.546875,
      "learning_rate": 4.338230728678888e-06,
      "loss": 1.144,
      "step": 703
    },
    {
      "epoch": 2.179984484096199,
      "grad_norm": 2.515625,
      "learning_rate": 4.326049819488157e-06,
      "loss": 1.0646,
      "step": 704
    },
    {
      "epoch": 2.183087664856478,
      "grad_norm": 2.46875,
      "learning_rate": 4.31387298283981e-06,
      "loss": 1.0774,
      "step": 705
    },
    {
      "epoch": 2.1861908456167574,
      "grad_norm": 2.4375,
      "learning_rate": 4.30170029231597e-06,
      "loss": 1.1138,
      "step": 706
    },
    {
      "epoch": 2.1892940263770364,
      "grad_norm": 2.59375,
      "learning_rate": 4.289531821473702e-06,
      "loss": 1.1083,
      "step": 707
    },
    {
      "epoch": 2.192397207137316,
      "grad_norm": 2.5625,
      "learning_rate": 4.277367643844574e-06,
      "loss": 1.0526,
      "step": 708
    },
    {
      "epoch": 2.195500387897595,
      "grad_norm": 2.53125,
      "learning_rate": 4.265207832934212e-06,
      "loss": 1.0819,
      "step": 709
    },
    {
      "epoch": 2.1986035686578744,
      "grad_norm": 2.5,
      "learning_rate": 4.253052462221852e-06,
      "loss": 1.1216,
      "step": 710
    },
    {
      "epoch": 2.2017067494181535,
      "grad_norm": 2.359375,
      "learning_rate": 4.2409016051599005e-06,
      "loss": 1.0637,
      "step": 711
    },
    {
      "epoch": 2.204809930178433,
      "grad_norm": 2.453125,
      "learning_rate": 4.228755335173488e-06,
      "loss": 1.0968,
      "step": 712
    },
    {
      "epoch": 2.207913110938712,
      "grad_norm": 2.46875,
      "learning_rate": 4.216613725660029e-06,
      "loss": 1.1471,
      "step": 713
    },
    {
      "epoch": 2.2110162916989915,
      "grad_norm": 2.546875,
      "learning_rate": 4.204476849988773e-06,
      "loss": 1.121,
      "step": 714
    },
    {
      "epoch": 2.2110162916989915,
      "eval_loss": 1.9649922847747803,
      "eval_runtime": 35.7739,
      "eval_samples_per_second": 137.195,
      "eval_steps_per_second": 13.725,
      "step": 714
    },
    {
      "epoch": 2.214119472459271,
      "grad_norm": 2.390625,
      "learning_rate": 4.192344781500362e-06,
      "loss": 1.1539,
      "step": 715
    },
    {
      "epoch": 2.21722265321955,
      "grad_norm": 2.609375,
      "learning_rate": 4.180217593506394e-06,
      "loss": 1.1566,
      "step": 716
    },
    {
      "epoch": 2.2203258339798295,
      "grad_norm": 2.5,
      "learning_rate": 4.168095359288973e-06,
      "loss": 1.1327,
      "step": 717
    },
    {
      "epoch": 2.2234290147401086,
      "grad_norm": 2.625,
      "learning_rate": 4.155978152100266e-06,
      "loss": 1.0694,
      "step": 718
    },
    {
      "epoch": 2.226532195500388,
      "grad_norm": 2.40625,
      "learning_rate": 4.143866045162067e-06,
      "loss": 1.1203,
      "step": 719
    },
    {
      "epoch": 2.229635376260667,
      "grad_norm": 2.5625,
      "learning_rate": 4.131759111665349e-06,
      "loss": 1.105,
      "step": 720
    },
    {
      "epoch": 2.2327385570209466,
      "grad_norm": 2.359375,
      "learning_rate": 4.119657424769819e-06,
      "loss": 1.0835,
      "step": 721
    },
    {
      "epoch": 2.2358417377812256,
      "grad_norm": 2.46875,
      "learning_rate": 4.1075610576034855e-06,
      "loss": 1.1392,
      "step": 722
    },
    {
      "epoch": 2.238944918541505,
      "grad_norm": 2.578125,
      "learning_rate": 4.095470083262209e-06,
      "loss": 1.122,
      "step": 723
    },
    {
      "epoch": 2.242048099301784,
      "grad_norm": 2.46875,
      "learning_rate": 4.0833845748092586e-06,
      "loss": 1.0996,
      "step": 724
    },
    {
      "epoch": 2.2451512800620637,
      "grad_norm": 2.484375,
      "learning_rate": 4.07130460527488e-06,
      "loss": 1.1147,
      "step": 725
    },
    {
      "epoch": 2.2482544608223427,
      "grad_norm": 2.484375,
      "learning_rate": 4.059230247655843e-06,
      "loss": 1.1167,
      "step": 726
    },
    {
      "epoch": 2.251357641582622,
      "grad_norm": 2.484375,
      "learning_rate": 4.047161574915012e-06,
      "loss": 1.085,
      "step": 727
    },
    {
      "epoch": 2.2544608223429012,
      "grad_norm": 2.484375,
      "learning_rate": 4.035098659980891e-06,
      "loss": 1.1194,
      "step": 728
    },
    {
      "epoch": 2.2575640031031807,
      "grad_norm": 2.546875,
      "learning_rate": 4.023041575747198e-06,
      "loss": 1.1382,
      "step": 729
    },
    {
      "epoch": 2.26066718386346,
      "grad_norm": 2.34375,
      "learning_rate": 4.010990395072414e-06,
      "loss": 1.0961,
      "step": 730
    },
    {
      "epoch": 2.2637703646237393,
      "grad_norm": 2.515625,
      "learning_rate": 3.9989451907793455e-06,
      "loss": 1.1489,
      "step": 731
    },
    {
      "epoch": 2.2668735453840188,
      "grad_norm": 2.515625,
      "learning_rate": 3.986906035654686e-06,
      "loss": 1.0916,
      "step": 732
    },
    {
      "epoch": 2.269976726144298,
      "grad_norm": 2.296875,
      "learning_rate": 3.974873002448575e-06,
      "loss": 1.1165,
      "step": 733
    },
    {
      "epoch": 2.2730799069045773,
      "grad_norm": 2.609375,
      "learning_rate": 3.962846163874159e-06,
      "loss": 1.037,
      "step": 734
    },
    {
      "epoch": 2.2761830876648563,
      "grad_norm": 2.640625,
      "learning_rate": 3.950825592607147e-06,
      "loss": 1.1065,
      "step": 735
    },
    {
      "epoch": 2.2761830876648563,
      "eval_loss": 1.9238359928131104,
      "eval_runtime": 35.7638,
      "eval_samples_per_second": 137.234,
      "eval_steps_per_second": 13.729,
      "step": 735
    },
    {
      "epoch": 2.279286268425136,
      "grad_norm": 2.53125,
      "learning_rate": 3.938811361285386e-06,
      "loss": 1.0867,
      "step": 736
    },
    {
      "epoch": 2.282389449185415,
      "grad_norm": 2.53125,
      "learning_rate": 3.926803542508401e-06,
      "loss": 1.0935,
      "step": 737
    },
    {
      "epoch": 2.2854926299456944,
      "grad_norm": 2.359375,
      "learning_rate": 3.914802208836973e-06,
      "loss": 1.0897,
      "step": 738
    },
    {
      "epoch": 2.2885958107059734,
      "grad_norm": 2.515625,
      "learning_rate": 3.902807432792698e-06,
      "loss": 1.074,
      "step": 739
    },
    {
      "epoch": 2.291698991466253,
      "grad_norm": 2.625,
      "learning_rate": 3.890819286857537e-06,
      "loss": 1.1133,
      "step": 740
    },
    {
      "epoch": 2.2948021722265324,
      "grad_norm": 2.34375,
      "learning_rate": 3.878837843473393e-06,
      "loss": 1.097,
      "step": 741
    },
    {
      "epoch": 2.2979053529868114,
      "grad_norm": 2.53125,
      "learning_rate": 3.866863175041666e-06,
      "loss": 1.0809,
      "step": 742
    },
    {
      "epoch": 2.301008533747091,
      "grad_norm": 2.625,
      "learning_rate": 3.854895353922816e-06,
      "loss": 1.1494,
      "step": 743
    },
    {
      "epoch": 2.30411171450737,
      "grad_norm": 2.53125,
      "learning_rate": 3.842934452435923e-06,
      "loss": 1.1029,
      "step": 744
    },
    {
      "epoch": 2.3072148952676494,
      "grad_norm": 2.546875,
      "learning_rate": 3.830980542858256e-06,
      "loss": 1.0904,
      "step": 745
    },
    {
      "epoch": 2.3103180760279285,
      "grad_norm": 2.3125,
      "learning_rate": 3.819033697424836e-06,
      "loss": 1.1015,
      "step": 746
    },
    {
      "epoch": 2.313421256788208,
      "grad_norm": 2.5625,
      "learning_rate": 3.80709398832799e-06,
      "loss": 1.1303,
      "step": 747
    },
    {
      "epoch": 2.316524437548487,
      "grad_norm": 2.40625,
      "learning_rate": 3.7951614877169285e-06,
      "loss": 1.1085,
      "step": 748
    },
    {
      "epoch": 2.3196276183087665,
      "grad_norm": 2.5,
      "learning_rate": 3.783236267697298e-06,
      "loss": 1.0755,
      "step": 749
    },
    {
      "epoch": 2.3227307990690456,
      "grad_norm": 2.421875,
      "learning_rate": 3.7713184003307524e-06,
      "loss": 1.1702,
      "step": 750
    },
    {
      "epoch": 2.325833979829325,
      "grad_norm": 2.453125,
      "learning_rate": 3.7594079576345123e-06,
      "loss": 1.1358,
      "step": 751
    },
    {
      "epoch": 2.3289371605896045,
      "grad_norm": 2.5,
      "learning_rate": 3.7475050115809365e-06,
      "loss": 1.096,
      "step": 752
    },
    {
      "epoch": 2.3320403413498836,
      "grad_norm": 2.46875,
      "learning_rate": 3.7356096340970804e-06,
      "loss": 1.1262,
      "step": 753
    },
    {
      "epoch": 2.335143522110163,
      "grad_norm": 2.453125,
      "learning_rate": 3.7237218970642624e-06,
      "loss": 1.1186,
      "step": 754
    },
    {
      "epoch": 2.338246702870442,
      "grad_norm": 2.515625,
      "learning_rate": 3.7118418723176365e-06,
      "loss": 1.1406,
      "step": 755
    },
    {
      "epoch": 2.3413498836307216,
      "grad_norm": 2.5,
      "learning_rate": 3.6999696316457468e-06,
      "loss": 1.1345,
      "step": 756
    },
    {
      "epoch": 2.3413498836307216,
      "eval_loss": 1.9017432928085327,
      "eval_runtime": 35.6863,
      "eval_samples_per_second": 137.532,
      "eval_steps_per_second": 13.759,
      "step": 756
    },
    {
      "epoch": 2.3444530643910007,
      "grad_norm": 2.484375,
      "learning_rate": 3.6881052467901056e-06,
      "loss": 1.1162,
      "step": 757
    },
    {
      "epoch": 2.34755624515128,
      "grad_norm": 2.375,
      "learning_rate": 3.67624878944475e-06,
      "loss": 1.1673,
      "step": 758
    },
    {
      "epoch": 2.350659425911559,
      "grad_norm": 2.5,
      "learning_rate": 3.6644003312558146e-06,
      "loss": 1.1423,
      "step": 759
    },
    {
      "epoch": 2.3537626066718387,
      "grad_norm": 2.4375,
      "learning_rate": 3.6525599438210956e-06,
      "loss": 1.0876,
      "step": 760
    },
    {
      "epoch": 2.3568657874321177,
      "grad_norm": 2.34375,
      "learning_rate": 3.640727698689622e-06,
      "loss": 1.0913,
      "step": 761
    },
    {
      "epoch": 2.359968968192397,
      "grad_norm": 2.375,
      "learning_rate": 3.6289036673612167e-06,
      "loss": 1.0975,
      "step": 762
    },
    {
      "epoch": 2.3630721489526767,
      "grad_norm": 2.46875,
      "learning_rate": 3.617087921286073e-06,
      "loss": 1.0962,
      "step": 763
    },
    {
      "epoch": 2.3661753297129557,
      "grad_norm": 2.40625,
      "learning_rate": 3.6052805318643114e-06,
      "loss": 1.1174,
      "step": 764
    },
    {
      "epoch": 2.3692785104732352,
      "grad_norm": 2.453125,
      "learning_rate": 3.5934815704455595e-06,
      "loss": 1.119,
      "step": 765
    },
    {
      "epoch": 2.3723816912335143,
      "grad_norm": 2.5,
      "learning_rate": 3.5816911083285165e-06,
      "loss": 1.0422,
      "step": 766
    },
    {
      "epoch": 2.3754848719937938,
      "grad_norm": 2.46875,
      "learning_rate": 3.569909216760522e-06,
      "loss": 1.0939,
      "step": 767
    },
    {
      "epoch": 2.378588052754073,
      "grad_norm": 2.421875,
      "learning_rate": 3.558135966937123e-06,
      "loss": 1.1353,
      "step": 768
    },
    {
      "epoch": 2.3816912335143523,
      "grad_norm": 2.515625,
      "learning_rate": 3.5463714300016483e-06,
      "loss": 1.1084,
      "step": 769
    },
    {
      "epoch": 2.3847944142746313,
      "grad_norm": 2.390625,
      "learning_rate": 3.534615677044776e-06,
      "loss": 1.1516,
      "step": 770
    },
    {
      "epoch": 2.387897595034911,
      "grad_norm": 2.4375,
      "learning_rate": 3.522868779104104e-06,
      "loss": 1.0946,
      "step": 771
    },
    {
      "epoch": 2.39100077579519,
      "grad_norm": 2.625,
      "learning_rate": 3.511130807163724e-06,
      "loss": 1.1497,
      "step": 772
    },
    {
      "epoch": 2.3941039565554694,
      "grad_norm": 2.546875,
      "learning_rate": 3.4994018321537824e-06,
      "loss": 1.1186,
      "step": 773
    },
    {
      "epoch": 2.397207137315749,
      "grad_norm": 2.5625,
      "learning_rate": 3.4876819249500675e-06,
      "loss": 1.1518,
      "step": 774
    },
    {
      "epoch": 2.400310318076028,
      "grad_norm": 2.484375,
      "learning_rate": 3.4759711563735676e-06,
      "loss": 1.0628,
      "step": 775
    },
    {
      "epoch": 2.4034134988363074,
      "grad_norm": 2.46875,
      "learning_rate": 3.4642695971900506e-06,
      "loss": 1.0757,
      "step": 776
    },
    {
      "epoch": 2.4065166795965864,
      "grad_norm": 2.609375,
      "learning_rate": 3.452577318109632e-06,
      "loss": 1.1215,
