{
  "best_metric": 0.9042022824287415,
  "best_model_checkpoint": "data/Llama-31-8B_task-1_120-samples_config-4_full/checkpoint-319",
  "epoch": 64.9090909090909,
  "eval_steps": 500,
  "global_step": 357,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.18181818181818182,
      "grad_norm": 1.9413753747940063,
      "learning_rate": 1.3333333333333336e-07,
      "loss": 2.4963,
      "step": 1
    },
    {
      "epoch": 0.36363636363636365,
      "grad_norm": 1.9993785619735718,
      "learning_rate": 2.666666666666667e-07,
      "loss": 2.515,
      "step": 2
    },
    {
      "epoch": 0.7272727272727273,
      "grad_norm": 1.8845915794372559,
      "learning_rate": 5.333333333333335e-07,
      "loss": 2.4687,
      "step": 4
    },
    {
      "epoch": 0.9090909090909091,
      "eval_loss": 2.458869218826294,
      "eval_runtime": 9.6295,
      "eval_samples_per_second": 2.492,
      "eval_steps_per_second": 2.492,
      "step": 5
    },
    {
      "epoch": 1.0909090909090908,
      "grad_norm": 2.0253942012786865,
      "learning_rate": 8.000000000000001e-07,
      "loss": 2.4759,
      "step": 6
    },
    {
      "epoch": 1.4545454545454546,
      "grad_norm": 1.8723983764648438,
      "learning_rate": 1.066666666666667e-06,
      "loss": 2.4786,
      "step": 8
    },
    {
      "epoch": 1.8181818181818183,
      "grad_norm": 1.924127459526062,
      "learning_rate": 1.3333333333333334e-06,
      "loss": 2.5083,
      "step": 10
    },
    {
      "epoch": 2.0,
      "eval_loss": 2.443969488143921,
      "eval_runtime": 9.6222,
      "eval_samples_per_second": 2.494,
      "eval_steps_per_second": 2.494,
      "step": 11
    },
    {
      "epoch": 2.1818181818181817,
      "grad_norm": 1.7455520629882812,
      "learning_rate": 1.6000000000000001e-06,
      "loss": 2.4117,
      "step": 12
    },
    {
      "epoch": 2.5454545454545454,
      "grad_norm": 1.710787296295166,
      "learning_rate": 1.8666666666666669e-06,
      "loss": 2.4583,
      "step": 14
    },
    {
      "epoch": 2.909090909090909,
      "grad_norm": 1.6907106637954712,
      "learning_rate": 2.133333333333334e-06,
      "loss": 2.4676,
      "step": 16
    },
    {
      "epoch": 2.909090909090909,
      "eval_loss": 2.421785593032837,
      "eval_runtime": 9.6333,
      "eval_samples_per_second": 2.491,
      "eval_steps_per_second": 2.491,
      "step": 16
    },
    {
      "epoch": 3.2727272727272725,
      "grad_norm": 1.5824416875839233,
      "learning_rate": 2.4000000000000003e-06,
      "loss": 2.4237,
      "step": 18
    },
    {
      "epoch": 3.6363636363636362,
      "grad_norm": 1.59761643409729,
      "learning_rate": 2.666666666666667e-06,
      "loss": 2.4148,
      "step": 20
    },
    {
      "epoch": 4.0,
      "grad_norm": 1.6097276210784912,
      "learning_rate": 2.9333333333333338e-06,
      "loss": 2.4562,
      "step": 22
    },
    {
      "epoch": 4.0,
      "eval_loss": 2.3870151042938232,
      "eval_runtime": 9.6315,
      "eval_samples_per_second": 2.492,
      "eval_steps_per_second": 2.492,
      "step": 22
    },
    {
      "epoch": 4.363636363636363,
      "grad_norm": 1.636257529258728,
      "learning_rate": 3.2000000000000003e-06,
      "loss": 2.4287,
      "step": 24
    },
    {
      "epoch": 4.7272727272727275,
      "grad_norm": 1.5569593906402588,
      "learning_rate": 3.4666666666666672e-06,
      "loss": 2.377,
      "step": 26
    },
    {
      "epoch": 4.909090909090909,
      "eval_loss": 2.3474807739257812,
      "eval_runtime": 9.6237,
      "eval_samples_per_second": 2.494,
      "eval_steps_per_second": 2.494,
      "step": 27
    },
    {
      "epoch": 5.090909090909091,
      "grad_norm": 1.808140516281128,
      "learning_rate": 3.7333333333333337e-06,
      "loss": 2.3605,
      "step": 28
    },
    {
      "epoch": 5.454545454545454,
      "grad_norm": 1.7729766368865967,
      "learning_rate": 4.000000000000001e-06,
      "loss": 2.3404,
      "step": 30
    },
    {
      "epoch": 5.818181818181818,
      "grad_norm": 1.9055501222610474,
      "learning_rate": 4.266666666666668e-06,
      "loss": 2.3303,
      "step": 32
    },
    {
      "epoch": 6.0,
      "eval_loss": 2.2793145179748535,
      "eval_runtime": 9.6178,
      "eval_samples_per_second": 2.495,
      "eval_steps_per_second": 2.495,
      "step": 33
    },
    {
      "epoch": 6.181818181818182,
      "grad_norm": 1.628233790397644,
      "learning_rate": 4.533333333333334e-06,
      "loss": 2.2739,
      "step": 34
    },
    {
      "epoch": 6.545454545454545,
      "grad_norm": 1.4241219758987427,
      "learning_rate": 4.800000000000001e-06,
      "loss": 2.3294,
      "step": 36
    },
    {
      "epoch": 6.909090909090909,
      "grad_norm": 1.5400785207748413,
      "learning_rate": 5.0666666666666676e-06,
      "loss": 2.2553,
      "step": 38
    },
    {
      "epoch": 6.909090909090909,
      "eval_loss": 2.2254207134246826,
      "eval_runtime": 9.6211,
      "eval_samples_per_second": 2.495,
      "eval_steps_per_second": 2.495,
      "step": 38
    },
    {
      "epoch": 7.2727272727272725,
      "grad_norm": 1.4445594549179077,
      "learning_rate": 5.333333333333334e-06,
      "loss": 2.2498,
      "step": 40
    },
    {
      "epoch": 7.636363636363637,
      "grad_norm": 1.7270216941833496,
      "learning_rate": 5.600000000000001e-06,
      "loss": 2.1853,
      "step": 42
    },
    {
      "epoch": 8.0,
      "grad_norm": 2.017972707748413,
      "learning_rate": 5.8666666666666675e-06,
      "loss": 2.174,
      "step": 44
    },
    {
