{
  "best_metric": 0.40008437633514404,
  "best_model_checkpoint": "mikhail-panzo/zlm-ceb_b64_le5_s8000/checkpoint-6000",
  "epoch": 313.72549019607845,
  "eval_steps": 500,
  "global_step": 8000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.9607843137254903,
      "grad_norm": 1.6111680269241333,
      "learning_rate": 2.5e-06,
      "loss": 0.821,
      "step": 50
    },
    {
      "epoch": 3.9215686274509802,
      "grad_norm": 1.9582017660140991,
      "learning_rate": 5e-06,
      "loss": 0.673,
      "step": 100
    },
    {
      "epoch": 5.882352941176471,
      "grad_norm": 0.9988609552383423,
      "learning_rate": 7.5e-06,
      "loss": 0.5518,
      "step": 150
    },
    {
      "epoch": 7.8431372549019605,
      "grad_norm": 1.0111173391342163,
      "learning_rate": 1e-05,
      "loss": 0.5172,
      "step": 200
    },
    {
      "epoch": 9.803921568627452,
      "grad_norm": 0.9361169338226318,
      "learning_rate": 1.25e-05,
      "loss": 0.4999,
      "step": 250
    },
    {
      "epoch": 11.764705882352942,
      "grad_norm": 0.8657887578010559,
      "learning_rate": 1.5e-05,
      "loss": 0.4911,
      "step": 300
    },
    {
      "epoch": 13.72549019607843,
      "grad_norm": 0.8616572618484497,
      "learning_rate": 1.75e-05,
      "loss": 0.4757,
      "step": 350
    },
    {
      "epoch": 15.686274509803921,
      "grad_norm": 1.100374698638916,
      "learning_rate": 2e-05,
      "loss": 0.47,
      "step": 400
    },
    {
      "epoch": 17.647058823529413,
      "grad_norm": 1.3860764503479004,
      "learning_rate": 2.25e-05,
      "loss": 0.4671,
      "step": 450
    },
    {
      "epoch": 19.607843137254903,
      "grad_norm": 1.2481627464294434,
      "learning_rate": 2.5e-05,
      "loss": 0.4626,
      "step": 500
    },
    {
      "epoch": 19.607843137254903,
      "eval_loss": 0.42633864283561707,
      "eval_runtime": 6.9188,
      "eval_samples_per_second": 26.016,
      "eval_steps_per_second": 3.324,
      "step": 500
    },
    {
      "epoch": 21.568627450980394,
      "grad_norm": 1.079897165298462,
      "learning_rate": 2.7500000000000004e-05,
      "loss": 0.4593,
      "step": 550
    },
    {
      "epoch": 23.529411764705884,
      "grad_norm": 0.9019867777824402,
      "learning_rate": 3e-05,
      "loss": 0.4474,
      "step": 600
    },
    {
      "epoch": 25.49019607843137,
      "grad_norm": 1.900229811668396,
      "learning_rate": 3.2500000000000004e-05,
      "loss": 0.4439,
      "step": 650
    },
    {
      "epoch": 27.45098039215686,
      "grad_norm": 1.3894062042236328,
      "learning_rate": 3.5e-05,
      "loss": 0.4397,
      "step": 700
    },
    {
      "epoch": 29.41176470588235,
      "grad_norm": 2.318305253982544,
      "learning_rate": 3.7500000000000003e-05,
      "loss": 0.4376,
      "step": 750
    },
    {
      "epoch": 31.372549019607842,
      "grad_norm": 0.9606600999832153,
      "learning_rate": 4e-05,
      "loss": 0.4394,
      "step": 800
    },
    {
      "epoch": 33.333333333333336,
      "grad_norm": 1.4766446352005005,
      "learning_rate": 4.25e-05,
      "loss": 0.4376,
      "step": 850
    },
    {
      "epoch": 35.294117647058826,
      "grad_norm": 1.1099992990493774,
      "learning_rate": 4.5e-05,
      "loss": 0.4333,
      "step": 900
    },
    {
      "epoch": 37.254901960784316,
      "grad_norm": 1.1501879692077637,
      "learning_rate": 4.75e-05,
      "loss": 0.4281,
      "step": 950
    },
    {
      "epoch": 39.21568627450981,
      "grad_norm": 0.9793707728385925,
      "learning_rate": 5e-05,
      "loss": 0.4288,
      "step": 1000
    },
    {
      "epoch": 39.21568627450981,
      "eval_loss": 0.4076782763004303,
      "eval_runtime": 6.5279,
      "eval_samples_per_second": 27.574,
      "eval_steps_per_second": 3.523,
      "step": 1000
    },
    {
