{
  "best_metric": 0.39249518513679504,
  "best_model_checkpoint": "mikhail-panzo/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": 2.2592813968658447,
      "learning_rate": 2.4500000000000004e-07,
      "loss": 0.8071,
      "step": 50
    },
    {
      "epoch": 3.9215686274509802,
      "grad_norm": 1.564003348350525,
      "learning_rate": 4.95e-07,
      "loss": 0.7677,
      "step": 100
    },
    {
      "epoch": 5.882352941176471,
      "grad_norm": 1.3204560279846191,
      "learning_rate": 7.450000000000001e-07,
      "loss": 0.7469,
      "step": 150
    },
    {
      "epoch": 7.8431372549019605,
      "grad_norm": 1.3252646923065186,
      "learning_rate": 9.950000000000002e-07,
      "loss": 0.7234,
      "step": 200
    },
    {
      "epoch": 9.803921568627452,
      "grad_norm": 1.9049391746520996,
      "learning_rate": 1.2450000000000002e-06,
      "loss": 0.7167,
      "step": 250
    },
    {
      "epoch": 11.764705882352942,
      "grad_norm": 1.5868775844573975,
      "learning_rate": 1.495e-06,
      "loss": 0.6949,
      "step": 300
    },
    {
      "epoch": 13.72549019607843,
      "grad_norm": 2.022406578063965,
      "learning_rate": 1.745e-06,
      "loss": 0.6581,
      "step": 350
    },
    {
      "epoch": 15.686274509803921,
      "grad_norm": 1.601341724395752,
      "learning_rate": 1.9950000000000004e-06,
      "loss": 0.6181,
      "step": 400
    },
    {
      "epoch": 17.647058823529413,
      "grad_norm": 1.671892523765564,
      "learning_rate": 2.245e-06,
      "loss": 0.5798,
      "step": 450
    },
    {
      "epoch": 19.607843137254903,
      "grad_norm": 1.9054476022720337,
      "learning_rate": 2.4950000000000003e-06,
      "loss": 0.5525,
      "step": 500
    },
    {
      "epoch": 19.607843137254903,
      "eval_loss": 0.4685615003108978,
      "eval_runtime": 6.3848,
      "eval_samples_per_second": 28.192,
      "eval_steps_per_second": 3.602,
      "step": 500
    },
    {
      "epoch": 21.568627450980394,
      "grad_norm": 1.5475994348526,
      "learning_rate": 2.7450000000000004e-06,
      "loss": 0.5407,
      "step": 550
    },
    {
      "epoch": 23.529411764705884,
      "grad_norm": 1.1005382537841797,
      "learning_rate": 2.995e-06,
      "loss": 0.5076,
      "step": 600
    },
    {
      "epoch": 25.49019607843137,
      "grad_norm": 0.979748547077179,
      "learning_rate": 3.2450000000000003e-06,
      "loss": 0.5071,
      "step": 650
    },
    {
      "epoch": 27.45098039215686,
      "grad_norm": 1.0884859561920166,
      "learning_rate": 3.495e-06,
      "loss": 0.4934,
      "step": 700
    },
    {
      "epoch": 29.41176470588235,
      "grad_norm": 1.8463184833526611,
      "learning_rate": 3.745e-06,
      "loss": 0.4919,
      "step": 750
    },
    {
      "epoch": 31.372549019607842,
      "grad_norm": 1.2262991666793823,
      "learning_rate": 3.995000000000001e-06,
      "loss": 0.4981,
      "step": 800
    },
    {
      "epoch": 33.333333333333336,
      "grad_norm": 1.0878300666809082,
      "learning_rate": 4.245e-06,
      "loss": 0.4928,
      "step": 850
    },
    {
      "epoch": 35.294117647058826,
      "grad_norm": 1.155561089515686,
      "learning_rate": 4.495e-06,
      "loss": 0.4848,
      "step": 900
    },
    {
      "epoch": 37.254901960784316,