      "step": 777
    },
    {
      "epoch": 2.4065166795965864,
      "eval_loss": 1.9178298711776733,
      "eval_runtime": 241.2459,
      "eval_samples_per_second": 20.344,
      "eval_steps_per_second": 2.035,
      "step": 777
    },
    {
      "epoch": 2.409619860356866,
      "grad_norm": 2.484375,
      "learning_rate": 3.440894389786352e-06,
      "loss": 1.1381,
      "step": 778
    },
    {
      "epoch": 2.412723041117145,
      "grad_norm": 2.5625,
      "learning_rate": 3.4292208828177433e-06,
      "loss": 1.1449,
      "step": 779
    },
    {
      "epoch": 2.4158262218774245,
      "grad_norm": 2.390625,
      "learning_rate": 3.4175568677444094e-06,
      "loss": 1.1079,
      "step": 780
    },
    {
      "epoch": 2.4189294026377035,
      "grad_norm": 2.46875,
      "learning_rate": 3.405902415049593e-06,
      "loss": 1.1132,
      "step": 781
    },
    {
      "epoch": 2.422032583397983,
      "grad_norm": 2.484375,
      "learning_rate": 3.3942575951587614e-06,
      "loss": 1.0993,
      "step": 782
    },
    {
      "epoch": 2.425135764158262,
      "grad_norm": 2.46875,
      "learning_rate": 3.3826224784391616e-06,
      "loss": 1.1136,
      "step": 783
    },
    {
      "epoch": 2.4282389449185415,
      "grad_norm": 2.34375,
      "learning_rate": 3.370997135199413e-06,
      "loss": 1.085,
      "step": 784
    },
    {
      "epoch": 2.431342125678821,
      "grad_norm": 2.609375,
      "learning_rate": 3.3593816356890773e-06,
      "loss": 1.116,
      "step": 785
    },
    {
      "epoch": 2.4344453064391,
      "grad_norm": 2.421875,
      "learning_rate": 3.3477760500982263e-06,
      "loss": 1.0815,
      "step": 786
    },
    {
      "epoch": 2.4375484871993796,
      "grad_norm": 2.390625,
      "learning_rate": 3.3361804485570315e-06,
      "loss": 1.1198,
      "step": 787
    },
    {
      "epoch": 2.4406516679596586,
      "grad_norm": 2.546875,
      "learning_rate": 3.324594901135326e-06,
      "loss": 1.1075,
      "step": 788
    },
    {
      "epoch": 2.443754848719938,
      "grad_norm": 2.46875,
      "learning_rate": 3.3130194778421944e-06,
      "loss": 1.1611,
      "step": 789
    },
    {
      "epoch": 2.446858029480217,
      "grad_norm": 2.53125,
      "learning_rate": 3.3014542486255365e-06,
      "loss": 1.1169,
      "step": 790
    },
    {
      "epoch": 2.4499612102404966,
      "grad_norm": 2.46875,
      "learning_rate": 3.289899283371657e-06,
      "loss": 1.1239,
      "step": 791
    },
    {
      "epoch": 2.4530643910007757,
      "grad_norm": 2.53125,
      "learning_rate": 3.278354651904836e-06,
      "loss": 1.1493,
      "step": 792
    },
    {
      "epoch": 2.456167571761055,
      "grad_norm": 2.515625,
      "learning_rate": 3.2668204239869046e-06,
      "loss": 1.0549,
      "step": 793
    },
    {
      "epoch": 2.459270752521334,
      "grad_norm": 2.546875,
      "learning_rate": 3.255296669316834e-06,
      "loss": 1.1453,
      "step": 794
    },
    {
      "epoch": 2.4623739332816137,
      "grad_norm": 2.453125,
      "learning_rate": 3.243783457530302e-06,
      "loss": 1.1085,
      "step": 795
    },
    {
      "epoch": 2.465477114041893,
      "grad_norm": 2.625,
      "learning_rate": 3.2322808581992825e-06,
      "loss": 1.0609,
      "step": 796
    },
    {
      "epoch": 2.4685802948021722,
      "grad_norm": 2.3125,
      "learning_rate": 3.2207889408316193e-06,
      "loss": 1.0604,
      "step": 797
    },
    {
      "epoch": 2.4716834755624513,
      "grad_norm": 2.546875,
      "learning_rate": 3.2093077748706037e-06,
      "loss": 1.1193,
      "step": 798
    },
    {
      "epoch": 2.4716834755624513,
      "eval_loss": 1.9621881246566772,
      "eval_runtime": 35.8266,
      "eval_samples_per_second": 136.993,
      "eval_steps_per_second": 13.705,
      "step": 798
    },
    {
      "epoch": 2.4747866563227308,
      "grad_norm": 2.59375,
      "learning_rate": 3.1978374296945635e-06,
      "loss": 1.101,
      "step": 799
    },
    {
      "epoch": 2.4778898370830102,
      "grad_norm": 2.609375,
      "learning_rate": 3.1863779746164355e-06,
      "loss": 1.0486,
      "step": 800
    },
    {
      "epoch": 2.4809930178432893,
      "grad_norm": 2.515625,
      "learning_rate": 3.1749294788833506e-06,
      "loss": 1.1238,
      "step": 801
    },
    {
      "epoch": 2.484096198603569,
      "grad_norm": 2.390625,
      "learning_rate": 3.1634920116762175e-06,
      "loss": 1.0923,
      "step": 802
    },
    {
      "epoch": 2.487199379363848,
      "grad_norm": 2.484375,
      "learning_rate": 3.1520656421092943e-06,
      "loss": 1.0818,
      "step": 803
    },
    {
      "epoch": 2.4903025601241273,
      "grad_norm": 2.453125,
      "learning_rate": 3.1406504392297867e-06,
      "loss": 1.0772,
      "step": 804
    },
    {
      "epoch": 2.4934057408844064,
      "grad_norm": 2.40625,
      "learning_rate": 3.1292464720174163e-06,
      "loss": 1.1305,
      "step": 805
    },
    {
      "epoch": 2.496508921644686,
      "grad_norm": 2.453125,
      "learning_rate": 3.1178538093840162e-06,
      "loss": 1.0953,
      "step": 806
    },
    {
      "epoch": 2.499612102404965,
      "grad_norm": 2.359375,
      "learning_rate": 3.106472520173101e-06,
      "loss": 1.15,
      "step": 807
    },
    {
      "epoch": 2.5027152831652444,
      "grad_norm": 2.5625,
      "learning_rate": 3.0951026731594634e-06,
      "loss": 1.1457,
      "step": 808
    },
    {
      "epoch": 2.5058184639255234,
      "grad_norm": 2.578125,
      "learning_rate": 3.0837443370487506e-06,
      "loss": 1.119,
      "step": 809
    },
    {
      "epoch": 2.508921644685803,
      "grad_norm": 2.59375,
      "learning_rate": 3.0723975804770546e-06,
      "loss": 1.1233,
      "step": 810
    },
    {
      "epoch": 2.5120248254460824,
      "grad_norm": 2.546875,
      "learning_rate": 3.061062472010489e-06,
      "loss": 1.1117,
      "step": 811
    },
    {
      "epoch": 2.5151280062063615,
      "grad_norm": 2.640625,
      "learning_rate": 3.0497390801447885e-06,
      "loss": 1.0744,
      "step": 812
    },
    {
      "epoch": 2.518231186966641,
      "grad_norm": 2.546875,
      "learning_rate": 3.038427473304878e-06,
      "loss": 1.1224,
      "step": 813
    },
    {
      "epoch": 2.52133436772692,
      "grad_norm": 2.53125,
      "learning_rate": 3.0271277198444737e-06,
      "loss": 1.144,
      "step": 814
    },
    {
      "epoch": 2.5244375484871995,
      "grad_norm": 2.609375,
      "learning_rate": 3.015839888045663e-06,
      "loss": 1.0959,
      "step": 815
    },
    {
      "epoch": 2.5275407292474785,
      "grad_norm": 2.453125,
      "learning_rate": 3.0045640461184917e-06,
      "loss": 1.0964,
      "step": 816
    },
    {
      "epoch": 2.530643910007758,
      "grad_norm": 2.59375,
      "learning_rate": 2.9933002622005563e-06,
      "loss": 1.1189,
      "step": 817
    },
    {
      "epoch": 2.5337470907680375,
      "grad_norm": 2.40625,
      "learning_rate": 2.9820486043565853e-06,
      "loss": 1.1489,
      "step": 818
    },
    {
      "epoch": 2.5368502715283165,
      "grad_norm": 2.5,
      "learning_rate": 2.9708091405780357e-06,
      "loss": 1.136,
      "step": 819
    },
    {
      "epoch": 2.5368502715283165,
      "eval_loss": 1.9721351861953735,
      "eval_runtime": 35.7663,
      "eval_samples_per_second": 137.224,
      "eval_steps_per_second": 13.728,
      "step": 819
    },
    {
      "epoch": 2.5399534522885956,
      "grad_norm": 2.4375,
      "learning_rate": 2.9595819387826753e-06,
      "loss": 1.0834,
      "step": 820
    },
    {
      "epoch": 2.543056633048875,
      "grad_norm": 2.40625,
      "learning_rate": 2.948367066814177e-06,
      "loss": 1.1013,
      "step": 821
    },
    {
      "epoch": 2.5461598138091546,
      "grad_norm": 2.546875,
      "learning_rate": 2.937164592441708e-06,
      "loss": 1.0714,
      "step": 822
    },
    {
      "epoch": 2.5492629945694336,
      "grad_norm": 2.375,
      "learning_rate": 2.9259745833595132e-06,
      "loss": 1.13,
      "step": 823
    },
    {
      "epoch": 2.552366175329713,
      "grad_norm": 2.640625,
      "learning_rate": 2.9147971071865215e-06,
      "loss": 1.1084,
      "step": 824
    },
    {
      "epoch": 2.555469356089992,
      "grad_norm": 2.546875,
      "learning_rate": 2.9036322314659204e-06,
      "loss": 1.113,
      "step": 825
    },
    {
      "epoch": 2.5585725368502716,
      "grad_norm": 2.578125,
      "learning_rate": 2.89248002366476e-06,
      "loss": 1.1307,
      "step": 826
    },
    {
      "epoch": 2.5616757176105507,
      "grad_norm": 2.375,
      "learning_rate": 2.8813405511735393e-06,
      "loss": 1.077,
      "step": 827
    },
    {
      "epoch": 2.56477889837083,
      "grad_norm": 2.4375,
      "learning_rate": 2.870213881305802e-06,
      "loss": 1.2077,
      "step": 828
    },
    {
      "epoch": 2.5678820791311097,
      "grad_norm": 2.453125,
      "learning_rate": 2.8591000812977245e-06,
      "loss": 1.135,
      "step": 829
    },
    {
      "epoch": 2.5709852598913887,
      "grad_norm": 2.4375,
      "learning_rate": 2.8479992183077144e-06,
      "loss": 1.1161,
      "step": 830
    },
    {
      "epoch": 2.5740884406516678,
      "grad_norm": 2.53125,
      "learning_rate": 2.836911359416008e-06,
      "loss": 1.07,
      "step": 831
    },
    {
      "epoch": 2.5771916214119472,
      "grad_norm": 2.5,
      "learning_rate": 2.8258365716242543e-06,
      "loss": 1.1105,
      "step": 832
    },
    {
      "epoch": 2.5802948021722267,
      "grad_norm": 2.390625,
      "learning_rate": 2.8147749218551205e-06,
      "loss": 1.1038,
      "step": 833
    },
    {
      "epoch": 2.5833979829325058,
      "grad_norm": 2.5,
      "learning_rate": 2.8037264769518802e-06,
      "loss": 1.1408,
      "step": 834
    },
    {
      "epoch": 2.5865011636927853,
      "grad_norm": 2.4375,
      "learning_rate": 2.792691303678016e-06,
      "loss": 1.1608,
      "step": 835
    },
    {
      "epoch": 2.5896043444530643,
      "grad_norm": 2.53125,
      "learning_rate": 2.781669468716811e-06,
      "loss": 1.1112,
      "step": 836
    },
    {
      "epoch": 2.592707525213344,
      "grad_norm": 2.328125,
      "learning_rate": 2.77066103867095e-06,
      "loss": 1.1338,
      "step": 837
    },
    {
      "epoch": 2.595810705973623,
      "grad_norm": 2.46875,
      "learning_rate": 2.7596660800621076e-06,
      "loss": 1.1125,
      "step": 838
    },
    {
      "epoch": 2.5989138867339023,
      "grad_norm": 2.40625,
      "learning_rate": 2.7486846593305626e-06,
      "loss": 1.1002,
      "step": 839
    },
    {
      "epoch": 2.6020170674941814,
      "grad_norm": 2.484375,
      "learning_rate": 2.737716842834781e-06,
      "loss": 1.0804,
      "step": 840
    },
    {
      "epoch": 2.6020170674941814,
      "eval_loss": 1.931480050086975,
      "eval_runtime": 36.7523,
      "eval_samples_per_second": 133.543,
      "eval_steps_per_second": 13.36,
      "step": 840
    },
    {
      "epoch": 2.605120248254461,
      "grad_norm": 2.515625,
      "learning_rate": 2.7267626968510264e-06,
      "loss": 1.0942,
      "step": 841
    },
    {
      "epoch": 2.60822342901474,
      "grad_norm": 2.5625,
      "learning_rate": 2.7158222875729474e-06,
      "loss": 1.105,
      "step": 842
    },
    {
      "epoch": 2.6113266097750194,
      "grad_norm": 2.578125,
      "learning_rate": 2.704895681111187e-06,
      "loss": 1.1656,
      "step": 843
    },
    {
      "epoch": 2.614429790535299,
      "grad_norm": 2.296875,
      "learning_rate": 2.6939829434929834e-06,
      "loss": 1.1065,
      "step": 844
    },
    {
      "epoch": 2.617532971295578,
      "grad_norm": 2.59375,
      "learning_rate": 2.6830841406617664e-06,
      "loss": 1.1014,
      "step": 845
    },
    {
      "epoch": 2.620636152055857,
      "grad_norm": 2.515625,
      "learning_rate": 2.672199338476763e-06,
      "loss": 1.1224,
      "step": 846
    },
    {
      "epoch": 2.6237393328161365,
      "grad_norm": 2.515625,
      "learning_rate": 2.661328602712592e-06,
      "loss": 1.0735,
      "step": 847
    },
    {
      "epoch": 2.626842513576416,
      "grad_norm": 2.5625,
      "learning_rate": 2.650471999058875e-06,
      "loss": 1.1483,
      "step": 848
    },
    {
      "epoch": 2.629945694336695,
      "grad_norm": 2.40625,
      "learning_rate": 2.639629593119837e-06,
      "loss": 1.0598,
      "step": 849
    },
    {
      "epoch": 2.6330488750969745,
      "grad_norm": 2.484375,
      "learning_rate": 2.6288014504139104e-06,