      "epoch": 8.0,
      "eval_loss": 2.139225721359253,
      "eval_runtime": 9.6197,
      "eval_samples_per_second": 2.495,
      "eval_steps_per_second": 2.495,
      "step": 44
    },
    {
      "epoch": 8.363636363636363,
      "grad_norm": 1.3301095962524414,
      "learning_rate": 6.133333333333334e-06,
      "loss": 2.1429,
      "step": 46
    },
    {
      "epoch": 8.727272727272727,
      "grad_norm": 1.083661675453186,
      "learning_rate": 6.4000000000000006e-06,
      "loss": 2.131,
      "step": 48
    },
    {
      "epoch": 8.909090909090908,
      "eval_loss": 2.0661048889160156,
      "eval_runtime": 9.6328,
      "eval_samples_per_second": 2.491,
      "eval_steps_per_second": 2.491,
      "step": 49
    },
    {
      "epoch": 9.090909090909092,
      "grad_norm": 1.0499473810195923,
      "learning_rate": 6.666666666666667e-06,
      "loss": 2.0813,
      "step": 50
    },
    {
      "epoch": 9.454545454545455,
      "grad_norm": 1.014916181564331,
      "learning_rate": 6.9333333333333344e-06,
      "loss": 2.0497,
      "step": 52
    },
    {
      "epoch": 9.818181818181818,
      "grad_norm": 1.0278065204620361,
      "learning_rate": 7.2000000000000005e-06,
      "loss": 2.0142,
      "step": 54
    },
    {
      "epoch": 10.0,
      "eval_loss": 1.9625515937805176,
      "eval_runtime": 9.6185,
      "eval_samples_per_second": 2.495,
      "eval_steps_per_second": 2.495,
      "step": 55
    },
    {
      "epoch": 10.181818181818182,
      "grad_norm": 0.9720411896705627,
      "learning_rate": 7.4666666666666675e-06,
      "loss": 1.9759,
      "step": 56
    },
    {
      "epoch": 10.545454545454545,
      "grad_norm": 0.9346638321876526,
      "learning_rate": 7.733333333333334e-06,
      "loss": 1.941,
      "step": 58
    },
    {
      "epoch": 10.909090909090908,
      "grad_norm": 0.8559221029281616,
      "learning_rate": 8.000000000000001e-06,
      "loss": 1.8873,
      "step": 60
    },
    {
      "epoch": 10.909090909090908,
      "eval_loss": 1.8745914697647095,
      "eval_runtime": 9.6239,
      "eval_samples_per_second": 2.494,
      "eval_steps_per_second": 2.494,
      "step": 60
    },
    {
      "epoch": 11.272727272727273,
      "grad_norm": 0.8817884922027588,
      "learning_rate": 8.266666666666667e-06,
      "loss": 1.9132,
      "step": 62
    },
    {
      "epoch": 11.636363636363637,
      "grad_norm": 0.8232048749923706,
      "learning_rate": 8.533333333333335e-06,
      "loss": 1.8387,
      "step": 64
    },
    {
      "epoch": 12.0,
      "grad_norm": 0.8017051815986633,
      "learning_rate": 8.8e-06,
      "loss": 1.7633,
      "step": 66
    },
    {
      "epoch": 12.0,
      "eval_loss": 1.7650254964828491,
      "eval_runtime": 9.6208,
      "eval_samples_per_second": 2.495,
      "eval_steps_per_second": 2.495,
      "step": 66
    },
    {
      "epoch": 12.363636363636363,
      "grad_norm": 0.9119341373443604,
      "learning_rate": 9.066666666666667e-06,
      "loss": 1.72,
      "step": 68
    },
    {
      "epoch": 12.727272727272727,
      "grad_norm": 0.8771039843559265,
      "learning_rate": 9.333333333333334e-06,
      "loss": 1.726,
      "step": 70
    },
    {
      "epoch": 12.909090909090908,
      "eval_loss": 1.6562695503234863,
      "eval_runtime": 9.6212,
      "eval_samples_per_second": 2.494,
      "eval_steps_per_second": 2.494,
      "step": 71
    },
    {
      "epoch": 13.090909090909092,
      "grad_norm": 0.9313778877258301,
      "learning_rate": 9.600000000000001e-06,
      "loss": 1.6816,
      "step": 72
    },
    {
      "epoch": 13.454545454545455,
      "grad_norm": 1.1438463926315308,
      "learning_rate": 9.866666666666668e-06,
      "loss": 1.6168,
      "step": 74
    },
    {
      "epoch": 13.818181818181818,
      "grad_norm": 1.0701647996902466,
      "learning_rate": 9.999945845889795e-06,
      "loss": 1.5711,
      "step": 76
    },
    {
      "epoch": 14.0,
      "eval_loss": 1.5122851133346558,
      "eval_runtime": 9.627,
      "eval_samples_per_second": 2.493,
      "eval_steps_per_second": 2.493,
      "step": 77
    },
    {
      "epoch": 14.181818181818182,
      "grad_norm": 0.9771044254302979,
      "learning_rate": 9.999512620046523e-06,
      "loss": 1.5488,
      "step": 78
    },
    {
      "epoch": 14.545454545454545,
      "grad_norm": 0.91764235496521,
      "learning_rate": 9.99864620589731e-06,
      "loss": 1.4504,
      "step": 80
    },
    {
      "epoch": 14.909090909090908,
      "grad_norm": 0.9226170182228088,
      "learning_rate": 9.99734667851357e-06,
      "loss": 1.4344,
      "step": 82
    },
    {
      "epoch": 14.909090909090908,
      "eval_loss": 1.3950275182724,
      "eval_runtime": 9.6306,
      "eval_samples_per_second": 2.492,
      "eval_steps_per_second": 2.492,
      "step": 82
    },
    {
      "epoch": 15.272727272727273,
      "grad_norm": 0.8576654195785522,
      "learning_rate": 9.995614150494293e-06,
      "loss": 1.3568,
      "step": 84
    },
    {
      "epoch": 15.636363636363637,
      "grad_norm": 0.9888388514518738,
      "learning_rate": 9.993448771956285e-06,
      "loss": 1.3195,
      "step": 86
    },
    {
      "epoch": 16.0,
      "grad_norm": 0.9264158010482788,
      "learning_rate": 9.99085073052117e-06,
      "loss": 1.3201,
      "step": 88
    },
    {
      "epoch": 16.0,
      "eval_loss": 1.2661280632019043,
      "eval_runtime": 9.6187,
      "eval_samples_per_second": 2.495,
      "eval_steps_per_second": 2.495,
      "step": 88
    },
    {
      "epoch": 16.363636363636363,