      "grad_norm": 1.4933425188064575,
      "learning_rate": 4.745e-06,
      "loss": 0.4754,
      "step": 950
    },
    {
      "epoch": 39.21568627450981,
      "grad_norm": 1.239863634109497,
      "learning_rate": 4.9950000000000005e-06,
      "loss": 0.4756,
      "step": 1000
    },
    {
      "epoch": 39.21568627450981,
      "eval_loss": 0.42764008045196533,
      "eval_runtime": 6.3802,
      "eval_samples_per_second": 28.212,
      "eval_steps_per_second": 3.605,
      "step": 1000
    },
    {
      "epoch": 41.1764705882353,
      "grad_norm": 1.9258134365081787,
      "learning_rate": 5.245e-06,
      "loss": 0.4653,
      "step": 1050
    },
    {
      "epoch": 43.13725490196079,
      "grad_norm": 0.9576085805892944,
      "learning_rate": 5.495000000000001e-06,
      "loss": 0.465,
      "step": 1100
    },
    {
      "epoch": 45.09803921568628,
      "grad_norm": 0.9215015769004822,
      "learning_rate": 5.745000000000001e-06,
      "loss": 0.4743,
      "step": 1150
    },
    {
      "epoch": 47.05882352941177,
      "grad_norm": 1.1917026042938232,
      "learning_rate": 5.995000000000001e-06,
      "loss": 0.4581,
      "step": 1200
    },
    {
      "epoch": 49.01960784313726,
      "grad_norm": 1.30499267578125,
      "learning_rate": 6.245000000000001e-06,
      "loss": 0.4626,
      "step": 1250
    },
    {
      "epoch": 50.98039215686274,
      "grad_norm": 1.4004439115524292,
      "learning_rate": 6.4950000000000005e-06,
      "loss": 0.4591,
      "step": 1300
    },
    {
      "epoch": 52.94117647058823,
      "grad_norm": 0.892646074295044,
      "learning_rate": 6.745000000000001e-06,
      "loss": 0.4545,
      "step": 1350
    },
    {
      "epoch": 54.90196078431372,
      "grad_norm": 1.0985666513442993,
      "learning_rate": 6.995000000000001e-06,
      "loss": 0.4501,
      "step": 1400
    },
    {
      "epoch": 56.86274509803921,
      "grad_norm": 1.271141529083252,
      "learning_rate": 7.245000000000001e-06,
      "loss": 0.4456,
      "step": 1450
    },
    {
      "epoch": 58.8235294117647,
      "grad_norm": 1.170074224472046,
      "learning_rate": 7.495000000000001e-06,
      "loss": 0.4543,
      "step": 1500
    },
    {
      "epoch": 58.8235294117647,
      "eval_loss": 0.4116212725639343,
      "eval_runtime": 6.2794,
      "eval_samples_per_second": 28.665,
      "eval_steps_per_second": 3.663,
      "step": 1500
    },
    {
      "epoch": 60.78431372549019,
      "grad_norm": 0.9767690896987915,
      "learning_rate": 7.745e-06,
      "loss": 0.4517,
      "step": 1550
    },
    {
      "epoch": 62.745098039215684,
      "grad_norm": 1.4193260669708252,
      "learning_rate": 7.995e-06,
      "loss": 0.4537,
      "step": 1600
    },
    {
      "epoch": 64.70588235294117,
      "grad_norm": 1.3294360637664795,
      "learning_rate": 8.245000000000002e-06,
      "loss": 0.4435,
      "step": 1650
    },
    {
      "epoch": 66.66666666666667,
      "grad_norm": 0.8386899828910828,
      "learning_rate": 8.495e-06,
      "loss": 0.4507,
      "step": 1700
    },
    {
      "epoch": 68.62745098039215,
      "grad_norm": 1.0917119979858398,
      "learning_rate": 8.745000000000002e-06,
      "loss": 0.4409,
      "step": 1750
    },
    {
      "epoch": 70.58823529411765,
      "grad_norm": 1.0725489854812622,