      "loss": 1.1319,
      "step": 850
    },
    {
      "epoch": 2.6361520558572535,
      "grad_norm": 2.40625,
      "learning_rate": 2.617987636373332e-06,
      "loss": 1.0678,
      "step": 851
    },
    {
      "epoch": 2.639255236617533,
      "grad_norm": 2.59375,
      "learning_rate": 2.6071882163437563e-06,
      "loss": 1.1224,
      "step": 852
    },
    {
      "epoch": 2.642358417377812,
      "grad_norm": 2.453125,
      "learning_rate": 2.5964032555838593e-06,
      "loss": 1.1164,
      "step": 853
    },
    {
      "epoch": 2.6454615981380916,
      "grad_norm": 2.453125,
      "learning_rate": 2.5856328192649426e-06,
      "loss": 1.1945,
      "step": 854
    },
    {
      "epoch": 2.648564778898371,
      "grad_norm": 2.4375,
      "learning_rate": 2.574876972470538e-06,
      "loss": 1.1355,
      "step": 855
    },
    {
      "epoch": 2.65166795965865,
      "grad_norm": 2.46875,
      "learning_rate": 2.5641357801960186e-06,
      "loss": 1.1319,
      "step": 856
    },
    {
      "epoch": 2.654771140418929,
      "grad_norm": 2.59375,
      "learning_rate": 2.553409307348197e-06,
      "loss": 1.1443,
      "step": 857
    },
    {
      "epoch": 2.6578743211792086,
      "grad_norm": 2.578125,
      "learning_rate": 2.542697618744945e-06,
      "loss": 1.1323,
      "step": 858
    },
    {
      "epoch": 2.660977501939488,
      "grad_norm": 2.515625,
      "learning_rate": 2.5320007791147954e-06,
      "loss": 1.0649,
      "step": 859
    },
    {
      "epoch": 2.664080682699767,
      "grad_norm": 2.515625,
      "learning_rate": 2.521318853096553e-06,
      "loss": 1.1179,
      "step": 860
    },
    {
      "epoch": 2.6671838634600467,
      "grad_norm": 2.40625,
      "learning_rate": 2.510651905238898e-06,
      "loss": 1.1333,
      "step": 861
    },
    {
      "epoch": 2.6671838634600467,
      "eval_loss": 1.9468801021575928,
      "eval_runtime": 35.7938,
      "eval_samples_per_second": 137.119,
      "eval_steps_per_second": 13.717,
      "step": 861
    },
    {
      "epoch": 2.6702870442203257,
      "grad_norm": 2.53125,
      "learning_rate": 2.5000000000000015e-06,
      "loss": 1.137,
      "step": 862
    },
    {
      "epoch": 2.673390224980605,
      "grad_norm": 2.359375,
      "learning_rate": 2.4893632017471386e-06,
      "loss": 1.0848,
      "step": 863
    },
    {
      "epoch": 2.6764934057408842,
      "grad_norm": 2.4375,
      "learning_rate": 2.478741574756296e-06,
      "loss": 1.1209,
      "step": 864
    },
    {
      "epoch": 2.6795965865011637,
      "grad_norm": 2.484375,
      "learning_rate": 2.4681351832117815e-06,
      "loss": 1.0679,
      "step": 865
    },
    {
      "epoch": 2.682699767261443,
      "grad_norm": 2.4375,
      "learning_rate": 2.457544091205837e-06,
      "loss": 1.0939,
      "step": 866
    },
    {
      "epoch": 2.6858029480217223,
      "grad_norm": 2.4375,
      "learning_rate": 2.446968362738255e-06,
      "loss": 1.1157,
      "step": 867
    },
    {
      "epoch": 2.6889061287820013,
      "grad_norm": 2.359375,
      "learning_rate": 2.4364080617159885e-06,
      "loss": 1.1199,
      "step": 868
    },
    {
      "epoch": 2.692009309542281,
      "grad_norm": 2.5,
      "learning_rate": 2.425863251952767e-06,
      "loss": 1.1516,
      "step": 869
    },
    {
      "epoch": 2.6951124903025603,
      "grad_norm": 2.5625,
      "learning_rate": 2.415333997168705e-06,
      "loss": 1.1805,
      "step": 870
    },
    {
      "epoch": 2.6982156710628393,
      "grad_norm": 2.328125,
      "learning_rate": 2.404820360989928e-06,
      "loss": 1.0636,
      "step": 871
    },
    {
      "epoch": 2.701318851823119,
      "grad_norm": 2.484375,
      "learning_rate": 2.3943224069481736e-06,
      "loss": 1.1254,
      "step": 872
    },
    {
      "epoch": 2.704422032583398,
      "grad_norm": 2.53125,
      "learning_rate": 2.3838401984804225e-06,
      "loss": 1.1262,
      "step": 873
    },
    {
      "epoch": 2.7075252133436774,
      "grad_norm": 2.484375,
      "learning_rate": 2.373373798928507e-06,
      "loss": 1.1243,
      "step": 874
    },
    {
      "epoch": 2.7106283941039564,
      "grad_norm": 2.484375,
      "learning_rate": 2.3629232715387245e-06,
      "loss": 1.0978,
      "step": 875
    },
    {
      "epoch": 2.713731574864236,
      "grad_norm": 2.609375,
      "learning_rate": 2.3524886794614653e-06,
      "loss": 1.1511,
      "step": 876
    },
    {
      "epoch": 2.7168347556245154,
      "grad_norm": 2.390625,
      "learning_rate": 2.3420700857508233e-06,
      "loss": 1.1187,
      "step": 877
    },
    {
      "epoch": 2.7199379363847944,
      "grad_norm": 2.421875,
      "learning_rate": 2.331667553364221e-06,
      "loss": 1.1805,
      "step": 878
    },
    {
      "epoch": 2.7230411171450735,
      "grad_norm": 2.53125,
      "learning_rate": 2.3212811451620194e-06,
      "loss": 1.0986,
      "step": 879
    },
    {
      "epoch": 2.726144297905353,
      "grad_norm": 2.546875,
      "learning_rate": 2.310910923907149e-06,
      "loss": 1.0976,
      "step": 880
    },
    {
      "epoch": 2.7292474786656324,
      "grad_norm": 2.734375,
      "learning_rate": 2.300556952264728e-06,
      "loss": 1.0876,
      "step": 881
    },
    {
      "epoch": 2.7323506594259115,
      "grad_norm": 2.515625,
      "learning_rate": 2.2902192928016737e-06,
      "loss": 1.0353,
      "step": 882
    },
    {
      "epoch": 2.7323506594259115,
      "eval_loss": 1.9487745761871338,
      "eval_runtime": 35.7775,
      "eval_samples_per_second": 137.181,
      "eval_steps_per_second": 13.724,
      "step": 882
    },
    {
      "epoch": 2.735453840186191,
      "grad_norm": 2.5,
      "learning_rate": 2.2798980079863386e-06,
      "loss": 1.0836,
      "step": 883
    },
    {
      "epoch": 2.73855702094647,
      "grad_norm": 2.390625,
      "learning_rate": 2.2695931601881277e-06,
      "loss": 1.0929,
      "step": 884
    },
    {
      "epoch": 2.7416602017067495,
      "grad_norm": 2.453125,
      "learning_rate": 2.2593048116771134e-06,
      "loss": 1.0934,
      "step": 885
    },
    {
      "epoch": 2.7447633824670286,
      "grad_norm": 2.5,
      "learning_rate": 2.249033024623672e-06,
      "loss": 1.1246,
      "step": 886
    },
    {
      "epoch": 2.747866563227308,
      "grad_norm": 2.484375,
      "learning_rate": 2.238777861098102e-06,
      "loss": 1.1809,
      "step": 887
    },
    {
      "epoch": 2.7509697439875875,
      "grad_norm": 2.703125,
      "learning_rate": 2.22853938307025e-06,
      "loss": 1.1397,
      "step": 888
    },
    {
      "epoch": 2.7540729247478666,
      "grad_norm": 2.453125,
      "learning_rate": 2.2183176524091293e-06,
      "loss": 1.1437,
      "step": 889
    },
    {
      "epoch": 2.7571761055081456,
      "grad_norm": 2.296875,
      "learning_rate": 2.208112730882561e-06,
      "loss": 1.1667,
      "step": 890
    },
    {
      "epoch": 2.760279286268425,
      "grad_norm": 2.515625,
      "learning_rate": 2.1979246801567844e-06,
      "loss": 1.139,
      "step": 891
    },
    {
      "epoch": 2.7633824670287046,
      "grad_norm": 2.375,
      "learning_rate": 2.187753561796097e-06,
      "loss": 1.1088,
      "step": 892
    },
    {
      "epoch": 2.7664856477889836,
      "grad_norm": 2.640625,
      "learning_rate": 2.1775994372624782e-06,
      "loss": 1.0801,
      "step": 893
    },
    {
      "epoch": 2.769588828549263,
      "grad_norm": 2.546875,
      "learning_rate": 2.1674623679152106e-06,
      "loss": 1.1265,
      "step": 894
    },
    {
      "epoch": 2.772692009309542,
      "grad_norm": 2.546875,
      "learning_rate": 2.1573424150105234e-06,
      "loss": 1.1449,
      "step": 895
    },
    {
      "epoch": 2.7757951900698217,
      "grad_norm": 2.59375,
      "learning_rate": 2.14723963970121e-06,
      "loss": 1.112,
      "step": 896
    },
    {
      "epoch": 2.7788983708301007,
      "grad_norm": 2.375,
      "learning_rate": 2.1371541030362675e-06,
      "loss": 1.111,
      "step": 897
    },
    {
      "epoch": 2.78200155159038,
      "grad_norm": 2.546875,
      "learning_rate": 2.127085865960516e-06,
      "loss": 1.1302,
      "step": 898
    },
    {
      "epoch": 2.7851047323506597,
      "grad_norm": 2.40625,
      "learning_rate": 2.117034989314245e-06,
      "loss": 1.1453,
      "step": 899
    },
    {
      "epoch": 2.7882079131109387,
      "grad_norm": 2.53125,
      "learning_rate": 2.107001533832837e-06,
      "loss": 1.1249,
      "step": 900
    },
    {
      "epoch": 2.791311093871218,
      "grad_norm": 2.5625,
      "learning_rate": 2.0969855601463966e-06,
      "loss": 1.114,
      "step": 901
    },
    {
      "epoch": 2.7944142746314973,
      "grad_norm": 2.578125,
      "learning_rate": 2.0869871287793986e-06,
      "loss": 1.1119,
      "step": 902
    },
    {
      "epoch": 2.7975174553917768,
      "grad_norm": 2.578125,
      "learning_rate": 2.0770063001503036e-06,
      "loss": 1.1589,
      "step": 903
    },
    {
      "epoch": 2.7975174553917768,
      "eval_loss": 1.9548412561416626,
      "eval_runtime": 35.6639,
      "eval_samples_per_second": 137.618,
      "eval_steps_per_second": 13.767,
      "step": 903
    },
    {
      "epoch": 2.800620636152056,
      "grad_norm": 2.515625,
      "learning_rate": 2.0670431345712092e-06,
      "loss": 1.0807,
      "step": 904
    },
    {
      "epoch": 2.8037238169123353,
      "grad_norm": 2.453125,
      "learning_rate": 2.057097692247477e-06,
      "loss": 1.107,
      "step": 905
    },
    {
      "epoch": 2.8068269976726143,
      "grad_norm": 2.453125,
      "learning_rate": 2.047170033277371e-06,
      "loss": 1.1204,
      "step": 906
    },
    {
      "epoch": 2.809930178432894,
      "grad_norm": 2.515625,
      "learning_rate": 2.0372602176516914e-06,
      "loss": 1.0887,
      "step": 907
    },
    {
      "epoch": 2.813033359193173,
      "grad_norm": 2.40625,
      "learning_rate": 2.027368305253418e-06,
      "loss": 1.1462,
      "step": 908
    },
    {
      "epoch": 2.8161365399534524,
      "grad_norm": 2.46875,
      "learning_rate": 2.0174943558573423e-06,
      "loss": 1.1308,
      "step": 909
    },
    {
      "epoch": 2.8192397207137314,
      "grad_norm": 2.421875,
      "learning_rate": 2.0076384291297134e-06,
      "loss": 1.0908,
      "step": 910
    },
    {
      "epoch": 2.822342901474011,
      "grad_norm": 2.46875,
      "learning_rate": 1.9978005846278693e-06,
      "loss": 1.1479,
      "step": 911
    },
    {
      "epoch": 2.82544608223429,
      "grad_norm": 2.46875,
      "learning_rate": 1.9879808817998793e-06,
      "loss": 1.1552,
      "step": 912
    },
    {
      "epoch": 2.8285492629945694,
      "grad_norm": 2.609375,
      "learning_rate": 1.97817937998419e-06,
      "loss": 1.1252,
      "step": 913
    },
    {
      "epoch": 2.831652443754849,
      "grad_norm": 2.4375,
      "learning_rate": 1.9683961384092624e-06,
      "loss": 1.112,
      "step": 914
    },
    {
      "epoch": 2.834755624515128,
      "grad_norm": 2.609375,
      "learning_rate": 1.958631216193214e-06,
      "loss": 1.1107,
      "step": 915
    },
    {
      "epoch": 2.8378588052754075,
      "grad_norm": 2.640625,
      "learning_rate": 1.9488846723434646e-06,
      "loss": 1.1531,
      "step": 916
    },
    {
      "epoch": 2.8409619860356865,
      "grad_norm": 2.453125,
      "learning_rate": 1.93915656575637e-06,
      "loss": 1.0911,
      "step": 917
    },
    {
      "epoch": 2.844065166795966,
      "grad_norm": 2.46875,
      "learning_rate": 1.9294469552168814e-06,
      "loss": 1.0753,
      "step": 918
    },
    {
      "epoch": 2.847168347556245,
      "grad_norm": 2.40625,
      "learning_rate": 1.9197558993981784e-06,
      "loss": 1.1207,
      "step": 919
    },
    {
      "epoch": 2.8502715283165245,
      "grad_norm": 2.65625,
      "learning_rate": 1.910083456861321e-06,
      "loss": 1.1572,
      "step": 920
    },
    {
      "epoch": 2.8533747090768036,
      "grad_norm": 2.4375,
      "learning_rate": 1.9004296860548887e-06,
      "loss": 1.1116,
      "step": 921
    },
    {
      "epoch": 2.856477889837083,
      "grad_norm": 2.59375,
      "learning_rate": 1.890794645314633e-06,
      "loss": 1.1176,
      "step": 922
    },
    {
      "epoch": 2.859581070597362,
      "grad_norm": 2.359375,
      "learning_rate": 1.881178392863125e-06,
      "loss": 1.1164,
      "step": 923
    },
    {
      "epoch": 2.8626842513576416,
      "grad_norm": 2.65625,
      "learning_rate": 1.871580986809402e-06,
      "loss": 1.1646,