      "epoch": 41.1764705882353,
      "grad_norm": 1.812929630279541,
      "learning_rate": 5.25e-05,
      "loss": 0.4224,
      "step": 1050
    },
    {
      "epoch": 43.13725490196079,
      "grad_norm": 1.066340446472168,
      "learning_rate": 5.500000000000001e-05,
      "loss": 0.4231,
      "step": 1100
    },
    {
      "epoch": 45.09803921568628,
      "grad_norm": 2.465958595275879,
      "learning_rate": 5.745e-05,
      "loss": 0.4261,
      "step": 1150
    },
    {
      "epoch": 47.05882352941177,
      "grad_norm": 1.3723503351211548,
      "learning_rate": 5.995000000000001e-05,
      "loss": 0.416,
      "step": 1200
    },
    {
      "epoch": 49.01960784313726,
      "grad_norm": 1.6065536737442017,
      "learning_rate": 6.245000000000001e-05,
      "loss": 0.416,
      "step": 1250
    },
    {
      "epoch": 50.98039215686274,
      "grad_norm": 3.8591885566711426,
      "learning_rate": 6.494999999999999e-05,
      "loss": 0.4162,
      "step": 1300
    },
    {
      "epoch": 52.94117647058823,
      "grad_norm": 2.415721893310547,
      "learning_rate": 6.745e-05,
      "loss": 0.4121,
      "step": 1350
    },
    {
      "epoch": 54.90196078431372,
      "grad_norm": 1.969329833984375,
      "learning_rate": 6.995e-05,
      "loss": 0.4104,
      "step": 1400
    },
    {
      "epoch": 56.86274509803921,
      "grad_norm": 3.190781593322754,
      "learning_rate": 7.245000000000001e-05,
      "loss": 0.4055,
      "step": 1450
    },
    {
      "epoch": 58.8235294117647,
      "grad_norm": 1.4585760831832886,
      "learning_rate": 7.495e-05,
      "loss": 0.4109,
      "step": 1500
    },
    {
      "epoch": 58.8235294117647,
      "eval_loss": 0.4012959897518158,
      "eval_runtime": 6.3508,
      "eval_samples_per_second": 28.343,
      "eval_steps_per_second": 3.622,
      "step": 1500
    },
    {
      "epoch": 60.78431372549019,
      "grad_norm": 4.141477584838867,
      "learning_rate": 7.745e-05,
      "loss": 0.4117,
      "step": 1550
    },
    {
      "epoch": 62.745098039215684,
      "grad_norm": 5.334190845489502,
      "learning_rate": 7.995e-05,
      "loss": 0.413,
      "step": 1600
    },
    {
      "epoch": 64.70588235294117,
      "grad_norm": 3.1039328575134277,
      "learning_rate": 8.245e-05,
      "loss": 0.405,
      "step": 1650
    },
    {
      "epoch": 66.66666666666667,
      "grad_norm": 2.6190948486328125,
      "learning_rate": 8.495e-05,
      "loss": 0.4098,
      "step": 1700
    },
    {
      "epoch": 68.62745098039215,
      "grad_norm": 1.8761863708496094,
      "learning_rate": 8.745000000000001e-05,
      "loss": 0.4035,
      "step": 1750
    },
    {
      "epoch": 70.58823529411765,
      "grad_norm": 0.9362647533416748,
      "learning_rate": 8.995e-05,
      "loss": 0.4253,
      "step": 1800
    },
    {
      "epoch": 72.54901960784314,
      "grad_norm": 1.6635262966156006,
      "learning_rate": 9.245e-05,
      "loss": 0.4295,
      "step": 1850
    },
    {
      "epoch": 74.50980392156863,
      "grad_norm": 1.3353509902954102,
      "learning_rate": 9.495e-05,
      "loss": 0.4078,
      "step": 1900
    },
    {
      "epoch": 76.47058823529412,
      "grad_norm": 8.328239440917969,
      "learning_rate": 9.745000000000001e-05,
      "loss": 0.4051,
      "step": 1950
    },
    {
      "epoch": 78.43137254901961,
      "grad_norm": 1.1230006217956543,
      "learning_rate": 9.995e-05,
      "loss": 0.3978,
      "step": 2000
    },
    {
      "epoch": 78.43137254901961,
      "eval_loss": 0.40350666642189026,
      "eval_runtime": 6.5593,
      "eval_samples_per_second": 27.442,
      "eval_steps_per_second": 3.506,
      "step": 2000
    },
    {
      "epoch": 80.3921568627451,
      "grad_norm": 3.8058626651763916,
      "learning_rate": 9.918333333333334e-05,
      "loss": 0.3997,
      "step": 2050