      "grad_norm": 0.8193872570991516,
      "learning_rate": 9.987820251299121e-06,
      "loss": 1.2496,
      "step": 90
    },
    {
      "epoch": 16.727272727272727,
      "grad_norm": 0.7646848559379578,
      "learning_rate": 9.984357596869369e-06,
      "loss": 1.1787,
      "step": 92
    },
    {
      "epoch": 16.90909090909091,
      "eval_loss": 1.18313729763031,
      "eval_runtime": 9.6587,
      "eval_samples_per_second": 2.485,
      "eval_steps_per_second": 2.485,
      "step": 93
    },
    {
      "epoch": 17.09090909090909,
      "grad_norm": 0.7123040556907654,
      "learning_rate": 9.980463067257437e-06,
      "loss": 1.2232,
      "step": 94
    },
    {
      "epoch": 17.454545454545453,
      "grad_norm": 0.6257199645042419,
      "learning_rate": 9.976136999909156e-06,
      "loss": 1.1068,
      "step": 96
    },
    {
      "epoch": 17.818181818181817,
      "grad_norm": 0.7334635257720947,
      "learning_rate": 9.971379769661422e-06,
      "loss": 1.1444,
      "step": 98
    },
    {
      "epoch": 18.0,
      "eval_loss": 1.1187793016433716,
      "eval_runtime": 9.6189,
      "eval_samples_per_second": 2.495,
      "eval_steps_per_second": 2.495,
      "step": 99
    },
    {
      "epoch": 18.181818181818183,
      "grad_norm": 0.589821457862854,
      "learning_rate": 9.966191788709716e-06,
      "loss": 1.1334,
      "step": 100
    },
    {
      "epoch": 18.545454545454547,
      "grad_norm": 0.5560262799263,
      "learning_rate": 9.960573506572391e-06,
      "loss": 1.0929,
      "step": 102
    },
    {
      "epoch": 18.90909090909091,
      "grad_norm": 0.5528337359428406,
      "learning_rate": 9.95452541005172e-06,
      "loss": 1.0591,
      "step": 104
    },
    {
      "epoch": 18.90909090909091,
      "eval_loss": 1.0836217403411865,
      "eval_runtime": 9.632,
      "eval_samples_per_second": 2.492,
      "eval_steps_per_second": 2.492,
      "step": 104
    },
    {
      "epoch": 19.272727272727273,
      "grad_norm": 0.5791244506835938,
      "learning_rate": 9.948048023191728e-06,
      "loss": 1.0522,
      "step": 106
    },
    {
      "epoch": 19.636363636363637,
      "grad_norm": 0.5012540817260742,
      "learning_rate": 9.941141907232766e-06,
      "loss": 1.0986,
      "step": 108
    },
    {
      "epoch": 20.0,
      "grad_norm": 0.4489583969116211,
      "learning_rate": 9.933807660562898e-06,
      "loss": 1.0151,
      "step": 110
    },
    {
      "epoch": 20.0,
      "eval_loss": 1.0539679527282715,
      "eval_runtime": 9.6338,
      "eval_samples_per_second": 2.491,
      "eval_steps_per_second": 2.491,
      "step": 110
    },
    {
      "epoch": 20.363636363636363,
      "grad_norm": 0.384694367647171,
      "learning_rate": 9.926045918666045e-06,
      "loss": 1.0363,
      "step": 112
    },
    {
      "epoch": 20.727272727272727,
      "grad_norm": 0.3782545030117035,
      "learning_rate": 9.91785735406693e-06,
      "loss": 1.0277,
      "step": 114
    },
    {
      "epoch": 20.90909090909091,
      "eval_loss": 1.0388442277908325,
      "eval_runtime": 9.6309,
      "eval_samples_per_second": 2.492,
      "eval_steps_per_second": 2.492,
      "step": 115
    },
    {
      "epoch": 21.09090909090909,
      "grad_norm": 0.3923183083534241,
      "learning_rate": 9.909242676272797e-06,
      "loss": 1.0385,
      "step": 116
    },
    {
      "epoch": 21.454545454545453,
      "grad_norm": 0.3774013817310333,
      "learning_rate": 9.90020263171194e-06,
      "loss": 1.0073,
      "step": 118
    },
    {
      "epoch": 21.818181818181817,
      "grad_norm": 0.3472942113876343,
      "learning_rate": 9.890738003669029e-06,
      "loss": 1.0025,
      "step": 120
    },
    {
      "epoch": 22.0,
      "eval_loss": 1.0249600410461426,
      "eval_runtime": 9.6161,
      "eval_samples_per_second": 2.496,
      "eval_steps_per_second": 2.496,
      "step": 121
    },
    {
      "epoch": 22.181818181818183,
      "grad_norm": 0.3280906081199646,
      "learning_rate": 9.880849612217238e-06,
      "loss": 0.9887,
      "step": 122
    },
    {
      "epoch": 22.545454545454547,
      "grad_norm": 0.3648756742477417,
      "learning_rate": 9.870538314147194e-06,
      "loss": 1.0023,
      "step": 124
    },
    {
      "epoch": 22.90909090909091,
      "grad_norm": 0.390601247549057,
      "learning_rate": 9.859805002892733e-06,
      "loss": 1.0161,
      "step": 126
    },
    {
      "epoch": 22.90909090909091,
      "eval_loss": 1.0154192447662354,
      "eval_runtime": 9.6244,
      "eval_samples_per_second": 2.494,
      "eval_steps_per_second": 2.494,
      "step": 126
    },
    {
      "epoch": 23.272727272727273,
      "grad_norm": 0.3790784180164337,
      "learning_rate": 9.84865060845349e-06,
      "loss": 0.9941,
      "step": 128
    },
    {
      "epoch": 23.636363636363637,
      "grad_norm": 0.36824390292167664,
      "learning_rate": 9.83707609731432e-06,
      "loss": 0.9697,
      "step": 130
    },
    {
      "epoch": 24.0,
      "grad_norm": 0.39850422739982605,
      "learning_rate": 9.825082472361558e-06,
      "loss": 0.9946,
      "step": 132
    },
    {
      "epoch": 24.0,
      "eval_loss": 1.0047398805618286,
      "eval_runtime": 9.6214,
      "eval_samples_per_second": 2.494,
      "eval_steps_per_second": 2.494,
      "step": 132
    },
    {
      "epoch": 24.363636363636363,
      "grad_norm": 0.34537094831466675,
      "learning_rate": 9.812670772796113e-06,
      "loss": 0.9699,
      "step": 134
    },
    {
      "epoch": 24.727272727272727,