      "step": 924
    },
    {
      "epoch": 2.8626842513576416,
      "eval_loss": 1.9641056060791016,
      "eval_runtime": 35.8345,
      "eval_samples_per_second": 136.963,
      "eval_steps_per_second": 13.702,
      "step": 924
    },
    {
      "epoch": 2.865787432117921,
      "grad_norm": 2.5,
      "learning_rate": 1.8620024851486174e-06,
      "loss": 1.1089,
      "step": 925
    },
    {
      "epoch": 2.8688906128782,
      "grad_norm": 2.453125,
      "learning_rate": 1.8524429457616832e-06,
      "loss": 1.0897,
      "step": 926
    },
    {
      "epoch": 2.871993793638479,
      "grad_norm": 2.46875,
      "learning_rate": 1.8429024264149336e-06,
      "loss": 1.1251,
      "step": 927
    },
    {
      "epoch": 2.8750969743987587,
      "grad_norm": 2.484375,
      "learning_rate": 1.8333809847597644e-06,
      "loss": 1.1078,
      "step": 928
    },
    {
      "epoch": 2.878200155159038,
      "grad_norm": 2.484375,
      "learning_rate": 1.8238786783322927e-06,
      "loss": 1.1094,
      "step": 929
    },
    {
      "epoch": 2.881303335919317,
      "grad_norm": 2.546875,
      "learning_rate": 1.8143955645530009e-06,
      "loss": 1.0751,
      "step": 930
    },
    {
      "epoch": 2.8844065166795967,
      "grad_norm": 2.546875,
      "learning_rate": 1.8049317007263933e-06,
      "loss": 1.1103,
      "step": 931
    },
    {
      "epoch": 2.8875096974398757,
      "grad_norm": 2.484375,
      "learning_rate": 1.7954871440406563e-06,
      "loss": 1.1042,
      "step": 932
    },
    {
      "epoch": 2.8906128782001552,
      "grad_norm": 2.421875,
      "learning_rate": 1.7860619515673034e-06,
      "loss": 1.1094,
      "step": 933
    },
    {
      "epoch": 2.8937160589604343,
      "grad_norm": 2.4375,
      "learning_rate": 1.7766561802608374e-06,
      "loss": 1.1285,
      "step": 934
    },
    {
      "epoch": 2.8968192397207138,
      "grad_norm": 2.640625,
      "learning_rate": 1.7672698869583964e-06,
      "loss": 1.0421,
      "step": 935
    },
    {
      "epoch": 2.8999224204809932,
      "grad_norm": 2.515625,
      "learning_rate": 1.7579031283794234e-06,
      "loss": 1.1379,
      "step": 936
    },
    {
      "epoch": 2.9030256012412723,
      "grad_norm": 2.5,
      "learning_rate": 1.748555961125315e-06,
      "loss": 1.0648,
      "step": 937
    },
    {
      "epoch": 2.9061287820015513,
      "grad_norm": 2.375,
      "learning_rate": 1.739228441679081e-06,
      "loss": 1.1566,
      "step": 938
    },
    {
      "epoch": 2.909231962761831,
      "grad_norm": 2.359375,
      "learning_rate": 1.729920626405005e-06,
      "loss": 1.0864,
      "step": 939
    },
    {
      "epoch": 2.9123351435221103,
      "grad_norm": 2.53125,
      "learning_rate": 1.7206325715483003e-06,
      "loss": 1.0605,
      "step": 940
    },
    {
      "epoch": 2.9154383242823894,
      "grad_norm": 2.453125,
      "learning_rate": 1.7113643332347708e-06,
      "loss": 1.1292,
      "step": 941
    },
    {
      "epoch": 2.918541505042669,
      "grad_norm": 2.515625,
      "learning_rate": 1.7021159674704768e-06,
      "loss": 1.1335,
      "step": 942
    },
    {
      "epoch": 2.921644685802948,
      "grad_norm": 2.515625,
      "learning_rate": 1.6928875301413916e-06,
      "loss": 1.123,
      "step": 943
    },
    {
      "epoch": 2.9247478665632274,
      "grad_norm": 2.375,
      "learning_rate": 1.6836790770130662e-06,
      "loss": 1.1007,
      "step": 944
    },
    {
      "epoch": 2.9278510473235064,
      "grad_norm": 2.484375,
      "learning_rate": 1.6744906637302865e-06,
      "loss": 1.0616,
      "step": 945
    },
    {
      "epoch": 2.9278510473235064,
      "eval_loss": 1.9717365503311157,
      "eval_runtime": 35.7663,
      "eval_samples_per_second": 137.224,
      "eval_steps_per_second": 13.728,
      "step": 945
    },
    {
      "epoch": 2.930954228083786,
      "grad_norm": 2.59375,
      "learning_rate": 1.665322345816746e-06,
      "loss": 1.14,
      "step": 946
    },
    {
      "epoch": 2.9340574088440654,
      "grad_norm": 2.5625,
      "learning_rate": 1.656174178674706e-06,
      "loss": 1.0957,
      "step": 947
    },
    {
      "epoch": 2.9371605896043445,
      "grad_norm": 2.484375,
      "learning_rate": 1.6470462175846608e-06,
      "loss": 1.0979,
      "step": 948
    },
    {
      "epoch": 2.9402637703646235,
      "grad_norm": 2.40625,
      "learning_rate": 1.637938517705001e-06,
      "loss": 1.1046,
      "step": 949
    },
    {
      "epoch": 2.943366951124903,
      "grad_norm": 2.40625,
      "learning_rate": 1.6288511340716833e-06,
      "loss": 1.1086,
      "step": 950
    },
    {
      "epoch": 2.9464701318851825,
      "grad_norm": 2.515625,
      "learning_rate": 1.6197841215978998e-06,
      "loss": 1.1187,
      "step": 951
    },
    {
      "epoch": 2.9495733126454615,
      "grad_norm": 2.46875,
      "learning_rate": 1.6107375350737437e-06,
      "loss": 1.1073,
      "step": 952
    },
    {
      "epoch": 2.952676493405741,
      "grad_norm": 2.453125,
      "learning_rate": 1.6017114291658786e-06,
      "loss": 1.1271,
      "step": 953
    },
    {
      "epoch": 2.95577967416602,
      "grad_norm": 2.296875,
      "learning_rate": 1.5927058584172034e-06,
      "loss": 1.0833,
      "step": 954
    },
    {
      "epoch": 2.9588828549262995,
      "grad_norm": 2.359375,
      "learning_rate": 1.583720877246533e-06,
      "loss": 1.0575,
      "step": 955
    },
    {
      "epoch": 2.9619860356865786,
      "grad_norm": 2.453125,
      "learning_rate": 1.5747565399482605e-06,
      "loss": 1.1433,
      "step": 956
    },
    {
      "epoch": 2.965089216446858,
      "grad_norm": 2.546875,
      "learning_rate": 1.5658129006920353e-06,
      "loss": 1.0694,
      "step": 957
    },
    {
      "epoch": 2.9681923972071376,
      "grad_norm": 2.453125,
      "learning_rate": 1.556890013522428e-06,
      "loss": 1.1197,
      "step": 958
    },
    {
      "epoch": 2.9712955779674166,
      "grad_norm": 2.4375,
      "learning_rate": 1.547987932358614e-06,
      "loss": 1.0486,
      "step": 959
    },
    {
      "epoch": 2.9743987587276957,
      "grad_norm": 2.5625,
      "learning_rate": 1.5391067109940366e-06,
      "loss": 1.1356,
      "step": 960
    },
    {
      "epoch": 2.977501939487975,
      "grad_norm": 2.46875,
      "learning_rate": 1.5302464030960924e-06,
      "loss": 1.1116,
      "step": 961
    },
    {
      "epoch": 2.9806051202482546,
      "grad_norm": 2.546875,
      "learning_rate": 1.5214070622058014e-06,
      "loss": 1.1158,
      "step": 962
    },
    {
      "epoch": 2.9837083010085337,
      "grad_norm": 2.421875,
      "learning_rate": 1.5125887417374814e-06,
      "loss": 1.0802,
      "step": 963
    },
    {
      "epoch": 2.986811481768813,
      "grad_norm": 2.4375,
      "learning_rate": 1.50379149497843e-06,
      "loss": 1.1045,
      "step": 964
    },
    {
      "epoch": 2.989914662529092,
      "grad_norm": 2.640625,
      "learning_rate": 1.495015375088602e-06,
      "loss": 1.1228,
      "step": 965
    },
    {
      "epoch": 2.9930178432893717,
      "grad_norm": 2.484375,
      "learning_rate": 1.4862604351002868e-06,
      "loss": 1.0812,
      "step": 966
    },
    {
      "epoch": 2.9930178432893717,
      "eval_loss": 1.9562337398529053,
      "eval_runtime": 35.9531,
      "eval_samples_per_second": 136.511,
      "eval_steps_per_second": 13.657,
      "step": 966
    },
    {
      "epoch": 2.9961210240496507,
      "grad_norm": 2.359375,
      "learning_rate": 1.477526727917784e-06,
      "loss": 1.1481,
      "step": 967
    },
    {
      "epoch": 2.9992242048099302,
      "grad_norm": 2.546875,
      "learning_rate": 1.4688143063170923e-06,
      "loss": 1.0925,
      "step": 968
    },
    {
      "epoch": 3.0,
      "grad_norm": 5.53125,
      "learning_rate": 1.4601232229455848e-06,
      "loss": 1.1229,
      "step": 969
    },
    {
      "epoch": 3.0031031807602795,
      "grad_norm": 2.53125,
      "learning_rate": 1.4514535303216893e-06,
      "loss": 1.0737,
      "step": 970
    },
    {
      "epoch": 3.0062063615205585,
      "grad_norm": 2.5,
      "learning_rate": 1.4428052808345776e-06,
      "loss": 1.0696,
      "step": 971
    },
    {
      "epoch": 3.009309542280838,
      "grad_norm": 2.4375,
      "learning_rate": 1.434178526743844e-06,
      "loss": 1.0481,
      "step": 972
    },
    {
      "epoch": 3.012412723041117,
      "grad_norm": 2.359375,
      "learning_rate": 1.4255733201791883e-06,
      "loss": 1.0884,
      "step": 973
    },
    {
      "epoch": 3.0155159038013966,
      "grad_norm": 2.484375,
      "learning_rate": 1.416989713140105e-06,
      "loss": 1.1118,
      "step": 974
    },
    {
      "epoch": 3.0186190845616756,
      "grad_norm": 2.421875,
      "learning_rate": 1.4084277574955667e-06,
      "loss": 1.1052,
      "step": 975
    },
    {
      "epoch": 3.021722265321955,
      "grad_norm": 2.546875,
      "learning_rate": 1.3998875049837141e-06,
      "loss": 1.1097,
      "step": 976
    },
    {
      "epoch": 3.024825446082234,
      "grad_norm": 2.453125,
      "learning_rate": 1.3913690072115328e-06,
      "loss": 1.0889,
      "step": 977
    },
    {
      "epoch": 3.0279286268425136,
      "grad_norm": 2.4375,
      "learning_rate": 1.3828723156545553e-06,
      "loss": 1.1205,
      "step": 978
    },
    {
      "epoch": 3.0310318076027927,
      "grad_norm": 2.390625,
      "learning_rate": 1.3743974816565441e-06,
      "loss": 1.1205,
      "step": 979
    },
    {
      "epoch": 3.034134988363072,
      "grad_norm": 2.5,
      "learning_rate": 1.3659445564291757e-06,
      "loss": 1.0956,
      "step": 980
    },
    {
      "epoch": 3.0372381691233516,
      "grad_norm": 2.546875,
      "learning_rate": 1.3575135910517412e-06,
      "loss": 1.1068,
      "step": 981
    },
    {
      "epoch": 3.0403413498836307,
      "grad_norm": 2.4375,
      "learning_rate": 1.3491046364708294e-06,
      "loss": 1.1011,
      "step": 982
    },
    {
      "epoch": 3.04344453064391,
      "grad_norm": 2.421875,
      "learning_rate": 1.3407177435000252e-06,
      "loss": 1.1397,
      "step": 983
    },
    {
      "epoch": 3.0465477114041892,
      "grad_norm": 2.46875,
      "learning_rate": 1.3323529628195986e-06,
      "loss": 1.1407,
      "step": 984
    },
    {
      "epoch": 3.0496508921644687,
      "grad_norm": 2.53125,
      "learning_rate": 1.3240103449762004e-06,
      "loss": 1.1292,
      "step": 985
    },
    {
      "epoch": 3.0527540729247478,
      "grad_norm": 2.53125,
      "learning_rate": 1.3156899403825534e-06,
      "loss": 1.1261,
      "step": 986
    },
    {
      "epoch": 3.0558572536850273,
      "grad_norm": 2.46875,
      "learning_rate": 1.3073917993171526e-06,
      "loss": 1.1582,
      "step": 987
    },
    {
      "epoch": 3.0558572536850273,
      "eval_loss": 1.9628909826278687,
      "eval_runtime": 35.7669,
      "eval_samples_per_second": 137.222,
      "eval_steps_per_second": 13.728,
      "step": 987
    },
    {
      "epoch": 3.0589604344453063,
      "grad_norm": 2.5625,
      "learning_rate": 1.2991159719239581e-06,
      "loss": 1.0701,
      "step": 988
    },
    {
      "epoch": 3.062063615205586,
      "grad_norm": 2.265625,
      "learning_rate": 1.2908625082120961e-06,
      "loss": 1.0675,
      "step": 989
    },
    {
      "epoch": 3.065166795965865,
      "grad_norm": 2.609375,
      "learning_rate": 1.2826314580555488e-06,
      "loss": 1.0605,
      "step": 990
    },
    {
      "epoch": 3.0682699767261443,
      "grad_norm": 2.40625,
      "learning_rate": 1.2744228711928585e-06,
      "loss": 1.0877,
      "step": 991
    },
    {
      "epoch": 3.0713731574864234,
      "grad_norm": 2.390625,
      "learning_rate": 1.2662367972268297e-06,
      "loss": 1.1251,
      "step": 992
    },
    {
      "epoch": 3.074476338246703,
      "grad_norm": 2.484375,
      "learning_rate": 1.2580732856242246e-06,
      "loss": 1.0899,
      "step": 993
    },
    {
      "epoch": 3.0775795190069823,
      "grad_norm": 2.28125,
      "learning_rate": 1.249932385715467e-06,
      "loss": 1.106,
      "step": 994
    },
    {
      "epoch": 3.0806826997672614,
      "grad_norm": 2.578125,
      "learning_rate": 1.2418141466943379e-06,
      "loss": 1.1513,
      "step": 995
    },
    {
      "epoch": 3.083785880527541,
      "grad_norm": 2.296875,
      "learning_rate": 1.233718617617689e-06,
      "loss": 1.049,
      "step": 996
    },
    {
      "epoch": 3.08688906128782,
      "grad_norm": 2.546875,
      "learning_rate": 1.2256458474051375e-06,