      "learning_rate": 8.995000000000001e-06,
      "loss": 0.4449,
      "step": 1800
    },
    {
      "epoch": 72.54901960784314,
      "grad_norm": 1.3506999015808105,
      "learning_rate": 9.245e-06,
      "loss": 0.4496,
      "step": 1850
    },
    {
      "epoch": 74.50980392156863,
      "grad_norm": 0.9701379537582397,
      "learning_rate": 9.495000000000001e-06,
      "loss": 0.4384,
      "step": 1900
    },
    {
      "epoch": 76.47058823529412,
      "grad_norm": 1.7079219818115234,
      "learning_rate": 9.745e-06,
      "loss": 0.4374,
      "step": 1950
    },
    {
      "epoch": 78.43137254901961,
      "grad_norm": 1.87998628616333,
      "learning_rate": 9.995000000000002e-06,
      "loss": 0.4346,
      "step": 2000
    },
    {
      "epoch": 78.43137254901961,
      "eval_loss": 0.4027920663356781,
      "eval_runtime": 6.3546,
      "eval_samples_per_second": 28.326,
      "eval_steps_per_second": 3.619,
      "step": 2000
    },
    {
      "epoch": 80.3921568627451,
      "grad_norm": 1.1510419845581055,
      "learning_rate": 9.918333333333335e-06,
      "loss": 0.4326,
      "step": 2050
    },
    {
      "epoch": 82.3529411764706,
      "grad_norm": 1.2605654001235962,
      "learning_rate": 9.835000000000002e-06,
      "loss": 0.4355,
      "step": 2100
    },
    {
      "epoch": 84.31372549019608,
      "grad_norm": 0.866606593132019,
      "learning_rate": 9.751666666666667e-06,
      "loss": 0.4286,
      "step": 2150
    },
    {
      "epoch": 86.27450980392157,
      "grad_norm": 2.0733227729797363,
      "learning_rate": 9.668333333333334e-06,
      "loss": 0.4365,
      "step": 2200
    },
    {
      "epoch": 88.23529411764706,
      "grad_norm": 0.9726402759552002,
      "learning_rate": 9.585e-06,
      "loss": 0.4367,
      "step": 2250
    },
    {
      "epoch": 90.19607843137256,
      "grad_norm": 1.0713222026824951,
      "learning_rate": 9.501666666666667e-06,
      "loss": 0.4288,
      "step": 2300
    },
    {
      "epoch": 92.15686274509804,
      "grad_norm": 1.5218483209609985,
      "learning_rate": 9.418333333333334e-06,
      "loss": 0.435,
      "step": 2350
    },
    {
      "epoch": 94.11764705882354,
      "grad_norm": 0.8391968011856079,
      "learning_rate": 9.335000000000001e-06,
      "loss": 0.431,
      "step": 2400
    },
    {
      "epoch": 96.07843137254902,
      "grad_norm": 1.3989890813827515,
      "learning_rate": 9.251666666666668e-06,
      "loss": 0.4251,
      "step": 2450
    },
    {
      "epoch": 98.03921568627452,
      "grad_norm": 0.9168123006820679,
      "learning_rate": 9.168333333333333e-06,
      "loss": 0.4292,
      "step": 2500
    },
    {
      "epoch": 98.03921568627452,
      "eval_loss": 0.3997121751308441,
      "eval_runtime": 6.4036,
      "eval_samples_per_second": 28.109,
      "eval_steps_per_second": 3.592,
      "step": 2500
    },
    {
      "epoch": 100.0,
      "grad_norm": 2.315229892730713,
      "learning_rate": 9.085e-06,
      "loss": 0.4295,
      "step": 2550
    },
    {
      "epoch": 101.96078431372548,
      "grad_norm": 1.0224589109420776,
      "learning_rate": 9.001666666666667e-06,
      "loss": 0.4212,
      "step": 2600
    },
    {
      "epoch": 103.92156862745098,
      "grad_norm": 1.389236330986023,
      "learning_rate": 8.918333333333334e-06,