      "grad_norm": 0.38018471002578735,
      "learning_rate": 9.799842074043438e-06,
      "loss": 0.9773,
      "step": 136
    },
    {
      "epoch": 24.90909090909091,
      "eval_loss": 0.9969633221626282,
      "eval_runtime": 9.6246,
      "eval_samples_per_second": 2.494,
      "eval_steps_per_second": 2.494,
      "step": 137
    },
    {
      "epoch": 25.09090909090909,
      "grad_norm": 0.40856873989105225,
      "learning_rate": 9.786597487660336e-06,
      "loss": 0.9836,
      "step": 138
    },
    {
      "epoch": 25.454545454545453,
      "grad_norm": 0.37090280652046204,
      "learning_rate": 9.77293816123866e-06,
      "loss": 0.9658,
      "step": 140
    },
    {
      "epoch": 25.818181818181817,
      "grad_norm": 0.4068634808063507,
      "learning_rate": 9.75886527830587e-06,
      "loss": 0.9708,
      "step": 142
    },
    {
      "epoch": 26.0,
      "eval_loss": 0.9890053272247314,
      "eval_runtime": 9.6268,
      "eval_samples_per_second": 2.493,
      "eval_steps_per_second": 2.493,
      "step": 143
    },
    {
      "epoch": 26.181818181818183,
      "grad_norm": 0.38360726833343506,
      "learning_rate": 9.744380058222483e-06,
      "loss": 0.9676,
      "step": 144
    },
    {
      "epoch": 26.545454545454547,
      "grad_norm": 0.38106632232666016,
      "learning_rate": 9.729483756076436e-06,
      "loss": 0.972,
      "step": 146
    },
    {
      "epoch": 26.90909090909091,
      "grad_norm": 0.36939358711242676,
      "learning_rate": 9.714177662574316e-06,
      "loss": 0.9374,
      "step": 148
    },
    {
      "epoch": 26.90909090909091,
      "eval_loss": 0.9821727275848389,
      "eval_runtime": 9.622,
      "eval_samples_per_second": 2.494,
      "eval_steps_per_second": 2.494,
      "step": 148
    },
    {
      "epoch": 27.272727272727273,
      "grad_norm": 0.37566348910331726,
      "learning_rate": 9.698463103929542e-06,
      "loss": 0.9219,
      "step": 150
    },
    {
      "epoch": 27.636363636363637,
      "grad_norm": 0.3677101731300354,
      "learning_rate": 9.682341441747446e-06,
      "loss": 0.9798,
      "step": 152
    },
    {
      "epoch": 28.0,
      "grad_norm": 0.3695693016052246,
      "learning_rate": 9.665814072907293e-06,
      "loss": 0.9403,
      "step": 154
    },
    {
      "epoch": 28.0,
      "eval_loss": 0.9750909209251404,
      "eval_runtime": 9.6244,
      "eval_samples_per_second": 2.494,
      "eval_steps_per_second": 2.494,
      "step": 154
    },
    {
      "epoch": 28.363636363636363,
      "grad_norm": 0.42501190304756165,
      "learning_rate": 9.648882429441258e-06,
      "loss": 0.9428,
      "step": 156
    },
    {
      "epoch": 28.727272727272727,
      "grad_norm": 0.3643590807914734,
      "learning_rate": 9.63154797841033e-06,
      "loss": 0.94,
      "step": 158
    },
    {
      "epoch": 28.90909090909091,
      "eval_loss": 0.9702978134155273,
      "eval_runtime": 9.6253,
      "eval_samples_per_second": 2.493,
      "eval_steps_per_second": 2.493,
      "step": 159
    },
    {
      "epoch": 29.09090909090909,
      "grad_norm": 0.3957996666431427,
      "learning_rate": 9.613812221777212e-06,
      "loss": 0.9274,
      "step": 160
    },
    {
      "epoch": 29.454545454545453,
      "grad_norm": 0.4291062355041504,
      "learning_rate": 9.595676696276173e-06,
      "loss": 0.9886,
      "step": 162
    },
    {
      "epoch": 29.818181818181817,
      "grad_norm": 0.5365828275680542,
      "learning_rate": 9.577142973279896e-06,
      "loss": 0.902,
      "step": 164
    },
    {
      "epoch": 30.0,
      "eval_loss": 0.9632946848869324,
      "eval_runtime": 9.6187,
      "eval_samples_per_second": 2.495,
      "eval_steps_per_second": 2.495,
      "step": 165
    },
    {
      "epoch": 30.181818181818183,
      "grad_norm": 0.38883283734321594,
      "learning_rate": 9.55821265866333e-06,
      "loss": 0.8967,
      "step": 166
    },
    {
      "epoch": 30.545454545454547,
      "grad_norm": 0.41333243250846863,
      "learning_rate": 9.538887392664544e-06,
      "loss": 0.9142,
      "step": 168
    },
    {
      "epoch": 30.90909090909091,
      "grad_norm": 0.4123990833759308,
      "learning_rate": 9.519168849742603e-06,
      "loss": 0.9215,
      "step": 170
    },
    {
      "epoch": 30.90909090909091,
      "eval_loss": 0.9604056477546692,
      "eval_runtime": 9.627,
      "eval_samples_per_second": 2.493,
      "eval_steps_per_second": 2.493,
      "step": 170
    },
    {
      "epoch": 31.272727272727273,
      "grad_norm": 0.407969206571579,
      "learning_rate": 9.499058738432492e-06,
      "loss": 0.9574,
      "step": 172
    },
    {
      "epoch": 31.636363636363637,
      "grad_norm": 0.4867004156112671,
      "learning_rate": 9.478558801197065e-06,
      "loss": 0.9208,
      "step": 174
    },
    {
      "epoch": 32.0,
      "grad_norm": 0.4684889316558838,
      "learning_rate": 9.457670814276083e-06,
      "loss": 0.8854,
      "step": 176
    },
    {
      "epoch": 32.0,
      "eval_loss": 0.9548270106315613,
      "eval_runtime": 9.6173,
      "eval_samples_per_second": 2.495,
      "eval_steps_per_second": 2.495,
      "step": 176
    },
    {
      "epoch": 32.36363636363637,
      "grad_norm": 0.527631938457489,
      "learning_rate": 9.436396587532297e-06,
      "loss": 0.8831,
      "step": 178
    },
    {
      "epoch": 32.72727272727273,
      "grad_norm": 0.44928282499313354,
      "learning_rate": 9.414737964294636e-06,
      "loss": 0.96,
      "step": 180
    },
    {
      "epoch": 32.90909090909091,
      "eval_loss": 0.9503173232078552,