      "loss": 1.1077,
      "step": 997
    },
    {
      "epoch": 3.0899922420480994,
      "grad_norm": 2.34375,
      "learning_rate": 1.2175958848387764e-06,
      "loss": 1.0813,
      "step": 998
    },
    {
      "epoch": 3.0930954228083785,
      "grad_norm": 2.5,
      "learning_rate": 1.2095687785628719e-06,
      "loss": 1.1129,
      "step": 999
    },
    {
      "epoch": 3.096198603568658,
      "grad_norm": 2.453125,
      "learning_rate": 1.2015645770835765e-06,
      "loss": 1.1086,
      "step": 1000
    },
    {
      "epoch": 3.099301784328937,
      "grad_norm": 2.640625,
      "learning_rate": 1.193583328768635e-06,
      "loss": 1.074,
      "step": 1001
    },
    {
      "epoch": 3.1024049650892165,
      "grad_norm": 2.453125,
      "learning_rate": 1.1856250818470905e-06,
      "loss": 1.1035,
      "step": 1002
    },
    {
      "epoch": 3.1055081458494955,
      "grad_norm": 2.375,
      "learning_rate": 1.177689884408993e-06,
      "loss": 1.1173,
      "step": 1003
    },
    {
      "epoch": 3.108611326609775,
      "grad_norm": 2.484375,
      "learning_rate": 1.1697777844051105e-06,
      "loss": 1.0987,
      "step": 1004
    },
    {
      "epoch": 3.1117145073700545,
      "grad_norm": 2.421875,
      "learning_rate": 1.1618888296466341e-06,
      "loss": 1.0899,
      "step": 1005
    },
    {
      "epoch": 3.1148176881303336,
      "grad_norm": 2.4375,
      "learning_rate": 1.1540230678048969e-06,
      "loss": 1.1299,
      "step": 1006
    },
    {
      "epoch": 3.117920868890613,
      "grad_norm": 2.3125,
      "learning_rate": 1.1461805464110804e-06,
      "loss": 1.0438,
      "step": 1007
    },
    {
      "epoch": 3.121024049650892,
      "grad_norm": 2.40625,
      "learning_rate": 1.1383613128559306e-06,
      "loss": 1.1585,
      "step": 1008
    },
    {
      "epoch": 3.121024049650892,
      "eval_loss": 1.9647064208984375,
      "eval_runtime": 35.6825,
      "eval_samples_per_second": 137.547,
      "eval_steps_per_second": 13.76,
      "step": 1008
    },
    {
      "epoch": 3.1241272304111716,
      "grad_norm": 2.5,
      "learning_rate": 1.1305654143894674e-06,
      "loss": 1.0902,
      "step": 1009
    },
    {
      "epoch": 3.1272304111714506,
      "grad_norm": 2.53125,
      "learning_rate": 1.1227928981207004e-06,
      "loss": 1.127,
      "step": 1010
    },
    {
      "epoch": 3.13033359193173,
      "grad_norm": 2.5,
      "learning_rate": 1.1150438110173488e-06,
      "loss": 1.0759,
      "step": 1011
    },
    {
      "epoch": 3.133436772692009,
      "grad_norm": 2.328125,
      "learning_rate": 1.1073181999055538e-06,
      "loss": 1.0511,
      "step": 1012
    },
    {
      "epoch": 3.1365399534522886,
      "grad_norm": 2.5625,
      "learning_rate": 1.0996161114695952e-06,
      "loss": 1.1516,
      "step": 1013
    },
    {
      "epoch": 3.1396431342125677,
      "grad_norm": 2.375,
      "learning_rate": 1.0919375922516085e-06,
      "loss": 1.0596,
      "step": 1014
    },
    {
      "epoch": 3.142746314972847,
      "grad_norm": 2.453125,
      "learning_rate": 1.0842826886513074e-06,
      "loss": 1.0837,
      "step": 1015
    },
    {
      "epoch": 3.1458494957331267,
      "grad_norm": 2.609375,
      "learning_rate": 1.0766514469257006e-06,
      "loss": 1.0751,
      "step": 1016
    },
    {
      "epoch": 3.1489526764934057,
      "grad_norm": 2.515625,
      "learning_rate": 1.0690439131888142e-06,
      "loss": 1.0758,
      "step": 1017
    },
    {
      "epoch": 3.152055857253685,
      "grad_norm": 2.546875,
      "learning_rate": 1.0614601334114099e-06,
      "loss": 1.0676,
      "step": 1018
    },
    {
      "epoch": 3.1551590380139642,
      "grad_norm": 2.421875,
      "learning_rate": 1.0539001534207077e-06,
      "loss": 1.0657,
      "step": 1019
    },
    {
      "epoch": 3.1582622187742437,
      "grad_norm": 2.484375,
      "learning_rate": 1.0463640189001146e-06,
      "loss": 1.0519,
      "step": 1020
    },
    {
      "epoch": 3.161365399534523,
      "grad_norm": 2.359375,
      "learning_rate": 1.038851775388941e-06,
      "loss": 1.1265,
      "step": 1021
    },
    {
      "epoch": 3.1644685802948023,
      "grad_norm": 2.359375,
      "learning_rate": 1.031363468282132e-06,
      "loss": 1.0716,
      "step": 1022
    },
    {
      "epoch": 3.1675717610550813,
      "grad_norm": 2.515625,
      "learning_rate": 1.0238991428299855e-06,
      "loss": 1.1013,
      "step": 1023
    },
    {
      "epoch": 3.170674941815361,
      "grad_norm": 2.484375,
      "learning_rate": 1.016458844137887e-06,
      "loss": 1.1279,
      "step": 1024
    },
    {
      "epoch": 3.17377812257564,
      "grad_norm": 2.5625,
      "learning_rate": 1.0090426171660317e-06,
      "loss": 1.1111,
      "step": 1025
    },
    {
      "epoch": 3.1768813033359193,
      "grad_norm": 2.546875,
      "learning_rate": 1.0016505067291542e-06,
      "loss": 1.149,
      "step": 1026
    },
    {
      "epoch": 3.179984484096199,
      "grad_norm": 2.515625,
      "learning_rate": 9.942825574962595e-07,
      "loss": 1.1115,
      "step": 1027
    },
    {
      "epoch": 3.183087664856478,
      "grad_norm": 2.484375,
      "learning_rate": 9.869388139903495e-07,
      "loss": 1.0909,
      "step": 1028
    },
    {
      "epoch": 3.1861908456167574,
      "grad_norm": 2.390625,
      "learning_rate": 9.796193205881544e-07,
      "loss": 1.0461,
      "step": 1029
    },
    {
      "epoch": 3.1861908456167574,
      "eval_loss": 1.9532619714736938,
      "eval_runtime": 243.003,
      "eval_samples_per_second": 20.197,
      "eval_steps_per_second": 2.021,
      "step": 1029
    },
    {
      "epoch": 3.1892940263770364,
      "grad_norm": 2.59375,
      "learning_rate": 9.723241215198692e-07,
      "loss": 1.1169,
      "step": 1030
    },
    {
      "epoch": 3.192397207137316,
      "grad_norm": 2.53125,
      "learning_rate": 9.650532608688827e-07,
      "loss": 1.1153,
      "step": 1031
    },
    {
      "epoch": 3.195500387897595,
      "grad_norm": 2.46875,
      "learning_rate": 9.578067825715126e-07,
      "loss": 1.1158,
      "step": 1032
    },
    {
      "epoch": 3.1986035686578744,
      "grad_norm": 2.546875,
      "learning_rate": 9.505847304167337e-07,
      "loss": 1.0921,
      "step": 1033
    },
    {
      "epoch": 3.2017067494181535,
      "grad_norm": 2.34375,
      "learning_rate": 9.433871480459244e-07,
      "loss": 1.1112,
      "step": 1034
    },
    {
      "epoch": 3.204809930178433,
      "grad_norm": 2.421875,
      "learning_rate": 9.362140789525953e-07,
      "loss": 1.0881,
      "step": 1035
    },
    {
      "epoch": 3.207913110938712,
      "grad_norm": 2.4375,
      "learning_rate": 9.290655664821296e-07,
      "loss": 1.1347,
      "step": 1036
    },
    {
      "epoch": 3.2110162916989915,
      "grad_norm": 2.5,
      "learning_rate": 9.219416538315156e-07,
      "loss": 1.1015,
      "step": 1037
    },
    {
      "epoch": 3.214119472459271,
      "grad_norm": 2.671875,
      "learning_rate": 9.148423840490955e-07,
      "loss": 1.1204,
      "step": 1038
    },
    {
      "epoch": 3.21722265321955,
      "grad_norm": 2.46875,
      "learning_rate": 9.077678000342949e-07,
      "loss": 1.103,
      "step": 1039
    },
    {
      "epoch": 3.2203258339798295,
      "grad_norm": 2.546875,
      "learning_rate": 9.007179445373715e-07,
      "loss": 1.0721,
      "step": 1040
    },
    {
      "epoch": 3.2234290147401086,
      "grad_norm": 2.625,
      "learning_rate": 8.936928601591538e-07,
      "loss": 1.0708,
      "step": 1041
    },
    {
      "epoch": 3.226532195500388,
      "grad_norm": 2.4375,
      "learning_rate": 8.866925893507805e-07,
      "loss": 1.0539,
      "step": 1042
    },
    {
      "epoch": 3.229635376260667,
      "grad_norm": 2.46875,
      "learning_rate": 8.797171744134491e-07,
      "loss": 1.1152,
      "step": 1043
    },
    {
      "epoch": 3.2327385570209466,
      "grad_norm": 2.453125,
      "learning_rate": 8.727666574981597e-07,
      "loss": 1.0724,
      "step": 1044
    },
    {
      "epoch": 3.2358417377812256,
      "grad_norm": 2.390625,
      "learning_rate": 8.658410806054568e-07,
      "loss": 1.1224,
      "step": 1045
    },
    {
      "epoch": 3.238944918541505,
      "grad_norm": 2.421875,
      "learning_rate": 8.589404855851763e-07,
      "loss": 1.1265,
      "step": 1046
    },
    {
      "epoch": 3.242048099301784,
      "grad_norm": 2.578125,
      "learning_rate": 8.520649141361959e-07,
      "loss": 1.1131,
      "step": 1047
    },
    {
      "epoch": 3.2451512800620637,
      "grad_norm": 2.390625,
      "learning_rate": 8.45214407806182e-07,
      "loss": 1.0743,
      "step": 1048
    },
    {
      "epoch": 3.2482544608223427,
      "grad_norm": 2.4375,
      "learning_rate": 8.383890079913326e-07,
      "loss": 1.0102,
      "step": 1049
    },
    {
      "epoch": 3.251357641582622,
      "grad_norm": 2.5625,
      "learning_rate": 8.315887559361391e-07,
      "loss": 1.1227,
      "step": 1050
    },
    {
      "epoch": 3.251357641582622,
      "eval_loss": 1.9681098461151123,
      "eval_runtime": 35.8524,
      "eval_samples_per_second": 136.895,
      "eval_steps_per_second": 13.695,
      "step": 1050
    },
    {
      "epoch": 3.2544608223429012,
      "grad_norm": 2.421875,
      "learning_rate": 8.248136927331241e-07,
      "loss": 1.0888,
      "step": 1051
    },
    {
      "epoch": 3.2575640031031807,
      "grad_norm": 2.453125,
      "learning_rate": 8.180638593226021e-07,
      "loss": 1.1358,
      "step": 1052
    },
    {
      "epoch": 3.26066718386346,
      "grad_norm": 2.4375,
      "learning_rate": 8.113392964924294e-07,
      "loss": 1.1092,
      "step": 1053
    },
    {
      "epoch": 3.2637703646237393,
      "grad_norm": 2.5,
      "learning_rate": 8.046400448777575e-07,
      "loss": 1.1659,
      "step": 1054
    },
    {
      "epoch": 3.2668735453840188,
      "grad_norm": 2.5625,
      "learning_rate": 7.979661449607845e-07,
      "loss": 1.118,
      "step": 1055
    },
    {
      "epoch": 3.269976726144298,
      "grad_norm": 2.5625,
      "learning_rate": 7.913176370705166e-07,
      "loss": 1.1223,
      "step": 1056
    },
    {
      "epoch": 3.2730799069045773,
      "grad_norm": 2.65625,
      "learning_rate": 7.846945613825213e-07,
      "loss": 1.1299,
      "step": 1057
    },
    {
      "epoch": 3.2761830876648563,
      "grad_norm": 2.578125,
      "learning_rate": 7.780969579186814e-07,
      "loss": 1.1033,
      "step": 1058
    },
    {
      "epoch": 3.279286268425136,
      "grad_norm": 2.5625,
      "learning_rate": 7.715248665469588e-07,
      "loss": 1.0752,
      "step": 1059
    },
    {
      "epoch": 3.282389449185415,
      "grad_norm": 2.53125,
      "learning_rate": 7.649783269811523e-07,
      "loss": 1.1143,
      "step": 1060
    },
    {
      "epoch": 3.2854926299456944,
      "grad_norm": 2.5625,
      "learning_rate": 7.58457378780652e-07,
      "loss": 1.0464,
      "step": 1061
    },
    {
      "epoch": 3.2885958107059734,
      "grad_norm": 2.46875,
      "learning_rate": 7.519620613502082e-07,
      "loss": 1.1117,
      "step": 1062
    },
    {
      "epoch": 3.291698991466253,
      "grad_norm": 2.546875,
      "learning_rate": 7.45492413939689e-07,
      "loss": 1.0911,
      "step": 1063
    },
    {
      "epoch": 3.2948021722265324,
      "grad_norm": 2.515625,
      "learning_rate": 7.390484756438449e-07,
      "loss": 1.1047,
      "step": 1064
    },
    {
      "epoch": 3.2979053529868114,
      "grad_norm": 2.453125,
      "learning_rate": 7.326302854020678e-07,
      "loss": 1.0991,
      "step": 1065
    },
    {
      "epoch": 3.301008533747091,
      "grad_norm": 2.453125,
      "learning_rate": 7.26237881998163e-07,
      "loss": 1.1368,
      "step": 1066
    },
    {
      "epoch": 3.30411171450737,
      "grad_norm": 2.421875,
      "learning_rate": 7.198713040601112e-07,
      "loss": 1.1097,
      "step": 1067
    },
    {
      "epoch": 3.3072148952676494,
      "grad_norm": 2.546875,
      "learning_rate": 7.135305900598321e-07,
      "loss": 1.1418,
      "step": 1068
    },
    {
      "epoch": 3.3103180760279285,
      "grad_norm": 2.59375,
      "learning_rate": 7.072157783129585e-07,
      "loss": 1.1065,
      "step": 1069
    },
    {
      "epoch": 3.313421256788208,
      "grad_norm": 2.484375,
      "learning_rate": 7.009269069785973e-07,
      "loss": 1.1047,
      "step": 1070
    },
    {
      "epoch": 3.316524437548487,
      "grad_norm": 2.5,
      "learning_rate": 6.946640140591071e-07,