    },
    {
      "epoch": 82.3529411764706,
      "grad_norm": 2.170177936553955,
      "learning_rate": 9.835e-05,
      "loss": 0.3988,
      "step": 2100
    },
    {
      "epoch": 84.31372549019608,
      "grad_norm": 0.6957865357398987,
      "learning_rate": 9.751666666666666e-05,
      "loss": 0.3938,
      "step": 2150
    },
    {
      "epoch": 86.27450980392157,
      "grad_norm": 1.1778301000595093,
      "learning_rate": 9.668333333333334e-05,
      "loss": 0.4008,
      "step": 2200
    },
    {
      "epoch": 88.23529411764706,
      "grad_norm": 1.1749225854873657,
      "learning_rate": 9.585000000000001e-05,
      "loss": 0.406,
      "step": 2250
    },
    {
      "epoch": 90.19607843137256,
      "grad_norm": 1.1235101222991943,
      "learning_rate": 9.501666666666668e-05,
      "loss": 0.3906,
      "step": 2300
    },
    {
      "epoch": 92.15686274509804,
      "grad_norm": 5.676591873168945,
      "learning_rate": 9.418333333333334e-05,
      "loss": 0.3954,
      "step": 2350
    },
    {
      "epoch": 94.11764705882354,
      "grad_norm": 1.4196738004684448,
      "learning_rate": 9.335e-05,
      "loss": 0.4018,
      "step": 2400
    },
    {
      "epoch": 96.07843137254902,
      "grad_norm": 2.848329544067383,
      "learning_rate": 9.251666666666667e-05,
      "loss": 0.387,
      "step": 2450
    },
    {
      "epoch": 98.03921568627452,
      "grad_norm": 1.7598848342895508,
      "learning_rate": 9.168333333333333e-05,
      "loss": 0.3898,
      "step": 2500
    },
    {
      "epoch": 98.03921568627452,
      "eval_loss": 0.40130773186683655,
      "eval_runtime": 6.4746,
      "eval_samples_per_second": 27.801,
      "eval_steps_per_second": 3.552,
      "step": 2500
    },
    {
      "epoch": 100.0,
      "grad_norm": 5.990013599395752,
      "learning_rate": 9.085e-05,
      "loss": 0.3982,
      "step": 2550
    },
    {
      "epoch": 101.96078431372548,
      "grad_norm": 2.984307050704956,
      "learning_rate": 9.001666666666667e-05,
      "loss": 0.3889,
      "step": 2600
    },
    {
      "epoch": 103.92156862745098,
      "grad_norm": 1.976651668548584,
      "learning_rate": 8.918333333333334e-05,
      "loss": 0.382,
      "step": 2650
    },
    {
      "epoch": 105.88235294117646,
      "grad_norm": 1.1698123216629028,
      "learning_rate": 8.834999999999999e-05,
      "loss": 0.3877,
      "step": 2700
    },
    {
      "epoch": 107.84313725490196,
      "grad_norm": 1.8311396837234497,
      "learning_rate": 8.751666666666668e-05,
      "loss": 0.3848,
      "step": 2750
    },
    {
      "epoch": 109.80392156862744,
      "grad_norm": 3.1962406635284424,
      "learning_rate": 8.668333333333334e-05,
      "loss": 0.384,
      "step": 2800
    },
    {
      "epoch": 111.76470588235294,
      "grad_norm": 1.4986674785614014,
      "learning_rate": 8.585000000000001e-05,
      "loss": 0.3737,
      "step": 2850
    },
    {
      "epoch": 113.72549019607843,
      "grad_norm": 4.846043109893799,
      "learning_rate": 8.501666666666667e-05,
      "loss": 0.3787,
      "step": 2900
    },
    {
      "epoch": 115.68627450980392,
      "grad_norm": 2.5160129070281982,
      "learning_rate": 8.418333333333334e-05,
      "loss": 0.3707,
      "step": 2950
    },
    {
      "epoch": 117.6470588235294,
      "grad_norm": 1.209688663482666,
      "learning_rate": 8.335e-05,
      "loss": 0.373,
      "step": 3000
    },
    {
      "epoch": 117.6470588235294,
      "eval_loss": 0.40103283524513245,
      "eval_runtime": 6.6876,
      "eval_samples_per_second": 26.916,
      "eval_steps_per_second": 3.439,
      "step": 3000
    },
    {
      "epoch": 119.6078431372549,
      "grad_norm": 1.4071707725524902,
      "learning_rate": 8.251666666666668e-05,
      "loss": 0.3731,
      "step": 3050
    },
    {
      "epoch": 121.56862745098039,