      "eval_runtime": 9.624,
      "eval_samples_per_second": 2.494,
      "eval_steps_per_second": 2.494,
      "step": 181
    },
    {
      "epoch": 33.09090909090909,
      "grad_norm": 0.5621985197067261,
      "learning_rate": 9.392696821198488e-06,
      "loss": 0.8666,
      "step": 182
    },
    {
      "epoch": 33.45454545454545,
      "grad_norm": 0.523452877998352,
      "learning_rate": 9.370275068023097e-06,
      "loss": 0.922,
      "step": 184
    },
    {
      "epoch": 33.81818181818182,
      "grad_norm": 0.5437294840812683,
      "learning_rate": 9.347474647526095e-06,
      "loss": 0.9162,
      "step": 186
    },
    {
      "epoch": 34.0,
      "eval_loss": 0.9452812075614929,
      "eval_runtime": 9.6313,
      "eval_samples_per_second": 2.492,
      "eval_steps_per_second": 2.492,
      "step": 187
    },
    {
      "epoch": 34.18181818181818,
      "grad_norm": 0.46963879466056824,
      "learning_rate": 9.324297535275156e-06,
      "loss": 0.8254,
      "step": 188
    },
    {
      "epoch": 34.54545454545455,
      "grad_norm": 0.48245498538017273,
      "learning_rate": 9.30074573947683e-06,
      "loss": 0.9174,
      "step": 190
    },
    {
      "epoch": 34.90909090909091,
      "grad_norm": 0.5139335989952087,
      "learning_rate": 9.276821300802535e-06,
      "loss": 0.8686,
      "step": 192
    },
    {
      "epoch": 34.90909090909091,
      "eval_loss": 0.9428532719612122,
      "eval_runtime": 9.6309,
      "eval_samples_per_second": 2.492,
      "eval_steps_per_second": 2.492,
      "step": 192
    },
    {
      "epoch": 35.27272727272727,
      "grad_norm": 0.45418813824653625,
      "learning_rate": 9.25252629221175e-06,
      "loss": 0.9011,
      "step": 194
    },
    {
      "epoch": 35.63636363636363,
      "grad_norm": 0.5155036449432373,
      "learning_rate": 9.227862818772392e-06,
      "loss": 0.8754,
      "step": 196
    },
    {
      "epoch": 36.0,
      "grad_norm": 0.4917118549346924,
      "learning_rate": 9.202833017478421e-06,
      "loss": 0.906,
      "step": 198
    },
    {
      "epoch": 36.0,
      "eval_loss": 0.9385306239128113,
      "eval_runtime": 9.6304,
      "eval_samples_per_second": 2.492,
      "eval_steps_per_second": 2.492,
      "step": 198
    },
    {
      "epoch": 36.36363636363637,
      "grad_norm": 0.5289394855499268,
      "learning_rate": 9.177439057064684e-06,
      "loss": 0.8751,
      "step": 200
    },
    {
      "epoch": 36.72727272727273,
      "grad_norm": 0.5498368144035339,
      "learning_rate": 9.151683137818989e-06,
      "loss": 0.8762,
      "step": 202
    },
    {
      "epoch": 36.90909090909091,
      "eval_loss": 0.9353806972503662,
      "eval_runtime": 9.6269,
      "eval_samples_per_second": 2.493,
      "eval_steps_per_second": 2.493,
      "step": 203
    },
    {
      "epoch": 37.09090909090909,
      "grad_norm": 0.516069233417511,
      "learning_rate": 9.125567491391476e-06,
      "loss": 0.869,
      "step": 204
    },
    {
      "epoch": 37.45454545454545,
      "grad_norm": 0.5102888345718384,
      "learning_rate": 9.099094380601244e-06,
      "loss": 0.8518,
      "step": 206
    },
    {
      "epoch": 37.81818181818182,
      "grad_norm": 0.5379929542541504,
      "learning_rate": 9.072266099240286e-06,
      "loss": 0.8929,
      "step": 208
    },
    {
      "epoch": 38.0,
      "eval_loss": 0.9331977963447571,
      "eval_runtime": 9.6277,
      "eval_samples_per_second": 2.493,
      "eval_steps_per_second": 2.493,
      "step": 209
    },
    {
      "epoch": 38.18181818181818,
      "grad_norm": 0.6433578729629517,
      "learning_rate": 9.045084971874738e-06,
      "loss": 0.8756,
      "step": 210
    },
    {
      "epoch": 38.54545454545455,
      "grad_norm": 0.6186140179634094,
      "learning_rate": 9.017553353643479e-06,
      "loss": 0.8582,
      "step": 212
    },
    {
      "epoch": 38.90909090909091,
      "grad_norm": 0.608066976070404,
      "learning_rate": 8.989673630054044e-06,
      "loss": 0.8687,
      "step": 214
    },
    {
      "epoch": 38.90909090909091,
      "eval_loss": 0.9301042556762695,
      "eval_runtime": 9.6307,
      "eval_samples_per_second": 2.492,
      "eval_steps_per_second": 2.492,
      "step": 214
    },
    {
      "epoch": 39.27272727272727,
      "grad_norm": 0.6045626401901245,
      "learning_rate": 8.961448216775955e-06,
      "loss": 0.8119,
      "step": 216
    },
    {
      "epoch": 39.63636363636363,
      "grad_norm": 0.6160129308700562,
      "learning_rate": 8.932879559431392e-06,
      "loss": 0.8301,
      "step": 218
    },
    {
      "epoch": 40.0,
      "grad_norm": 0.6550566554069519,
      "learning_rate": 8.903970133383297e-06,
      "loss": 0.8933,
      "step": 220
    },
    {
      "epoch": 40.0,
      "eval_loss": 0.9279410243034363,
      "eval_runtime": 9.6288,
      "eval_samples_per_second": 2.493,
      "eval_steps_per_second": 2.493,
      "step": 220
    },
    {
      "epoch": 40.36363636363637,
      "grad_norm": 0.6415209770202637,
      "learning_rate": 8.874722443520898e-06,
      "loss": 0.8325,
      "step": 222
    },
    {
      "epoch": 40.72727272727273,
      "grad_norm": 0.6836015582084656,
      "learning_rate": 8.845139024042664e-06,
      "loss": 0.858,
      "step": 224
    },
    {
      "epoch": 40.90909090909091,
      "eval_loss": 0.9241297841072083,
      "eval_runtime": 9.6293,
      "eval_samples_per_second": 2.492,
      "eval_steps_per_second": 2.492,
      "step": 225
    },
    {
      "epoch": 41.09090909090909,