      "loss": 1.1045,
      "step": 1071
    },
    {
      "epoch": 3.316524437548487,
      "eval_loss": 1.955759048461914,
      "eval_runtime": 35.8214,
      "eval_samples_per_second": 137.013,
      "eval_steps_per_second": 13.707,
      "step": 1071
    },
    {
      "epoch": 3.3196276183087665,
      "grad_norm": 2.453125,
      "learning_rate": 6.884271373998608e-07,
      "loss": 1.1084,
      "step": 1072
    },
    {
      "epoch": 3.3227307990690456,
      "grad_norm": 2.5,
      "learning_rate": 6.822163146890242e-07,
      "loss": 1.1018,
      "step": 1073
    },
    {
      "epoch": 3.325833979829325,
      "grad_norm": 2.515625,
      "learning_rate": 6.760315834573194e-07,
      "loss": 1.138,
      "step": 1074
    },
    {
      "epoch": 3.3289371605896045,
      "grad_norm": 2.390625,
      "learning_rate": 6.698729810778065e-07,
      "loss": 1.1032,
      "step": 1075
    },
    {
      "epoch": 3.3320403413498836,
      "grad_norm": 2.46875,
      "learning_rate": 6.637405447656542e-07,
      "loss": 1.1216,
      "step": 1076
    },
    {
      "epoch": 3.335143522110163,
      "grad_norm": 2.546875,
      "learning_rate": 6.576343115779155e-07,
      "loss": 1.0576,
      "step": 1077
    },
    {
      "epoch": 3.338246702870442,
      "grad_norm": 2.546875,
      "learning_rate": 6.515543184133e-07,
      "loss": 1.0866,
      "step": 1078
    },
    {
      "epoch": 3.3413498836307216,
      "grad_norm": 2.390625,
      "learning_rate": 6.45500602011957e-07,
      "loss": 1.1177,
      "step": 1079
    },
    {
      "epoch": 3.3444530643910007,
      "grad_norm": 2.578125,
      "learning_rate": 6.394731989552494e-07,
      "loss": 1.1264,
      "step": 1080
    },
    {
      "epoch": 3.34755624515128,
      "grad_norm": 2.53125,
      "learning_rate": 6.334721456655363e-07,
      "loss": 1.0791,
      "step": 1081
    },
    {
      "epoch": 3.350659425911559,
      "grad_norm": 2.375,
      "learning_rate": 6.274974784059495e-07,
      "loss": 1.121,
      "step": 1082
    },
    {
      "epoch": 3.3537626066718387,
      "grad_norm": 2.4375,
      "learning_rate": 6.215492332801725e-07,
      "loss": 1.0811,
      "step": 1083
    },
    {
      "epoch": 3.3568657874321177,
      "grad_norm": 2.421875,
      "learning_rate": 6.156274462322292e-07,
      "loss": 1.1112,
      "step": 1084
    },
    {
      "epoch": 3.359968968192397,
      "grad_norm": 2.671875,
      "learning_rate": 6.097321530462607e-07,
      "loss": 1.0571,
      "step": 1085
    },
    {
      "epoch": 3.3630721489526767,
      "grad_norm": 2.546875,
      "learning_rate": 6.038633893463125e-07,
      "loss": 1.0917,
      "step": 1086
    },
    {
      "epoch": 3.3661753297129557,
      "grad_norm": 2.375,
      "learning_rate": 5.980211905961181e-07,
      "loss": 1.0542,
      "step": 1087
    },
    {
      "epoch": 3.3692785104732352,
      "grad_norm": 2.421875,
      "learning_rate": 5.922055920988818e-07,
      "loss": 1.1087,
      "step": 1088
    },
    {
      "epoch": 3.3723816912335143,
      "grad_norm": 2.421875,
      "learning_rate": 5.86416628997069e-07,
      "loss": 1.1278,
      "step": 1089
    },
    {
      "epoch": 3.3754848719937938,
      "grad_norm": 2.703125,
      "learning_rate": 5.806543362721945e-07,
      "loss": 1.1085,
      "step": 1090
    },
    {
      "epoch": 3.378588052754073,
      "grad_norm": 2.40625,
      "learning_rate": 5.74918748744609e-07,
      "loss": 1.0801,
      "step": 1091
    },
    {
      "epoch": 3.3816912335143523,
      "grad_norm": 2.625,
      "learning_rate": 5.692099010732893e-07,
      "loss": 1.1241,
      "step": 1092
    },
    {
      "epoch": 3.3816912335143523,
      "eval_loss": 1.9659039974212646,
      "eval_runtime": 35.8224,
      "eval_samples_per_second": 137.009,
      "eval_steps_per_second": 13.707,
      "step": 1092
    },
    {
      "epoch": 3.3847944142746313,
      "grad_norm": 2.484375,
      "learning_rate": 5.635278277556267e-07,
      "loss": 1.1254,
      "step": 1093
    },
    {
      "epoch": 3.387897595034911,
      "grad_norm": 2.609375,
      "learning_rate": 5.578725631272225e-07,
      "loss": 1.0805,
      "step": 1094
    },
    {
      "epoch": 3.39100077579519,
      "grad_norm": 2.5,
      "learning_rate": 5.522441413616786e-07,
      "loss": 1.0751,
      "step": 1095
    },
    {
      "epoch": 3.3941039565554694,
      "grad_norm": 2.515625,
      "learning_rate": 5.466425964703914e-07,
      "loss": 1.1418,
      "step": 1096
    },
    {
      "epoch": 3.397207137315749,
      "grad_norm": 2.578125,
      "learning_rate": 5.410679623023446e-07,
      "loss": 1.1077,
      "step": 1097
    },
    {
      "epoch": 3.400310318076028,
      "grad_norm": 2.484375,
      "learning_rate": 5.355202725439046e-07,
      "loss": 1.1064,
      "step": 1098
    },
    {
      "epoch": 3.4034134988363074,
      "grad_norm": 2.578125,
      "learning_rate": 5.29999560718622e-07,
      "loss": 1.1506,
      "step": 1099
    },
    {
      "epoch": 3.4065166795965864,
      "grad_norm": 2.625,
      "learning_rate": 5.245058601870223e-07,
      "loss": 1.1384,
      "step": 1100
    },
    {
      "epoch": 3.409619860356866,
      "grad_norm": 2.421875,
      "learning_rate": 5.190392041464098e-07,
      "loss": 1.0852,
      "step": 1101
    },
    {
      "epoch": 3.412723041117145,
      "grad_norm": 2.671875,
      "learning_rate": 5.135996256306619e-07,
      "loss": 1.1237,
      "step": 1102
    },
    {
      "epoch": 3.4158262218774245,
      "grad_norm": 2.53125,
      "learning_rate": 5.081871575100333e-07,
      "loss": 1.0518,
      "step": 1103
    },
    {
      "epoch": 3.4189294026377035,
      "grad_norm": 2.484375,
      "learning_rate": 5.028018324909561e-07,
      "loss": 1.088,
      "step": 1104
    },
    {
      "epoch": 3.422032583397983,
      "grad_norm": 2.46875,
      "learning_rate": 4.974436831158441e-07,
      "loss": 1.0793,
      "step": 1105
    },
    {
      "epoch": 3.425135764158262,
      "grad_norm": 2.375,
      "learning_rate": 4.921127417628896e-07,
      "loss": 1.0585,
      "step": 1106
    },
    {
      "epoch": 3.4282389449185415,
      "grad_norm": 2.53125,
      "learning_rate": 4.868090406458776e-07,
      "loss": 1.0814,
      "step": 1107
    },
    {
      "epoch": 3.431342125678821,
      "grad_norm": 2.5625,
      "learning_rate": 4.815326118139813e-07,
      "loss": 1.0919,
      "step": 1108
    },
    {
      "epoch": 3.4344453064391,
      "grad_norm": 2.53125,
      "learning_rate": 4.7628348715157646e-07,
      "loss": 1.1196,
      "step": 1109
    },
    {
      "epoch": 3.4375484871993796,
      "grad_norm": 2.46875,
      "learning_rate": 4.7106169837804526e-07,
      "loss": 1.0955,
      "step": 1110
    },
    {
      "epoch": 3.4406516679596586,
      "grad_norm": 2.46875,
      "learning_rate": 4.658672770475825e-07,
      "loss": 1.091,
      "step": 1111
    },
    {
      "epoch": 3.443754848719938,
      "grad_norm": 2.375,
      "learning_rate": 4.60700254549008e-07,
      "loss": 1.1002,
      "step": 1112
    },
    {
      "epoch": 3.446858029480217,
      "grad_norm": 2.515625,
      "learning_rate": 4.5556066210557747e-07,
      "loss": 1.0713,
      "step": 1113
    },
    {
      "epoch": 3.446858029480217,
      "eval_loss": 1.9555165767669678,
      "eval_runtime": 35.8018,
      "eval_samples_per_second": 137.088,
      "eval_steps_per_second": 13.714,
      "step": 1113
    },
    {
      "epoch": 3.4499612102404966,
      "grad_norm": 2.484375,
      "learning_rate": 4.5044853077479134e-07,
      "loss": 1.1026,
      "step": 1114
    },
    {
      "epoch": 3.4530643910007757,
      "grad_norm": 2.515625,
      "learning_rate": 4.453638914482084e-07,
      "loss": 1.1472,
      "step": 1115
    },
    {
      "epoch": 3.456167571761055,
      "grad_norm": 2.421875,
      "learning_rate": 4.4030677485125906e-07,
      "loss": 1.0861,
      "step": 1116
    },
    {
      "epoch": 3.459270752521334,
      "grad_norm": 2.453125,
      "learning_rate": 4.3527721154305703e-07,
      "loss": 1.1309,
      "step": 1117
    },
    {
      "epoch": 3.4623739332816137,
      "grad_norm": 2.421875,
      "learning_rate": 4.302752319162212e-07,
      "loss": 1.0781,
      "step": 1118
    },
    {
      "epoch": 3.465477114041893,
      "grad_norm": 2.421875,
      "learning_rate": 4.2530086619668565e-07,
      "loss": 1.1006,
      "step": 1119
    },
    {
      "epoch": 3.4685802948021722,
      "grad_norm": 2.640625,
      "learning_rate": 4.203541444435211e-07,
      "loss": 1.0945,
      "step": 1120
    },
    {
      "epoch": 3.4716834755624513,
      "grad_norm": 2.515625,
      "learning_rate": 4.1543509654874745e-07,
      "loss": 1.0755,
      "step": 1121
    },
    {
      "epoch": 3.4747866563227308,
      "grad_norm": 2.4375,
      "learning_rate": 4.105437522371619e-07,
      "loss": 1.0962,
      "step": 1122
    },
    {
      "epoch": 3.4778898370830102,
      "grad_norm": 2.5,
      "learning_rate": 4.0568014106615205e-07,
      "loss": 1.1154,
      "step": 1123
    },
    {
      "epoch": 3.4809930178432893,
      "grad_norm": 2.375,
      "learning_rate": 4.008442924255207e-07,
      "loss": 1.1284,
      "step": 1124
    },
    {
      "epoch": 3.484096198603569,
      "grad_norm": 2.390625,
      "learning_rate": 3.9603623553730597e-07,
      "loss": 1.0838,
      "step": 1125
    },
    {
      "epoch": 3.487199379363848,
      "grad_norm": 2.515625,
      "learning_rate": 3.9125599945560866e-07,
      "loss": 1.0773,
      "step": 1126
    },
    {
      "epoch": 3.4903025601241273,
      "grad_norm": 2.546875,
      "learning_rate": 3.865036130664107e-07,
      "loss": 1.0938,
      "step": 1127
    },
    {
      "epoch": 3.4934057408844064,
      "grad_norm": 2.578125,
      "learning_rate": 3.8177910508740703e-07,
      "loss": 1.0808,
      "step": 1128
    },
    {
      "epoch": 3.496508921644686,
      "grad_norm": 2.53125,
      "learning_rate": 3.770825040678294e-07,
      "loss": 1.0891,
      "step": 1129
    },
    {
      "epoch": 3.499612102404965,
      "grad_norm": 2.5,
      "learning_rate": 3.724138383882697e-07,
      "loss": 1.0656,
      "step": 1130
    },
    {
      "epoch": 3.5027152831652444,
      "grad_norm": 2.40625,
      "learning_rate": 3.677731362605158e-07,
      "loss": 1.0524,
      "step": 1131
    },
    {
      "epoch": 3.5058184639255234,
      "grad_norm": 2.5,
      "learning_rate": 3.631604257273774e-07,
      "loss": 1.1188,
      "step": 1132
    },
    {
      "epoch": 3.508921644685803,
      "grad_norm": 2.515625,
      "learning_rate": 3.585757346625157e-07,
      "loss": 1.0609,
      "step": 1133
    },
    {
      "epoch": 3.5120248254460824,
      "grad_norm": 2.46875,
      "learning_rate": 3.5401909077027506e-07,
      "loss": 1.056,
      "step": 1134
    },
    {
      "epoch": 3.5120248254460824,
      "eval_loss": 1.9670829772949219,
      "eval_runtime": 35.732,
      "eval_samples_per_second": 137.356,
      "eval_steps_per_second": 13.741,
      "step": 1134
    },
    {
      "epoch": 3.5151280062063615,
      "grad_norm": 2.53125,
      "learning_rate": 3.4949052158551875e-07,
      "loss": 1.0648,
      "step": 1135
    },
    {
      "epoch": 3.518231186966641,
      "grad_norm": 2.4375,
      "learning_rate": 3.4499005447346024e-07,
      "loss": 1.1,
      "step": 1136
    },
    {
      "epoch": 3.52133436772692,
      "grad_norm": 2.390625,
      "learning_rate": 3.4051771662949483e-07,
      "loss": 1.1277,
      "step": 1137
    },
    {
      "epoch": 3.5244375484871995,
      "grad_norm": 2.453125,
      "learning_rate": 3.360735350790428e-07,
      "loss": 1.068,
      "step": 1138
    },
    {
      "epoch": 3.5275407292474785,
      "grad_norm": 2.484375,
      "learning_rate": 3.3165753667737776e-07,
      "loss": 1.067,
      "step": 1139
    },
    {
      "epoch": 3.530643910007758,
      "grad_norm": 2.515625,
      "learning_rate": 3.272697481094711e-07,
      "loss": 1.0651,
      "step": 1140
    },
    {
      "epoch": 3.5337470907680375,
      "grad_norm": 2.484375,
      "learning_rate": 3.229101958898273e-07,
      "loss": 1.0714,
      "step": 1141
    },
    {
      "epoch": 3.5368502715283165,
      "grad_norm": 2.4375,
      "learning_rate": 3.1857890636232405e-07,
      "loss": 1.1078,
      "step": 1142
    },
    {
      "epoch": 3.5399534522885956,
      "grad_norm": 2.375,
      "learning_rate": 3.142759057000533e-07,
      "loss": 1.0906,
      "step": 1143
    },
    {
      "epoch": 3.543056633048875,