      "grad_norm": 2.3906798362731934,
      "learning_rate": 8.168333333333333e-05,
      "loss": 0.3705,
      "step": 3100
    },
    {
      "epoch": 123.52941176470588,
      "grad_norm": 2.833566188812256,
      "learning_rate": 8.085e-05,
      "loss": 0.3754,
      "step": 3150
    },
    {
      "epoch": 125.49019607843137,
      "grad_norm": 1.7871581315994263,
      "learning_rate": 8.001666666666667e-05,
      "loss": 0.3734,
      "step": 3200
    },
    {
      "epoch": 127.45098039215686,
      "grad_norm": 1.0529006719589233,
      "learning_rate": 7.918333333333334e-05,
      "loss": 0.3792,
      "step": 3250
    },
    {
      "epoch": 129.41176470588235,
      "grad_norm": 2.939656972885132,
      "learning_rate": 7.835000000000001e-05,
      "loss": 0.3693,
      "step": 3300
    },
    {
      "epoch": 131.37254901960785,
      "grad_norm": 0.9791694283485413,
      "learning_rate": 7.751666666666668e-05,
      "loss": 0.366,
      "step": 3350
    },
    {
      "epoch": 133.33333333333334,
      "grad_norm": 1.6775403022766113,
      "learning_rate": 7.668333333333335e-05,
      "loss": 0.3709,
      "step": 3400
    },
    {
      "epoch": 135.2941176470588,
      "grad_norm": 1.0679121017456055,
      "learning_rate": 7.585e-05,
      "loss": 0.3677,
      "step": 3450
    },
    {
      "epoch": 137.2549019607843,
      "grad_norm": 2.068532705307007,
      "learning_rate": 7.501666666666667e-05,
      "loss": 0.3644,
      "step": 3500
    },
    {
      "epoch": 137.2549019607843,
      "eval_loss": 0.4005117118358612,
      "eval_runtime": 6.7669,
      "eval_samples_per_second": 26.6,
      "eval_steps_per_second": 3.399,
      "step": 3500
    },
    {
      "epoch": 139.2156862745098,
      "grad_norm": 0.9116536378860474,
      "learning_rate": 7.418333333333334e-05,
      "loss": 0.365,
      "step": 3550
    },
    {
      "epoch": 141.1764705882353,
      "grad_norm": 1.6524943113327026,
      "learning_rate": 7.335000000000001e-05,
      "loss": 0.362,
      "step": 3600
    },
    {
      "epoch": 143.13725490196077,
      "grad_norm": 0.9939906597137451,
      "learning_rate": 7.251666666666666e-05,
      "loss": 0.3641,
      "step": 3650
    },
    {
      "epoch": 145.09803921568627,
      "grad_norm": 1.7879794836044312,
      "learning_rate": 7.168333333333333e-05,
      "loss": 0.3649,
      "step": 3700
    },
    {
      "epoch": 147.05882352941177,
      "grad_norm": 2.4735560417175293,
      "learning_rate": 7.085e-05,
      "loss": 0.3591,
      "step": 3750
    },
    {
      "epoch": 149.01960784313727,
      "grad_norm": 1.4222075939178467,
      "learning_rate": 7.001666666666667e-05,
      "loss": 0.3605,
      "step": 3800
    },
    {
      "epoch": 150.98039215686273,
      "grad_norm": 2.095998525619507,
      "learning_rate": 6.918333333333334e-05,
      "loss": 0.3597,
      "step": 3850
    },
    {
      "epoch": 152.94117647058823,
      "grad_norm": 3.8910038471221924,
      "learning_rate": 6.835000000000001e-05,
      "loss": 0.3605,
      "step": 3900
    },
    {
      "epoch": 154.90196078431373,
      "grad_norm": 0.7127563953399658,
      "learning_rate": 6.751666666666668e-05,
      "loss": 0.3629,
      "step": 3950
    },
    {
      "epoch": 156.86274509803923,
      "grad_norm": 1.66582190990448,
      "learning_rate": 6.668333333333333e-05,
      "loss": 0.3569,
      "step": 4000
    },
    {
      "epoch": 156.86274509803923,
      "eval_loss": 0.4029478132724762,
      "eval_runtime": 6.4438,
      "eval_samples_per_second": 27.934,
      "eval_steps_per_second": 3.569,
      "step": 4000
    },
    {
      "epoch": 158.8235294117647,
      "grad_norm": 1.4073445796966553,
      "learning_rate": 6.585e-05,
      "loss": 0.3648,
      "step": 4050
    },
    {
      "epoch": 160.7843137254902,