      "grad_norm": 0.6644122004508972,
      "learning_rate": 8.815222438236726e-06,
      "loss": 0.8649,
      "step": 226
    },
    {
      "epoch": 41.45454545454545,
      "grad_norm": 0.6619220972061157,
      "learning_rate": 8.784975278258783e-06,
      "loss": 0.8085,
      "step": 228
    },
    {
      "epoch": 41.81818181818182,
      "grad_norm": 0.6005414724349976,
      "learning_rate": 8.754400164907496e-06,
      "loss": 0.8481,
      "step": 230
    },
    {
      "epoch": 42.0,
      "eval_loss": 0.9222747683525085,
      "eval_runtime": 9.6426,
      "eval_samples_per_second": 2.489,
      "eval_steps_per_second": 2.489,
      "step": 231
    },
    {
      "epoch": 42.18181818181818,
      "grad_norm": 0.722902238368988,
      "learning_rate": 8.723499747397415e-06,
      "loss": 0.8578,
      "step": 232
    },
    {
      "epoch": 42.54545454545455,
      "grad_norm": 0.7436155080795288,
      "learning_rate": 8.692276703129421e-06,
      "loss": 0.7996,
      "step": 234
    },
    {
      "epoch": 42.90909090909091,
      "grad_norm": 0.6658902168273926,
      "learning_rate": 8.660733737458751e-06,
      "loss": 0.8228,
      "step": 236
    },
    {
      "epoch": 42.90909090909091,
      "eval_loss": 0.9217340350151062,
      "eval_runtime": 9.6277,
      "eval_samples_per_second": 2.493,
      "eval_steps_per_second": 2.493,
      "step": 236
    },
    {
      "epoch": 43.27272727272727,
      "grad_norm": 0.6352283358573914,
      "learning_rate": 8.628873583460593e-06,
      "loss": 0.8113,
      "step": 238
    },
    {
      "epoch": 43.63636363636363,
      "grad_norm": 1.0223489999771118,
      "learning_rate": 8.596699001693257e-06,
      "loss": 0.8149,
      "step": 240
    },
    {
      "epoch": 44.0,
      "grad_norm": 0.7334797978401184,
      "learning_rate": 8.564212779959003e-06,
      "loss": 0.8593,
      "step": 242
    },
    {
      "epoch": 44.0,
      "eval_loss": 0.9185922741889954,
      "eval_runtime": 9.6303,
      "eval_samples_per_second": 2.492,
      "eval_steps_per_second": 2.492,
      "step": 242
    },
    {
      "epoch": 44.36363636363637,
      "grad_norm": 0.7544272541999817,
      "learning_rate": 8.531417733062476e-06,
      "loss": 0.7958,
      "step": 244
    },
    {
      "epoch": 44.72727272727273,
      "grad_norm": 0.8189204335212708,
      "learning_rate": 8.498316702566828e-06,
      "loss": 0.8238,
      "step": 246
    },
    {
      "epoch": 44.90909090909091,
      "eval_loss": 0.9156233668327332,
      "eval_runtime": 9.6451,
      "eval_samples_per_second": 2.488,
      "eval_steps_per_second": 2.488,
      "step": 247
    },
    {
      "epoch": 45.09090909090909,
      "grad_norm": 0.6729193329811096,
      "learning_rate": 8.464912556547486e-06,
      "loss": 0.835,
      "step": 248
    },
    {
      "epoch": 45.45454545454545,
      "grad_norm": 0.6723213195800781,
      "learning_rate": 8.43120818934367e-06,
      "loss": 0.7991,
      "step": 250
    },
    {
      "epoch": 45.81818181818182,
      "grad_norm": 0.8917332887649536,
      "learning_rate": 8.397206521307584e-06,
      "loss": 0.8081,
      "step": 252
    },
    {
      "epoch": 46.0,
      "eval_loss": 0.9161267876625061,
      "eval_runtime": 9.6325,
      "eval_samples_per_second": 2.492,
      "eval_steps_per_second": 2.492,
      "step": 253
    },
    {
      "epoch": 46.18181818181818,
      "grad_norm": 0.7718498110771179,
      "learning_rate": 8.362910498551402e-06,
      "loss": 0.8071,
      "step": 254
    },
    {
      "epoch": 46.54545454545455,
      "grad_norm": 0.7421916127204895,
      "learning_rate": 8.328323092691985e-06,
      "loss": 0.7838,
      "step": 256
    },
    {
      "epoch": 46.90909090909091,
      "grad_norm": 0.775203287601471,
      "learning_rate": 8.293447300593402e-06,
      "loss": 0.8327,
      "step": 258
    },
    {
      "epoch": 46.90909090909091,
      "eval_loss": 0.912854015827179,
      "eval_runtime": 9.6464,
      "eval_samples_per_second": 2.488,
      "eval_steps_per_second": 2.488,
      "step": 258
    },
    {
      "epoch": 47.27272727272727,
      "grad_norm": 0.6994480490684509,
      "learning_rate": 8.258286144107277e-06,
      "loss": 0.7949,
      "step": 260
    },
    {
      "epoch": 47.63636363636363,
      "grad_norm": 0.8607519865036011,
      "learning_rate": 8.222842669810936e-06,
      "loss": 0.7794,
      "step": 262
    },
    {
      "epoch": 48.0,
      "grad_norm": 0.8172978758811951,
      "learning_rate": 8.18711994874345e-06,
      "loss": 0.8029,
      "step": 264
    },
    {
      "epoch": 48.0,
      "eval_loss": 0.9110000133514404,
      "eval_runtime": 9.6168,
      "eval_samples_per_second": 2.496,
      "eval_steps_per_second": 2.496,
      "step": 264
    },
    {
      "epoch": 48.36363636363637,
      "grad_norm": 0.8061463236808777,
      "learning_rate": 8.151121076139534e-06,
      "loss": 0.8099,
      "step": 266
    },
    {
      "epoch": 48.72727272727273,
      "grad_norm": 0.9735673069953918,
      "learning_rate": 8.11484917116136e-06,
      "loss": 0.7909,
      "step": 268
    },
    {
      "epoch": 48.90909090909091,
      "eval_loss": 0.9093864560127258,
      "eval_runtime": 9.6205,
      "eval_samples_per_second": 2.495,
      "eval_steps_per_second": 2.495,
      "step": 269
    },
    {
      "epoch": 49.09090909090909,
      "grad_norm": 0.8723132014274597,
      "learning_rate": 8.078307376628292e-06,
      "loss": 0.7628,
      "step": 270
    },
    {
      "epoch": 49.45454545454545,
      "grad_norm": 0.7607284188270569,