      "grad_norm": 2.546875,
      "learning_rate": 3.100012199051627e-07,
      "loss": 1.1269,
      "step": 1144
    },
    {
      "epoch": 3.5461598138091546,
      "grad_norm": 2.53125,
      "learning_rate": 3.057548748087008e-07,
      "loss": 1.0772,
      "step": 1145
    },
    {
      "epoch": 3.5492629945694336,
      "grad_norm": 2.609375,
      "learning_rate": 3.015368960704584e-07,
      "loss": 1.1014,
      "step": 1146
    },
    {
      "epoch": 3.552366175329713,
      "grad_norm": 2.453125,
      "learning_rate": 2.973473091788121e-07,
      "loss": 1.0721,
      "step": 1147
    },
    {
      "epoch": 3.555469356089992,
      "grad_norm": 2.40625,
      "learning_rate": 2.931861394505764e-07,
      "loss": 1.1104,
      "step": 1148
    },
    {
      "epoch": 3.5585725368502716,
      "grad_norm": 2.421875,
      "learning_rate": 2.890534120308425e-07,
      "loss": 1.1192,
      "step": 1149
    },
    {
      "epoch": 3.5616757176105507,
      "grad_norm": 2.484375,
      "learning_rate": 2.8494915189283325e-07,
      "loss": 1.1091,
      "step": 1150
    },
    {
      "epoch": 3.56477889837083,
      "grad_norm": 2.390625,
      "learning_rate": 2.8087338383774866e-07,
      "loss": 1.0715,
      "step": 1151
    },
    {
      "epoch": 3.5678820791311097,
      "grad_norm": 2.453125,
      "learning_rate": 2.7682613249461866e-07,
      "loss": 1.0827,
      "step": 1152
    },
    {
      "epoch": 3.5709852598913887,
      "grad_norm": 2.359375,
      "learning_rate": 2.728074223201488e-07,
      "loss": 1.0728,
      "step": 1153
    },
    {
      "epoch": 3.5740884406516678,
      "grad_norm": 2.53125,
      "learning_rate": 2.6881727759857925e-07,
      "loss": 1.0884,
      "step": 1154
    },
    {
      "epoch": 3.5771916214119472,
      "grad_norm": 2.484375,
      "learning_rate": 2.6485572244153546e-07,
      "loss": 1.093,
      "step": 1155
    },
    {
      "epoch": 3.5771916214119472,
      "eval_loss": 1.9634618759155273,
      "eval_runtime": 35.7669,
      "eval_samples_per_second": 137.222,
      "eval_steps_per_second": 13.728,
      "step": 1155
    },
    {
      "epoch": 3.5802948021722267,
      "grad_norm": 2.515625,
      "learning_rate": 2.6092278078788004e-07,
      "loss": 1.1575,
      "step": 1156
    },
    {
      "epoch": 3.5833979829325058,
      "grad_norm": 2.671875,
      "learning_rate": 2.570184764035705e-07,
      "loss": 1.1287,
      "step": 1157
    },
    {
      "epoch": 3.5865011636927853,
      "grad_norm": 2.546875,
      "learning_rate": 2.5314283288151556e-07,
      "loss": 1.0785,
      "step": 1158
    },
    {
      "epoch": 3.5896043444530643,
      "grad_norm": 2.53125,
      "learning_rate": 2.4929587364143195e-07,
      "loss": 1.1155,
      "step": 1159
    },
    {
      "epoch": 3.592707525213344,
      "grad_norm": 2.515625,
      "learning_rate": 2.4547762192970394e-07,
      "loss": 1.0768,
      "step": 1160
    },
    {
      "epoch": 3.595810705973623,
      "grad_norm": 2.421875,
      "learning_rate": 2.416881008192429e-07,
      "loss": 1.0985,
      "step": 1161
    },
    {
      "epoch": 3.5989138867339023,
      "grad_norm": 2.46875,
      "learning_rate": 2.3792733320934348e-07,
      "loss": 1.0813,
      "step": 1162
    },
    {
      "epoch": 3.6020170674941814,
      "grad_norm": 2.515625,
      "learning_rate": 2.3419534182555215e-07,
      "loss": 1.0998,
      "step": 1163
    },
    {
      "epoch": 3.605120248254461,
      "grad_norm": 2.484375,
      "learning_rate": 2.3049214921952612e-07,
      "loss": 1.1112,
      "step": 1164
    },
    {
      "epoch": 3.60822342901474,
      "grad_norm": 2.390625,
      "learning_rate": 2.2681777776889736e-07,
      "loss": 1.116,
      "step": 1165
    },
    {
      "epoch": 3.6113266097750194,
      "grad_norm": 2.484375,
      "learning_rate": 2.2317224967713602e-07,
      "loss": 1.0912,
      "step": 1166
    },
    {
      "epoch": 3.614429790535299,
      "grad_norm": 2.46875,
      "learning_rate": 2.1955558697341894e-07,
      "loss": 1.0826,
      "step": 1167
    },
    {
      "epoch": 3.617532971295578,
      "grad_norm": 2.515625,
      "learning_rate": 2.1596781151249524e-07,
      "loss": 1.1068,
      "step": 1168
    },
    {
      "epoch": 3.620636152055857,
      "grad_norm": 2.546875,
      "learning_rate": 2.1240894497455478e-07,
      "loss": 1.1126,
      "step": 1169
    },
    {
      "epoch": 3.6237393328161365,
      "grad_norm": 2.515625,
      "learning_rate": 2.088790088650977e-07,
      "loss": 1.0756,
      "step": 1170
    },
    {
      "epoch": 3.626842513576416,
      "grad_norm": 2.46875,
      "learning_rate": 2.053780245147996e-07,
      "loss": 1.0842,
      "step": 1171
    },
    {
      "epoch": 3.629945694336695,
      "grad_norm": 2.78125,
      "learning_rate": 2.0190601307939094e-07,
      "loss": 1.1086,
      "step": 1172
    },
    {
      "epoch": 3.6330488750969745,
      "grad_norm": 2.484375,
      "learning_rate": 1.984629955395223e-07,
      "loss": 1.0931,
      "step": 1173
    },
    {
      "epoch": 3.6361520558572535,
      "grad_norm": 2.4375,
      "learning_rate": 1.9504899270064105e-07,
      "loss": 1.0801,
      "step": 1174
    },
    {
      "epoch": 3.639255236617533,
      "grad_norm": 2.328125,
      "learning_rate": 1.916640251928653e-07,
      "loss": 1.0895,
      "step": 1175
    },
    {
      "epoch": 3.642358417377812,
      "grad_norm": 2.515625,
      "learning_rate": 1.8830811347085697e-07,
      "loss": 1.0508,
      "step": 1176
    },
    {
      "epoch": 3.642358417377812,
      "eval_loss": 1.9662879705429077,
      "eval_runtime": 35.7888,
      "eval_samples_per_second": 137.138,
      "eval_steps_per_second": 13.719,
      "step": 1176
    },
    {
      "epoch": 3.6454615981380916,
      "grad_norm": 2.59375,
      "learning_rate": 1.849812778137e-07,
      "loss": 1.1187,
      "step": 1177
    },
    {
      "epoch": 3.648564778898371,
      "grad_norm": 2.53125,
      "learning_rate": 1.8168353832477948e-07,
      "loss": 1.1027,
      "step": 1178
    },
    {
      "epoch": 3.65166795965865,
      "grad_norm": 2.5625,
      "learning_rate": 1.7841491493165607e-07,
      "loss": 1.1628,
      "step": 1179
    },
    {
      "epoch": 3.654771140418929,
      "grad_norm": 2.625,
      "learning_rate": 1.7517542738595071e-07,
      "loss": 1.0819,
      "step": 1180
    },
    {
      "epoch": 3.6578743211792086,
      "grad_norm": 2.5625,
      "learning_rate": 1.7196509526321847e-07,
      "loss": 1.0802,
      "step": 1181
    },
    {
      "epoch": 3.660977501939488,
      "grad_norm": 2.453125,
      "learning_rate": 1.687839379628381e-07,
      "loss": 1.0775,
      "step": 1182
    },
    {
      "epoch": 3.664080682699767,
      "grad_norm": 2.515625,
      "learning_rate": 1.6563197470788827e-07,
      "loss": 1.106,
      "step": 1183
    },
    {
      "epoch": 3.6671838634600467,
      "grad_norm": 2.484375,
      "learning_rate": 1.6250922454503548e-07,
      "loss": 1.093,
      "step": 1184
    },
    {
      "epoch": 3.6702870442203257,
      "grad_norm": 2.53125,
      "learning_rate": 1.5941570634441683e-07,
      "loss": 1.092,
      "step": 1185
    },
    {
      "epoch": 3.673390224980605,
      "grad_norm": 2.40625,
      "learning_rate": 1.5635143879952575e-07,
      "loss": 1.0813,
      "step": 1186
    },
    {
      "epoch": 3.6764934057408842,
      "grad_norm": 2.578125,
      "learning_rate": 1.5331644042710202e-07,
      "loss": 1.1016,
      "step": 1187
    },
    {
      "epoch": 3.6795965865011637,
      "grad_norm": 2.46875,
      "learning_rate": 1.5031072956701698e-07,
      "loss": 1.1133,
      "step": 1188
    },
    {
      "epoch": 3.682699767261443,
      "grad_norm": 2.609375,
      "learning_rate": 1.47334324382164e-07,
      "loss": 1.1154,
      "step": 1189
    },
    {
      "epoch": 3.6858029480217223,
      "grad_norm": 2.578125,
      "learning_rate": 1.4438724285834593e-07,
      "loss": 1.1042,
      "step": 1190
    },
    {
      "epoch": 3.6889061287820013,
      "grad_norm": 2.328125,
      "learning_rate": 1.414695028041718e-07,
      "loss": 1.048,
      "step": 1191
    },
    {
      "epoch": 3.692009309542281,
      "grad_norm": 2.484375,
      "learning_rate": 1.3858112185094418e-07,
      "loss": 1.0658,
      "step": 1192
    },
    {
      "epoch": 3.6951124903025603,
      "grad_norm": 2.484375,
      "learning_rate": 1.3572211745255692e-07,
      "loss": 1.1317,
      "step": 1193
    },
    {
      "epoch": 3.6982156710628393,
      "grad_norm": 2.453125,
      "learning_rate": 1.328925068853837e-07,
      "loss": 1.0854,
      "step": 1194
    },
    {
      "epoch": 3.701318851823119,
      "grad_norm": 2.53125,
      "learning_rate": 1.3009230724818134e-07,
      "loss": 1.0903,
      "step": 1195
    },
    {
      "epoch": 3.704422032583398,
      "grad_norm": 2.53125,
      "learning_rate": 1.273215354619789e-07,
      "loss": 1.0948,
      "step": 1196
    },
    {
      "epoch": 3.7075252133436774,
      "grad_norm": 2.53125,
      "learning_rate": 1.2458020826998152e-07,
      "loss": 1.1521,
      "step": 1197
    },
    {
      "epoch": 3.7075252133436774,
      "eval_loss": 1.9763387441635132,
      "eval_runtime": 35.8027,
      "eval_samples_per_second": 137.085,
      "eval_steps_per_second": 13.714,
      "step": 1197
    },
    {
      "epoch": 3.7106283941039564,
      "grad_norm": 2.5,
      "learning_rate": 1.2186834223746612e-07,
      "loss": 1.0798,
      "step": 1198
    },
    {
      "epoch": 3.713731574864236,
      "grad_norm": 2.484375,
      "learning_rate": 1.1918595375168096e-07,
      "loss": 1.0509,
      "step": 1199
    },
    {
      "epoch": 3.7168347556245154,
      "grad_norm": 2.46875,
      "learning_rate": 1.1653305902174894e-07,
      "loss": 1.0835,
      "step": 1200
    },
    {
      "epoch": 3.7199379363847944,
      "grad_norm": 2.5625,
      "learning_rate": 1.1390967407856612e-07,
      "loss": 1.0957,
      "step": 1201
    },
    {
      "epoch": 3.7230411171450735,
      "grad_norm": 2.4375,
      "learning_rate": 1.113158147747101e-07,
      "loss": 1.0917,
      "step": 1202
    },
    {
      "epoch": 3.726144297905353,
      "grad_norm": 2.421875,
      "learning_rate": 1.0875149678433893e-07,
      "loss": 1.1323,
      "step": 1203
    },
    {
      "epoch": 3.7292474786656324,
      "grad_norm": 2.515625,
      "learning_rate": 1.0621673560309798e-07,
      "loss": 1.1414,
      "step": 1204
    },
    {
      "epoch": 3.7323506594259115,
      "grad_norm": 2.390625,
      "learning_rate": 1.0371154654802929e-07,
      "loss": 1.1168,
      "step": 1205
    },
    {
      "epoch": 3.735453840186191,
      "grad_norm": 2.546875,
      "learning_rate": 1.01235944757474e-07,
      "loss": 1.1107,
      "step": 1206
    },
    {
      "epoch": 3.73855702094647,
      "grad_norm": 2.4375,
      "learning_rate": 9.878994519098573e-08,
      "loss": 1.1083,
      "step": 1207
    },
    {
      "epoch": 3.7416602017067495,
      "grad_norm": 2.515625,
      "learning_rate": 9.637356262923724e-08,
      "loss": 1.105,
      "step": 1208
    },
    {
      "epoch": 3.7447633824670286,
      "grad_norm": 2.53125,
      "learning_rate": 9.398681167393064e-08,
      "loss": 1.1188,
      "step": 1209
    },
    {
      "epoch": 3.747866563227308,
      "grad_norm": 2.515625,
      "learning_rate": 9.162970674771177e-08,
      "loss": 1.1101,
      "step": 1210
    },
    {
      "epoch": 3.7509697439875875,
      "grad_norm": 2.546875,
      "learning_rate": 8.930226209408089e-08,
      "loss": 1.0959,
      "step": 1211
    },
    {
      "epoch": 3.7540729247478666,
      "grad_norm": 2.46875,
      "learning_rate": 8.700449177730775e-08,
      "loss": 1.08,
      "step": 1212
    },
    {
      "epoch": 3.7571761055081456,
      "grad_norm": 2.515625,
      "learning_rate": 8.473640968234609e-08,
      "loss": 1.0717,
      "step": 1213
    },
    {
      "epoch": 3.760279286268425,
      "grad_norm": 2.59375,
      "learning_rate": 8.249802951474928e-08,
      "loss": 1.111,
      "step": 1214
    },
    {
      "epoch": 3.7633824670287046,
      "grad_norm": 2.609375,
      "learning_rate": 8.028936480058924e-08,
      "loss": 1.0903,
      "step": 1215
    },
    {
      "epoch": 3.7664856477889836,
      "grad_norm": 2.53125,
      "learning_rate": 7.81104288863721e-08,
      "loss": 1.1384,
      "step": 1216
    },
    {
      "epoch": 3.769588828549263,
      "grad_norm": 2.453125,
      "learning_rate": 7.59612349389599e-08,
      "loss": 1.1499,