      "grad_norm": 1.070461392402649,
      "learning_rate": 6.501666666666667e-05,
      "loss": 0.3569,
      "step": 4100
    },
    {
      "epoch": 162.7450980392157,
      "grad_norm": 2.141096353530884,
      "learning_rate": 6.418333333333334e-05,
      "loss": 0.3607,
      "step": 4150
    },
    {
      "epoch": 164.7058823529412,
      "grad_norm": 1.1931374073028564,
      "learning_rate": 6.335e-05,
      "loss": 0.3575,
      "step": 4200
    },
    {
      "epoch": 166.66666666666666,
      "grad_norm": 1.0879640579223633,
      "learning_rate": 6.251666666666666e-05,
      "loss": 0.356,
      "step": 4250
    },
    {
      "epoch": 168.62745098039215,
      "grad_norm": 1.5755019187927246,
      "learning_rate": 6.168333333333333e-05,
      "loss": 0.362,
      "step": 4300
    },
    {
      "epoch": 170.58823529411765,
      "grad_norm": 9.463594436645508,
      "learning_rate": 6.085000000000001e-05,
      "loss": 0.3612,
      "step": 4350
    },
    {
      "epoch": 172.54901960784315,
      "grad_norm": 5.174690246582031,
      "learning_rate": 6.0016666666666664e-05,
      "loss": 0.3537,
      "step": 4400
    },
    {
      "epoch": 174.50980392156862,
      "grad_norm": 1.342671513557434,
      "learning_rate": 5.918333333333333e-05,
      "loss": 0.3521,
      "step": 4450
    },
    {
      "epoch": 176.47058823529412,
      "grad_norm": 1.7866413593292236,
      "learning_rate": 5.835e-05,
      "loss": 0.3515,
      "step": 4500
    },
    {
      "epoch": 176.47058823529412,
      "eval_loss": 0.4038984179496765,
      "eval_runtime": 6.8698,
      "eval_samples_per_second": 26.202,
      "eval_steps_per_second": 3.348,
      "step": 4500
    },
    {
      "epoch": 178.4313725490196,
      "grad_norm": 2.6986000537872314,
      "learning_rate": 5.751666666666667e-05,
      "loss": 0.3539,
      "step": 4550
    },
    {
      "epoch": 180.3921568627451,
      "grad_norm": 1.255347728729248,
      "learning_rate": 5.668333333333333e-05,
      "loss": 0.3578,
      "step": 4600
    },
    {
      "epoch": 182.35294117647058,
      "grad_norm": 2.3221983909606934,
      "learning_rate": 5.585e-05,
      "loss": 0.3594,
      "step": 4650
    },
    {
      "epoch": 184.31372549019608,
      "grad_norm": 1.0801488161087036,
      "learning_rate": 5.501666666666667e-05,
      "loss": 0.3481,
      "step": 4700
    },
    {
      "epoch": 186.27450980392157,
      "grad_norm": 0.7307039499282837,
      "learning_rate": 5.4183333333333334e-05,
      "loss": 0.3486,
      "step": 4750
    },
    {
      "epoch": 188.23529411764707,
      "grad_norm": 0.9629709720611572,
      "learning_rate": 5.335e-05,
      "loss": 0.3465,
      "step": 4800
    },
    {
      "epoch": 190.19607843137254,
      "grad_norm": 0.5935477614402771,
      "learning_rate": 5.251666666666667e-05,
      "loss": 0.3481,
      "step": 4850
    },
    {
      "epoch": 192.15686274509804,
      "grad_norm": 1.965649127960205,
      "learning_rate": 5.168333333333334e-05,
      "loss": 0.3468,
      "step": 4900
    },
    {
      "epoch": 194.11764705882354,
      "grad_norm": 0.9675062894821167,
      "learning_rate": 5.0849999999999996e-05,
      "loss": 0.3529,
      "step": 4950
    },
    {
      "epoch": 196.07843137254903,
      "grad_norm": 0.7859313488006592,
      "learning_rate": 5.0016666666666665e-05,
      "loss": 0.3443,
      "step": 5000
    },
    {
      "epoch": 196.07843137254903,
      "eval_loss": 0.4005332589149475,
      "eval_runtime": 6.4088,
      "eval_samples_per_second": 28.086,
      "eval_steps_per_second": 3.589,
      "step": 5000
    },
    {
      "epoch": 198.0392156862745,
      "grad_norm": 0.9622226357460022,
      "learning_rate": 4.9183333333333334e-05,
      "loss": 0.349,
      "step": 5050
    },
    {
      "epoch": 200.0,
      "grad_norm": 0.7713702321052551,