      "learning_rate": 8.041498858744572e-06,
      "loss": 0.7665,
      "step": 272
    },
    {
      "epoch": 49.81818181818182,
      "grad_norm": 0.8277180194854736,
      "learning_rate": 8.004426806824985e-06,
      "loss": 0.7826,
      "step": 274
    },
    {
      "epoch": 50.0,
      "eval_loss": 0.9079095721244812,
      "eval_runtime": 9.6142,
      "eval_samples_per_second": 2.496,
      "eval_steps_per_second": 2.496,
      "step": 275
    },
    {
      "epoch": 50.18181818181818,
      "grad_norm": 0.8411371111869812,
      "learning_rate": 7.967094433018508e-06,
      "loss": 0.7943,
      "step": 276
    },
    {
      "epoch": 50.54545454545455,
      "grad_norm": 0.834507167339325,
      "learning_rate": 7.929504972030003e-06,
      "loss": 0.7586,
      "step": 278
    },
    {
      "epoch": 50.90909090909091,
      "grad_norm": 1.0113625526428223,
      "learning_rate": 7.891661680839932e-06,
      "loss": 0.773,
      "step": 280
    },
    {
      "epoch": 50.90909090909091,
      "eval_loss": 0.9122073650360107,
      "eval_runtime": 9.6327,
      "eval_samples_per_second": 2.492,
      "eval_steps_per_second": 2.492,
      "step": 280
    },
    {
      "epoch": 51.27272727272727,
      "grad_norm": 0.8380469083786011,
      "learning_rate": 7.85356783842216e-06,
      "loss": 0.7737,
      "step": 282
    },
    {
      "epoch": 51.63636363636363,
      "grad_norm": 0.8534033894538879,
      "learning_rate": 7.815226745459831e-06,
      "loss": 0.7941,
      "step": 284
    },
    {
      "epoch": 52.0,
      "grad_norm": 0.8890909552574158,
      "learning_rate": 7.776641724059398e-06,
      "loss": 0.7377,
      "step": 286
    },
    {
      "epoch": 52.0,
      "eval_loss": 0.9077624678611755,
      "eval_runtime": 9.6175,
      "eval_samples_per_second": 2.495,
      "eval_steps_per_second": 2.495,
      "step": 286
    },
    {
      "epoch": 52.36363636363637,
      "grad_norm": 0.9524686336517334,
      "learning_rate": 7.737816117462752e-06,
      "loss": 0.7699,
      "step": 288
    },
    {
      "epoch": 52.72727272727273,
      "grad_norm": 0.8625631928443909,
      "learning_rate": 7.698753289757565e-06,
      "loss": 0.7491,
      "step": 290
    },
    {
      "epoch": 52.90909090909091,
      "eval_loss": 0.9050046801567078,
      "eval_runtime": 9.6225,
      "eval_samples_per_second": 2.494,
      "eval_steps_per_second": 2.494,
      "step": 291
    },
    {
      "epoch": 53.09090909090909,
      "grad_norm": 1.0375274419784546,
      "learning_rate": 7.65945662558579e-06,
      "loss": 0.7661,
      "step": 292
    },
    {
      "epoch": 53.45454545454545,
      "grad_norm": 0.8255937695503235,
      "learning_rate": 7.619929529850397e-06,
      "loss": 0.7606,
      "step": 294
    },
    {
      "epoch": 53.81818181818182,
      "grad_norm": 1.0094412565231323,
      "learning_rate": 7.580175427420358e-06,
      "loss": 0.7414,
      "step": 296
    },
    {
      "epoch": 54.0,
      "eval_loss": 0.9093080163002014,
      "eval_runtime": 9.6164,
      "eval_samples_per_second": 2.496,
      "eval_steps_per_second": 2.496,
      "step": 297
    },
    {
      "epoch": 54.18181818181818,
      "grad_norm": 0.8360889554023743,
      "learning_rate": 7.54019776283389e-06,
      "loss": 0.7467,
      "step": 298
    },
    {
      "epoch": 54.54545454545455,
      "grad_norm": 0.9806857109069824,
      "learning_rate": 7.500000000000001e-06,
      "loss": 0.7445,
      "step": 300
    },
    {
      "epoch": 54.90909090909091,
      "grad_norm": 1.0003647804260254,
      "learning_rate": 7.459585621898353e-06,
      "loss": 0.7275,
      "step": 302
    },
    {
      "epoch": 54.90909090909091,
      "eval_loss": 0.9052907824516296,
      "eval_runtime": 9.6259,
      "eval_samples_per_second": 2.493,
      "eval_steps_per_second": 2.493,
      "step": 302
    },
    {
      "epoch": 55.27272727272727,
      "grad_norm": 1.080519437789917,
      "learning_rate": 7.418958130277483e-06,
      "loss": 0.7526,
      "step": 304
    },
    {
      "epoch": 55.63636363636363,
      "grad_norm": 1.2365224361419678,
      "learning_rate": 7.378121045351378e-06,
      "loss": 0.7289,
      "step": 306
    },
    {
      "epoch": 56.0,
      "grad_norm": 0.9447788000106812,
      "learning_rate": 7.337077905494472e-06,
      "loss": 0.7198,
      "step": 308
    },
    {
      "epoch": 56.0,
      "eval_loss": 0.9046055674552917,
      "eval_runtime": 9.6309,
      "eval_samples_per_second": 2.492,
      "eval_steps_per_second": 2.492,
      "step": 308
    },
    {
      "epoch": 56.36363636363637,
      "grad_norm": 0.9524107575416565,
      "learning_rate": 7.295832266935059e-06,
      "loss": 0.766,
      "step": 310
    },
    {
      "epoch": 56.72727272727273,
      "grad_norm": 0.9705169796943665,
      "learning_rate": 7.254387703447154e-06,
      "loss": 0.7203,
      "step": 312
    },
    {
      "epoch": 56.90909090909091,
      "eval_loss": 0.9092791676521301,
      "eval_runtime": 9.6325,
      "eval_samples_per_second": 2.492,
      "eval_steps_per_second": 2.492,
      "step": 313
    },
    {
      "epoch": 57.09090909090909,
      "grad_norm": 1.0789105892181396,
      "learning_rate": 7.212747806040845e-06,
      "loss": 0.6951,
      "step": 314
    },
    {
      "epoch": 57.45454545454545,
      "grad_norm": 1.1204413175582886,
      "learning_rate": 7.170916182651141e-06,
      "loss": 0.7249,
      "step": 316
    },
    {
      "epoch": 57.81818181818182,
      "grad_norm": 1.0801540613174438,