      "loss": 0.4215,
      "step": 2650
    },
    {
      "epoch": 105.88235294117646,
      "grad_norm": 0.9601902961730957,
      "learning_rate": 8.836666666666668e-06,
      "loss": 0.4271,
      "step": 2700
    },
    {
      "epoch": 107.84313725490196,
      "grad_norm": 0.8070810437202454,
      "learning_rate": 8.753333333333333e-06,
      "loss": 0.4269,
      "step": 2750
    },
    {
      "epoch": 109.80392156862744,
      "grad_norm": 1.3389467000961304,
      "learning_rate": 8.67e-06,
      "loss": 0.4209,
      "step": 2800
    },
    {
      "epoch": 111.76470588235294,
      "grad_norm": 1.0083783864974976,
      "learning_rate": 8.586666666666667e-06,
      "loss": 0.4143,
      "step": 2850
    },
    {
      "epoch": 113.72549019607843,
      "grad_norm": 1.390769600868225,
      "learning_rate": 8.503333333333334e-06,
      "loss": 0.4208,
      "step": 2900
    },
    {
      "epoch": 115.68627450980392,
      "grad_norm": 1.0967605113983154,
      "learning_rate": 8.42e-06,
      "loss": 0.4127,
      "step": 2950
    },
    {
      "epoch": 117.6470588235294,
      "grad_norm": 0.9284217357635498,
      "learning_rate": 8.336666666666668e-06,
      "loss": 0.4166,
      "step": 3000
    },
    {
      "epoch": 117.6470588235294,
      "eval_loss": 0.395207017660141,
      "eval_runtime": 6.3246,
      "eval_samples_per_second": 28.46,
      "eval_steps_per_second": 3.637,
      "step": 3000
    },
    {
      "epoch": 119.6078431372549,
      "grad_norm": 1.1663891077041626,
      "learning_rate": 8.253333333333334e-06,
      "loss": 0.4155,
      "step": 3050
    },
    {
      "epoch": 121.56862745098039,
      "grad_norm": 5.156078338623047,
      "learning_rate": 8.17e-06,
      "loss": 0.4156,
      "step": 3100
    },
    {
      "epoch": 123.52941176470588,
      "grad_norm": 0.9861869215965271,
      "learning_rate": 8.086666666666667e-06,
      "loss": 0.4163,
      "step": 3150
    },
    {
      "epoch": 125.49019607843137,
      "grad_norm": 0.8449072241783142,
      "learning_rate": 8.003333333333334e-06,
      "loss": 0.4176,
      "step": 3200
    },
    {
      "epoch": 127.45098039215686,
      "grad_norm": 1.300947904586792,
      "learning_rate": 7.92e-06,
      "loss": 0.4245,
      "step": 3250
    },
    {
      "epoch": 129.41176470588235,
      "grad_norm": 0.7830358147621155,
      "learning_rate": 7.836666666666667e-06,
      "loss": 0.4124,
      "step": 3300
    },
    {
      "epoch": 131.37254901960785,
      "grad_norm": 1.3250160217285156,
      "learning_rate": 7.753333333333334e-06,
      "loss": 0.4119,
      "step": 3350
    },
    {
      "epoch": 133.33333333333334,
      "grad_norm": 0.8610659241676331,
      "learning_rate": 7.670000000000001e-06,
      "loss": 0.4154,
      "step": 3400
    },
    {
      "epoch": 135.2941176470588,
      "grad_norm": 1.3154343366622925,
      "learning_rate": 7.586666666666668e-06,
      "loss": 0.4141,
      "step": 3450
    },
    {
      "epoch": 137.2549019607843,
      "grad_norm": 0.9761775732040405,
      "learning_rate": 7.503333333333333e-06,
      "loss": 0.4122,
      "step": 3500
    },
    {
      "epoch": 137.2549019607843,
      "eval_loss": 0.3956769108772278,
      "eval_runtime": 6.4498,
      "eval_samples_per_second": 27.908,