      "learning_rate": 4.835e-05,
      "loss": 0.3446,
      "step": 5100
    },
    {
      "epoch": 201.9607843137255,
      "grad_norm": 1.2544292211532593,
      "learning_rate": 4.751666666666667e-05,
      "loss": 0.3443,
      "step": 5150
    },
    {
      "epoch": 203.92156862745097,
      "grad_norm": 0.7886931300163269,
      "learning_rate": 4.6683333333333334e-05,
      "loss": 0.3424,
      "step": 5200
    },
    {
      "epoch": 205.88235294117646,
      "grad_norm": 1.3799712657928467,
      "learning_rate": 4.585e-05,
      "loss": 0.348,
      "step": 5250
    },
    {
      "epoch": 207.84313725490196,
      "grad_norm": 0.9509637355804443,
      "learning_rate": 4.5016666666666665e-05,
      "loss": 0.3447,
      "step": 5300
    },
    {
      "epoch": 209.80392156862746,
      "grad_norm": 1.3633594512939453,
      "learning_rate": 4.4183333333333334e-05,
      "loss": 0.343,
      "step": 5350
    },
    {
      "epoch": 211.76470588235293,
      "grad_norm": 3.379939317703247,
      "learning_rate": 4.335e-05,
      "loss": 0.3451,
      "step": 5400
    },
    {
      "epoch": 213.72549019607843,
      "grad_norm": 0.7113842368125916,
      "learning_rate": 4.251666666666667e-05,
      "loss": 0.3429,
      "step": 5450
    },
    {
      "epoch": 215.68627450980392,
      "grad_norm": 0.8408060073852539,
      "learning_rate": 4.1683333333333335e-05,
      "loss": 0.3469,
      "step": 5500
    },
    {
      "epoch": 215.68627450980392,
      "eval_loss": 0.40183132886886597,
      "eval_runtime": 6.5032,
      "eval_samples_per_second": 27.679,
      "eval_steps_per_second": 3.537,
      "step": 5500
    },
    {
      "epoch": 217.64705882352942,
      "grad_norm": 1.2636085748672485,
      "learning_rate": 4.085e-05,
      "loss": 0.3371,
      "step": 5550
    },
    {
      "epoch": 219.6078431372549,
      "grad_norm": 2.5460524559020996,
      "learning_rate": 4.0016666666666666e-05,
      "loss": 0.3449,
      "step": 5600
    },
    {
      "epoch": 221.5686274509804,
      "grad_norm": 0.6922531723976135,
      "learning_rate": 3.9183333333333335e-05,
      "loss": 0.3432,
      "step": 5650
    },
    {
      "epoch": 223.52941176470588,
      "grad_norm": 0.5262002944946289,
      "learning_rate": 3.8350000000000004e-05,
      "loss": 0.3408,
      "step": 5700
    },
    {
      "epoch": 225.49019607843138,
      "grad_norm": 0.9883546233177185,
      "learning_rate": 3.7516666666666666e-05,
      "loss": 0.3399,
      "step": 5750
    },
    {
      "epoch": 227.45098039215685,
      "grad_norm": 0.7871970534324646,
      "learning_rate": 3.6683333333333335e-05,
      "loss": 0.3417,
      "step": 5800
    },
    {
      "epoch": 229.41176470588235,
      "grad_norm": 0.7385707497596741,
      "learning_rate": 3.585e-05,
      "loss": 0.3399,
      "step": 5850
    },
    {
      "epoch": 231.37254901960785,
      "grad_norm": 0.5304046869277954,
      "learning_rate": 3.501666666666667e-05,
      "loss": 0.3388,
      "step": 5900
    },
    {
      "epoch": 233.33333333333334,
      "grad_norm": 0.5726514458656311,
      "learning_rate": 3.4183333333333335e-05,
      "loss": 0.3361,
      "step": 5950
    },
    {
      "epoch": 235.2941176470588,
      "grad_norm": 0.49944621324539185,
      "learning_rate": 3.3350000000000004e-05,
      "loss": 0.3427,
      "step": 6000
    },
    {
      "epoch": 235.2941176470588,
      "eval_loss": 0.40008437633514404,
      "eval_runtime": 6.512,
      "eval_samples_per_second": 27.641,
      "eval_steps_per_second": 3.532,
      "step": 6000
    },
    {
      "epoch": 237.2549019607843,
      "grad_norm": 1.1355714797973633,
      "learning_rate": 3.2516666666666666e-05,
      "loss": 0.3377,
      "step": 6050
    },
    {
      "epoch": 239.2156862745098,
      "grad_norm": 0.7759362459182739,