      "learning_rate": 7.128896457825364e-06,
      "loss": 0.6903,
      "step": 318
    },
    {
      "epoch": 58.0,
      "eval_loss": 0.9042022824287415,
      "eval_runtime": 9.621,
      "eval_samples_per_second": 2.495,
      "eval_steps_per_second": 2.495,
      "step": 319
    },
    {
      "epoch": 58.18181818181818,
      "grad_norm": 1.052799105644226,
      "learning_rate": 7.08669227240909e-06,
      "loss": 0.7306,
      "step": 320
    },
    {
      "epoch": 58.54545454545455,
      "grad_norm": 1.020494818687439,
      "learning_rate": 7.04430728323069e-06,
      "loss": 0.7288,
      "step": 322
    },
    {
      "epoch": 58.90909090909091,
      "grad_norm": 1.0670812129974365,
      "learning_rate": 7.0017451627844765e-06,
      "loss": 0.6987,
      "step": 324
    },
    {
      "epoch": 58.90909090909091,
      "eval_loss": 0.9106718897819519,
      "eval_runtime": 9.6324,
      "eval_samples_per_second": 2.492,
      "eval_steps_per_second": 2.492,
      "step": 324
    },
    {
      "epoch": 59.27272727272727,
      "grad_norm": 1.0681415796279907,
      "learning_rate": 6.959009598912493e-06,
      "loss": 0.6906,
      "step": 326
    },
    {
      "epoch": 59.63636363636363,
      "grad_norm": 1.1053000688552856,
      "learning_rate": 6.916104294484988e-06,
      "loss": 0.7063,
      "step": 328
    },
    {
      "epoch": 60.0,
      "grad_norm": 1.0191996097564697,
      "learning_rate": 6.873032967079562e-06,
      "loss": 0.7141,
      "step": 330
    },
    {
      "epoch": 60.0,
      "eval_loss": 0.9078884124755859,
      "eval_runtime": 9.6326,
      "eval_samples_per_second": 2.492,
      "eval_steps_per_second": 2.492,
      "step": 330
    },
    {
      "epoch": 60.36363636363637,
      "grad_norm": 1.0764459371566772,
      "learning_rate": 6.829799348659061e-06,
      "loss": 0.7079,
      "step": 332
    },
    {
      "epoch": 60.72727272727273,
      "grad_norm": 1.146618366241455,
      "learning_rate": 6.7864071852482205e-06,
      "loss": 0.7023,
      "step": 334
    },
    {
      "epoch": 60.90909090909091,
      "eval_loss": 0.9119828343391418,
      "eval_runtime": 9.6211,
      "eval_samples_per_second": 2.495,
      "eval_steps_per_second": 2.495,
      "step": 335
    },
    {
      "epoch": 61.09090909090909,
      "grad_norm": 1.274398684501648,
      "learning_rate": 6.7428602366090764e-06,
      "loss": 0.6856,
      "step": 336
    },
    {
      "epoch": 61.45454545454545,
      "grad_norm": 1.1239506006240845,
      "learning_rate": 6.699162275915208e-06,
      "loss": 0.6603,
      "step": 338
    },
    {
      "epoch": 61.81818181818182,
      "grad_norm": 1.3075493574142456,
      "learning_rate": 6.655317089424791e-06,
      "loss": 0.6945,
      "step": 340
    },
    {
      "epoch": 62.0,
      "eval_loss": 0.9086711406707764,
      "eval_runtime": 9.6164,
      "eval_samples_per_second": 2.496,
      "eval_steps_per_second": 2.496,
      "step": 341
    },
    {
      "epoch": 62.18181818181818,
      "grad_norm": 1.1385191679000854,
      "learning_rate": 6.611328476152557e-06,
      "loss": 0.7058,
      "step": 342
    },
    {
      "epoch": 62.54545454545455,
      "grad_norm": 1.2736291885375977,
      "learning_rate": 6.567200247540599e-06,
      "loss": 0.6662,
      "step": 344
    },
    {
      "epoch": 62.90909090909091,
      "grad_norm": 1.1537060737609863,
      "learning_rate": 6.522936227128139e-06,
      "loss": 0.6897,
      "step": 346
    },
    {
      "epoch": 62.90909090909091,
      "eval_loss": 0.9129719734191895,
      "eval_runtime": 9.6291,
      "eval_samples_per_second": 2.492,
      "eval_steps_per_second": 2.492,
      "step": 346
    },
    {
      "epoch": 63.27272727272727,
      "grad_norm": 1.4773812294006348,
      "learning_rate": 6.4785402502202345e-06,
      "loss": 0.6822,
      "step": 348
    },
    {
      "epoch": 63.63636363636363,
      "grad_norm": 1.3053030967712402,
      "learning_rate": 6.434016163555452e-06,
      "loss": 0.6596,
      "step": 350
    },
    {
      "epoch": 64.0,
      "grad_norm": 1.1616432666778564,
      "learning_rate": 6.389367824972575e-06,
      "loss": 0.6597,
      "step": 352
    },
    {
      "epoch": 64.0,
      "eval_loss": 0.9133894443511963,
      "eval_runtime": 9.6317,
      "eval_samples_per_second": 2.492,
      "eval_steps_per_second": 2.492,
      "step": 352
    },
    {
      "epoch": 64.36363636363636,
      "grad_norm": 1.0900788307189941,
      "learning_rate": 6.344599103076329e-06,
      "loss": 0.6563,
      "step": 354
    },
    {
      "epoch": 64.72727272727273,
      "grad_norm": 1.2286537885665894,
      "learning_rate": 6.299713876902188e-06,
      "loss": 0.6954,
      "step": 356
    },
    {
      "epoch": 64.9090909090909,
      "eval_loss": 0.9120491147041321,
      "eval_runtime": 9.6347,
      "eval_samples_per_second": 2.491,
      "eval_steps_per_second": 2.491,
      "step": 357
    },
    {
      "epoch": 64.9090909090909,
      "step": 357,
      "total_flos": 8.780093794235187e+16,
      "train_loss": 1.1563805774146436,
      "train_runtime": 6696.4639,
      "train_samples_per_second": 1.971,
      "train_steps_per_second": 0.112
    }
  ],
  "logging_steps": 2,
  "max_steps": 750,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 150,
  "save_steps": 25,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 7,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 0
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 8.780093794235187e+16,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}