      "step": 1217
    },
    {
      "epoch": 3.772692009309542,
      "grad_norm": 2.53125,
      "learning_rate": 7.384179594548957e-08,
      "loss": 1.1154,
      "step": 1218
    },
    {
      "epoch": 3.772692009309542,
      "eval_loss": 1.969642996788025,
      "eval_runtime": 35.7449,
      "eval_samples_per_second": 137.306,
      "eval_steps_per_second": 13.736,
      "step": 1218
    },
    {
      "epoch": 3.7757951900698217,
      "grad_norm": 2.65625,
      "learning_rate": 7.175212471329573e-08,
      "loss": 1.095,
      "step": 1219
    },
    {
      "epoch": 3.7788983708301007,
      "grad_norm": 2.484375,
      "learning_rate": 6.969223386983137e-08,
      "loss": 1.0595,
      "step": 1220
    },
    {
      "epoch": 3.78200155159038,
      "grad_norm": 2.46875,
      "learning_rate": 6.766213586259396e-08,
      "loss": 1.1559,
      "step": 1221
    },
    {
      "epoch": 3.7851047323506597,
      "grad_norm": 2.40625,
      "learning_rate": 6.566184295904777e-08,
      "loss": 1.0697,
      "step": 1222
    },
    {
      "epoch": 3.7882079131109387,
      "grad_norm": 2.53125,
      "learning_rate": 6.369136724655222e-08,
      "loss": 1.1103,
      "step": 1223
    },
    {
      "epoch": 3.791311093871218,
      "grad_norm": 2.453125,
      "learning_rate": 6.175072063228648e-08,
      "loss": 1.1123,
      "step": 1224
    },
    {
      "epoch": 3.7944142746314973,
      "grad_norm": 2.515625,
      "learning_rate": 5.983991484317997e-08,
      "loss": 1.1226,
      "step": 1225
    },
    {
      "epoch": 3.7975174553917768,
      "grad_norm": 2.59375,
      "learning_rate": 5.795896142583801e-08,
      "loss": 1.0709,
      "step": 1226
    },
    {
      "epoch": 3.800620636152056,
      "grad_norm": 2.34375,
      "learning_rate": 5.610787174647581e-08,
      "loss": 1.1146,
      "step": 1227
    },
    {
      "epoch": 3.8037238169123353,
      "grad_norm": 2.5625,
      "learning_rate": 5.4286656990847897e-08,
      "loss": 1.0504,
      "step": 1228
    },
    {
      "epoch": 3.8068269976726143,
      "grad_norm": 2.640625,
      "learning_rate": 5.249532816418101e-08,
      "loss": 1.1316,
      "step": 1229
    },
    {
      "epoch": 3.809930178432894,
      "grad_norm": 2.53125,
      "learning_rate": 5.0733896091105795e-08,
      "loss": 1.1507,
      "step": 1230
    },
    {
      "epoch": 3.813033359193173,
      "grad_norm": 2.40625,
      "learning_rate": 4.9002371415595716e-08,
      "loss": 1.0859,
      "step": 1231
    },
    {
      "epoch": 3.8161365399534524,
      "grad_norm": 2.28125,
      "learning_rate": 4.730076460089716e-08,
      "loss": 1.0807,
      "step": 1232
    },
    {
      "epoch": 3.8192397207137314,
      "grad_norm": 2.46875,
      "learning_rate": 4.5629085929470576e-08,
      "loss": 1.0137,
      "step": 1233
    },
    {
      "epoch": 3.822342901474011,
      "grad_norm": 2.46875,
      "learning_rate": 4.398734550292716e-08,
      "loss": 1.1228,
      "step": 1234
    },
    {
      "epoch": 3.82544608223429,
      "grad_norm": 2.5,
      "learning_rate": 4.237555324196563e-08,
      "loss": 1.1313,
      "step": 1235
    },
    {
      "epoch": 3.8285492629945694,
      "grad_norm": 2.578125,
      "learning_rate": 4.079371888631667e-08,
      "loss": 1.121,
      "step": 1236
    },
    {
      "epoch": 3.831652443754849,
      "grad_norm": 2.53125,
      "learning_rate": 3.9241851994678556e-08,
      "loss": 1.1492,
      "step": 1237
    },
    {
      "epoch": 3.834755624515128,
      "grad_norm": 2.421875,
      "learning_rate": 3.771996194466498e-08,
      "loss": 1.0961,
      "step": 1238
    },
    {
      "epoch": 3.8378588052754075,
      "grad_norm": 2.59375,
      "learning_rate": 3.622805793274397e-08,
      "loss": 1.0677,
      "step": 1239
    },
    {
      "epoch": 3.8378588052754075,
      "eval_loss": 1.9602329730987549,
      "eval_runtime": 35.7984,
      "eval_samples_per_second": 137.101,
      "eval_steps_per_second": 13.716,
      "step": 1239
    },
    {
      "epoch": 3.8409619860356865,
      "grad_norm": 2.484375,
      "learning_rate": 3.476614897418573e-08,
      "loss": 1.111,
      "step": 1240
    },
    {
      "epoch": 3.844065166795966,
      "grad_norm": 2.4375,
      "learning_rate": 3.333424390300433e-08,
      "loss": 1.0736,
      "step": 1241
    },
    {
      "epoch": 3.847168347556245,
      "grad_norm": 2.5625,
      "learning_rate": 3.193235137190831e-08,
      "loss": 1.1825,
      "step": 1242
    },
    {
      "epoch": 3.8502715283165245,
      "grad_norm": 2.46875,
      "learning_rate": 3.0560479852246304e-08,
      "loss": 1.0377,
      "step": 1243
    },
    {
      "epoch": 3.8533747090768036,
      "grad_norm": 2.515625,
      "learning_rate": 2.9218637633955382e-08,
      "loss": 1.132,
      "step": 1244
    },
    {
      "epoch": 3.856477889837083,
      "grad_norm": 2.390625,
      "learning_rate": 2.7906832825512764e-08,
      "loss": 1.0784,
      "step": 1245
    },
    {
      "epoch": 3.859581070597362,
      "grad_norm": 2.421875,
      "learning_rate": 2.6625073353884756e-08,
      "loss": 1.0914,
      "step": 1246
    },
    {
      "epoch": 3.8626842513576416,
      "grad_norm": 2.46875,
      "learning_rate": 2.5373366964480116e-08,
      "loss": 1.0986,
      "step": 1247
    },
    {
      "epoch": 3.865787432117921,
      "grad_norm": 2.5,
      "learning_rate": 2.415172122110343e-08,
      "loss": 1.0791,
      "step": 1248
    },
    {
      "epoch": 3.8688906128782,
      "grad_norm": 2.40625,
      "learning_rate": 2.2960143505909027e-08,
      "loss": 1.1128,
      "step": 1249
    },
    {
      "epoch": 3.871993793638479,
      "grad_norm": 2.515625,
      "learning_rate": 2.1798641019356027e-08,
      "loss": 1.1111,
      "step": 1250
    },
    {
      "epoch": 3.8750969743987587,
      "grad_norm": 2.546875,
      "learning_rate": 2.0667220780165033e-08,
      "loss": 1.1733,
      "step": 1251
    },
    {
      "epoch": 3.878200155159038,
      "grad_norm": 2.375,
      "learning_rate": 1.9565889625275945e-08,
      "loss": 1.0525,
      "step": 1252
    },
    {
      "epoch": 3.881303335919317,
      "grad_norm": 2.671875,
      "learning_rate": 1.849465420980634e-08,
      "loss": 1.0608,
      "step": 1253
    },
    {
      "epoch": 3.8844065166795967,
      "grad_norm": 2.484375,
      "learning_rate": 1.745352100701203e-08,
      "loss": 1.1161,
      "step": 1254
    },
    {
      "epoch": 3.8875096974398757,
      "grad_norm": 2.515625,
      "learning_rate": 1.644249630824657e-08,
      "loss": 1.1333,
      "step": 1255
    },
    {
      "epoch": 3.8906128782001552,
      "grad_norm": 2.390625,
      "learning_rate": 1.5461586222924596e-08,
      "loss": 1.0928,
      "step": 1256
    },
    {
      "epoch": 3.8937160589604343,
      "grad_norm": 2.421875,
      "learning_rate": 1.4510796678484096e-08,
      "loss": 1.1115,
      "step": 1257
    },
    {
      "epoch": 3.8968192397207138,
      "grad_norm": 2.421875,
      "learning_rate": 1.3590133420350315e-08,
      "loss": 1.1049,
      "step": 1258
    },
    {
      "epoch": 3.8999224204809932,
      "grad_norm": 2.578125,
      "learning_rate": 1.2699602011902456e-08,
      "loss": 1.0713,
      "step": 1259
    },
    {
      "epoch": 3.9030256012412723,
      "grad_norm": 2.53125,
      "learning_rate": 1.1839207834439259e-08,
      "loss": 1.1474,
      "step": 1260
    },
    {
      "epoch": 3.9030256012412723,
      "eval_loss": 1.9617058038711548,
      "eval_runtime": 35.7395,
      "eval_samples_per_second": 137.327,
      "eval_steps_per_second": 13.738,
      "step": 1260
    },
    {
      "epoch": 3.9061287820015513,
      "grad_norm": 2.453125,
      "learning_rate": 1.1008956087144585e-08,
      "loss": 1.0678,
      "step": 1261
    },
    {
      "epoch": 3.909231962761831,
      "grad_norm": 2.671875,
      "learning_rate": 1.0208851787060215e-08,
      "loss": 1.083,
      "step": 1262
    },
    {
      "epoch": 3.9123351435221103,
      "grad_norm": 2.453125,
      "learning_rate": 9.438899769051435e-09,
      "loss": 1.0748,
      "step": 1263
    },
    {
      "epoch": 3.9154383242823894,
      "grad_norm": 2.40625,
      "learning_rate": 8.699104685779835e-09,
      "loss": 1.1172,
      "step": 1264
    },
    {
      "epoch": 3.918541505042669,
      "grad_norm": 2.359375,
      "learning_rate": 7.989471007675554e-09,
      "loss": 1.1162,
      "step": 1265
    },
    {
      "epoch": 3.921644685802948,
      "grad_norm": 2.53125,
      "learning_rate": 7.310003022908407e-09,
      "loss": 1.0883,
      "step": 1266
    },
    {
      "epoch": 3.9247478665632274,
      "grad_norm": 2.46875,
      "learning_rate": 6.660704837364029e-09,
      "loss": 1.0532,
      "step": 1267
    },
    {
      "epoch": 3.9278510473235064,
      "grad_norm": 2.453125,
      "learning_rate": 6.0415803746183275e-09,
      "loss": 1.0812,
      "step": 1268
    },
    {
      "epoch": 3.930954228083786,
      "grad_norm": 2.578125,
      "learning_rate": 5.452633375912509e-09,
      "loss": 1.1095,
      "step": 1269
    },
    {
      "epoch": 3.9340574088440654,
      "grad_norm": 2.4375,
      "learning_rate": 4.89386740013198e-09,
      "loss": 1.1008,
      "step": 1270
    },
    {
      "epoch": 3.9371605896043445,
      "grad_norm": 2.421875,
      "learning_rate": 4.365285823785259e-09,
      "loss": 1.0762,
      "step": 1271
    },
    {
      "epoch": 3.9402637703646235,
      "grad_norm": 2.578125,
      "learning_rate": 3.866891840982323e-09,
      "loss": 1.1288,
      "step": 1272
    },
    {
      "epoch": 3.943366951124903,
      "grad_norm": 2.53125,
      "learning_rate": 3.3986884634146232e-09,
      "loss": 1.0526,
      "step": 1273
    },
    {
      "epoch": 3.9464701318851825,
      "grad_norm": 2.390625,
      "learning_rate": 2.960678520340099e-09,
      "loss": 1.0841,
      "step": 1274
    },
    {
      "epoch": 3.9495733126454615,
      "grad_norm": 2.375,
      "learning_rate": 2.552864658561527e-09,
      "loss": 1.1006,
      "step": 1275
    },
    {
      "epoch": 3.952676493405741,
      "grad_norm": 2.546875,
      "learning_rate": 2.1752493424148647e-09,
      "loss": 1.1046,
      "step": 1276
    },
    {
      "epoch": 3.95577967416602,
      "grad_norm": 2.546875,
      "learning_rate": 1.8278348537525969e-09,
      "loss": 1.136,
      "step": 1277
    },
    {
      "epoch": 3.9588828549262995,
      "grad_norm": 2.359375,
      "learning_rate": 1.5106232919276376e-09,
      "loss": 1.0831,
      "step": 1278
    },
    {
      "epoch": 3.9619860356865786,
      "grad_norm": 2.515625,
      "learning_rate": 1.2236165737850025e-09,
      "loss": 1.153,
      "step": 1279
    },
    {
      "epoch": 3.965089216446858,
      "grad_norm": 2.6875,
      "learning_rate": 9.668164336473772e-10,
      "loss": 1.065,
      "step": 1280
    },
    {
      "epoch": 3.9681923972071376,
      "grad_norm": 2.34375,
      "learning_rate": 7.402244233045697e-10,
      "loss": 1.059,
      "step": 1281
    },
    {
      "epoch": 3.9681923972071376,
      "eval_loss": 1.9586632251739502,
      "eval_runtime": 35.6392,
      "eval_samples_per_second": 137.713,
      "eval_steps_per_second": 13.777,
      "step": 1281
    },
    {
      "epoch": 3.9712955779674166,
      "grad_norm": 2.484375,
      "learning_rate": 5.438419120062933e-10,
      "loss": 1.0562,
      "step": 1282
    },
    {
      "epoch": 3.9743987587276957,
      "grad_norm": 2.59375,
      "learning_rate": 3.7767008645106517e-10,
      "loss": 1.0734,
      "step": 1283
    },
    {
      "epoch": 3.977501939487975,
      "grad_norm": 2.53125,
      "learning_rate": 2.417099507817655e-10,
      "loss": 1.0631,
      "step": 1284
    },
    {
      "epoch": 3.9806051202482546,
      "grad_norm": 2.625,
      "learning_rate": 1.3596232657675512e-10,
      "loss": 1.0862,
      "step": 1285
    },
    {
      "epoch": 3.9837083010085337,
      "grad_norm": 2.453125,
      "learning_rate": 6.042785284821051e-11,
      "loss": 1.0846,
      "step": 1286
    },
    {
      "epoch": 3.986811481768813,
      "grad_norm": 2.5,
      "learning_rate": 1.5106986034352234e-11,
      "loss": 1.0656,
      "step": 1287
    },
    {
      "epoch": 3.989914662529092,
      "grad_norm": 2.625,
      "learning_rate": 0.0,
      "loss": 1.0467,
      "step": 1288
    }
  ],
  "logging_steps": 1,
  "max_steps": 1288,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 81,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.8971604216100946e+18,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}