      "learning_rate": 3.1683333333333335e-05,
      "loss": 0.339,
      "step": 6100
    },
    {
      "epoch": 241.1764705882353,
      "grad_norm": 1.1163146495819092,
      "learning_rate": 3.0850000000000004e-05,
      "loss": 0.337,
      "step": 6150
    },
    {
      "epoch": 243.13725490196077,
      "grad_norm": 0.8132624626159668,
      "learning_rate": 3.001666666666667e-05,
      "loss": 0.3378,
      "step": 6200
    },
    {
      "epoch": 245.09803921568627,
      "grad_norm": 0.7331820130348206,
      "learning_rate": 2.9183333333333336e-05,
      "loss": 0.3323,
      "step": 6250
    },
    {
      "epoch": 247.05882352941177,
      "grad_norm": 0.9192725419998169,
      "learning_rate": 2.8349999999999998e-05,
      "loss": 0.3382,
      "step": 6300
    },
    {
      "epoch": 249.01960784313727,
      "grad_norm": 3.9876034259796143,
      "learning_rate": 2.7516666666666667e-05,
      "loss": 0.3384,
      "step": 6350
    },
    {
      "epoch": 250.98039215686273,
      "grad_norm": 0.6646643280982971,
      "learning_rate": 2.6683333333333333e-05,
      "loss": 0.3365,
      "step": 6400
    },
    {
      "epoch": 252.94117647058823,
      "grad_norm": 0.5878006815910339,
      "learning_rate": 2.585e-05,
      "loss": 0.3403,
      "step": 6450
    },
    {
      "epoch": 254.90196078431373,
      "grad_norm": 0.8833298087120056,
      "learning_rate": 2.5016666666666667e-05,
      "loss": 0.3401,
      "step": 6500
    },
    {
      "epoch": 254.90196078431373,
      "eval_loss": 0.4041617810726166,
      "eval_runtime": 6.702,
      "eval_samples_per_second": 26.858,
      "eval_steps_per_second": 3.432,
      "step": 6500
    },
    {
      "epoch": 256.8627450980392,
      "grad_norm": 0.751977801322937,
      "learning_rate": 2.4183333333333336e-05,
      "loss": 0.3319,
      "step": 6550
    },
    {
      "epoch": 258.8235294117647,
      "grad_norm": 1.289437174797058,
      "learning_rate": 2.3350000000000002e-05,
      "loss": 0.3382,
      "step": 6600
    },
    {
      "epoch": 260.7843137254902,
      "grad_norm": 0.5142917633056641,
      "learning_rate": 2.2516666666666667e-05,
      "loss": 0.334,
      "step": 6650
    },
    {
      "epoch": 262.7450980392157,
      "grad_norm": 0.4598856568336487,
      "learning_rate": 2.1683333333333333e-05,
      "loss": 0.338,
      "step": 6700
    },
    {
      "epoch": 264.70588235294116,
      "grad_norm": 0.47424769401550293,
      "learning_rate": 2.085e-05,
      "loss": 0.3369,
      "step": 6750
    },
    {
      "epoch": 266.6666666666667,
      "grad_norm": 0.6651924848556519,
      "learning_rate": 2.0016666666666668e-05,
      "loss": 0.3379,
      "step": 6800
    },
    {
      "epoch": 268.62745098039215,
      "grad_norm": 0.6406331062316895,
      "learning_rate": 1.9183333333333333e-05,
      "loss": 0.3409,
      "step": 6850
    },
    {
      "epoch": 270.5882352941176,
      "grad_norm": 0.8162440061569214,
      "learning_rate": 1.8350000000000002e-05,
      "loss": 0.3336,
      "step": 6900
    },
    {
      "epoch": 272.54901960784315,
      "grad_norm": 0.5905545353889465,
      "learning_rate": 1.7516666666666668e-05,
      "loss": 0.3324,
      "step": 6950
    },
    {
      "epoch": 274.5098039215686,
      "grad_norm": 0.7422587871551514,
      "learning_rate": 1.6700000000000003e-05,
      "loss": 0.3419,
      "step": 7000
    },
    {
      "epoch": 274.5098039215686,
      "eval_loss": 0.40541204810142517,
      "eval_runtime": 6.8059,
      "eval_samples_per_second": 26.447,
      "eval_steps_per_second": 3.379,
      "step": 7000
    },
    {
      "epoch": 276.47058823529414,
      "grad_norm": 0.5827329158782959,
      "learning_rate": 1.586666666666667e-05,
      "loss": 0.3372,
      "step": 7050
    },
    {
      "epoch": 278.4313725490196,
      "grad_norm": 0.6354573369026184,
      "learning_rate": 1.5033333333333336e-05,
      "loss": 0.3316,
      "step": 7100
    },
    {
      "epoch": 280.3921568627451,
      "grad_norm": 0.47346118092536926,
      "learning_rate": 1.42e-05,
      "loss": 0.335,
      "step": 7150
    },
    {
      "epoch": 282.3529411764706,
      "grad_norm": 0.6653403043746948,
      "learning_rate": 1.3366666666666667e-05,
      "loss": 0.333,
      "step": 7200
    },
    {
      "epoch": 284.3137254901961,
      "grad_norm": 0.44320473074913025,
      "learning_rate": 1.2533333333333332e-05,
      "loss": 0.3322,
      "step": 7250
    },
    {
      "epoch": 286.27450980392155,
      "grad_norm": 0.40339425206184387,
      "learning_rate": 1.1700000000000001e-05,
      "loss": 0.3286,
      "step": 7300
    },
    {
      "epoch": 288.2352941176471,
      "grad_norm": 0.6550925374031067,
      "learning_rate": 1.0866666666666667e-05,
      "loss": 0.3351,
      "step": 7350
    },
    {
      "epoch": 290.19607843137254,
      "grad_norm": 0.5210171341896057,
      "learning_rate": 1.0033333333333333e-05,
      "loss": 0.3309,
      "step": 7400
    },
    {
      "epoch": 292.15686274509807,
      "grad_norm": 0.4037918448448181,
      "learning_rate": 9.2e-06,
      "loss": 0.3291,
      "step": 7450
    },
    {
      "epoch": 294.11764705882354,
      "grad_norm": 0.7901822328567505,
      "learning_rate": 8.366666666666667e-06,
      "loss": 0.3318,
      "step": 7500
    },
    {
      "epoch": 294.11764705882354,
      "eval_loss": 0.405687540769577,
      "eval_runtime": 6.5967,
      "eval_samples_per_second": 27.286,
      "eval_steps_per_second": 3.487,
      "step": 7500
    },
    {
      "epoch": 296.078431372549,
      "grad_norm": 0.6064783930778503,
      "learning_rate": 7.533333333333334e-06,
      "loss": 0.3314,
      "step": 7550
    },
    {
      "epoch": 298.03921568627453,
      "grad_norm": 0.5984176397323608,
      "learning_rate": 6.700000000000001e-06,
      "loss": 0.3312,
      "step": 7600
    },
    {
      "epoch": 300.0,
      "grad_norm": 0.6286676526069641,
      "learning_rate": 5.866666666666667e-06,
      "loss": 0.3286,
      "step": 7650
    },
    {
      "epoch": 301.96078431372547,
      "grad_norm": 0.45633211731910706,
      "learning_rate": 5.033333333333334e-06,
      "loss": 0.3285,
      "step": 7700
    },
    {
      "epoch": 303.921568627451,
      "grad_norm": 0.41933074593544006,
      "learning_rate": 4.2000000000000004e-06,
      "loss": 0.3272,
      "step": 7750
    },
    {
      "epoch": 305.88235294117646,
      "grad_norm": 0.4385930597782135,
      "learning_rate": 3.3666666666666665e-06,
      "loss": 0.3297,
      "step": 7800
    },
    {
      "epoch": 307.84313725490193,
      "grad_norm": 0.49151408672332764,
      "learning_rate": 2.5333333333333334e-06,
      "loss": 0.3315,
      "step": 7850
    },
    {
      "epoch": 309.80392156862746,
      "grad_norm": 0.5050336718559265,
      "learning_rate": 1.7000000000000002e-06,
      "loss": 0.331,
      "step": 7900
    },
    {
      "epoch": 311.7647058823529,
      "grad_norm": 0.3783471882343292,
      "learning_rate": 8.666666666666667e-07,
      "loss": 0.3269,
      "step": 7950
    },
    {
      "epoch": 313.72549019607845,
      "grad_norm": 0.4698627293109894,
      "learning_rate": 3.3333333333333334e-08,
      "loss": 0.3312,
      "step": 8000
    },
    {
      "epoch": 313.72549019607845,
      "eval_loss": 0.4050500988960266,
      "eval_runtime": 6.4964,
      "eval_samples_per_second": 27.708,
      "eval_steps_per_second": 3.54,
      "step": 8000
    }
  ],
  "logging_steps": 50,
  "max_steps": 8000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 320,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 8.70355911863679e+16,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}