      "eval_steps_per_second": 3.566,
      "step": 3500
    },
    {
      "epoch": 139.2156862745098,
      "grad_norm": 1.142308235168457,
      "learning_rate": 7.420000000000001e-06,
      "loss": 0.4107,
      "step": 3550
    },
    {
      "epoch": 141.1764705882353,
      "grad_norm": 0.863447368144989,
      "learning_rate": 7.336666666666668e-06,
      "loss": 0.4093,
      "step": 3600
    },
    {
      "epoch": 143.13725490196077,
      "grad_norm": 0.9665449261665344,
      "learning_rate": 7.253333333333335e-06,
      "loss": 0.4114,
      "step": 3650
    },
    {
      "epoch": 145.09803921568627,
      "grad_norm": 1.046099305152893,
      "learning_rate": 7.17e-06,
      "loss": 0.4131,
      "step": 3700
    },
    {
      "epoch": 147.05882352941177,
      "grad_norm": 0.9744811654090881,
      "learning_rate": 7.086666666666667e-06,
      "loss": 0.4068,
      "step": 3750
    },
    {
      "epoch": 149.01960784313727,
      "grad_norm": 1.3594048023223877,
      "learning_rate": 7.003333333333335e-06,
      "loss": 0.4086,
      "step": 3800
    },
    {
      "epoch": 150.98039215686273,
      "grad_norm": 0.889851450920105,
      "learning_rate": 6.92e-06,
      "loss": 0.4098,
      "step": 3850
    },
    {
      "epoch": 152.94117647058823,
      "grad_norm": 0.940967321395874,
      "learning_rate": 6.836666666666667e-06,
      "loss": 0.4071,
      "step": 3900
    },
    {
      "epoch": 154.90196078431373,
      "grad_norm": 1.5695271492004395,
      "learning_rate": 6.753333333333334e-06,
      "loss": 0.4097,
      "step": 3950
    },
    {
      "epoch": 156.86274509803923,
      "grad_norm": 1.224783182144165,
      "learning_rate": 6.6700000000000005e-06,
      "loss": 0.4063,
      "step": 4000
    },
    {
      "epoch": 156.86274509803923,
      "eval_loss": 0.3939879834651947,
      "eval_runtime": 6.6684,
      "eval_samples_per_second": 26.993,
      "eval_steps_per_second": 3.449,
      "step": 4000
    },
    {
      "epoch": 158.8235294117647,
      "grad_norm": 0.8880926370620728,
      "learning_rate": 6.5866666666666666e-06,
      "loss": 0.4149,
      "step": 4050
    },
    {
      "epoch": 160.7843137254902,
      "grad_norm": 0.8387724757194519,
      "learning_rate": 6.5033333333333335e-06,
      "loss": 0.4066,
      "step": 4100
    },
    {
      "epoch": 162.7450980392157,
      "grad_norm": 0.9014448523521423,
      "learning_rate": 6.42e-06,
      "loss": 0.4102,
      "step": 4150
    },
    {
      "epoch": 164.7058823529412,
      "grad_norm": 0.7693892121315002,
      "learning_rate": 6.336666666666667e-06,
      "loss": 0.4066,
      "step": 4200
    },
    {
      "epoch": 166.66666666666666,
      "grad_norm": 0.9693483114242554,
      "learning_rate": 6.253333333333333e-06,
      "loss": 0.4084,
      "step": 4250
    },
    {
      "epoch": 168.62745098039215,
      "grad_norm": 0.8914594054222107,
      "learning_rate": 6.17e-06,
      "loss": 0.4109,
      "step": 4300
    },
    {
      "epoch": 170.58823529411765,
      "grad_norm": 11.76923942565918,
      "learning_rate": 6.086666666666667e-06,
      "loss": 0.4109,
      "step": 4350
    },
    {
      "epoch": 172.54901960784315,
      "grad_norm": 0.8837347626686096,
      "learning_rate": 6.003333333333334e-06,
      "loss": 0.4051,
      "step": 4400
    },
    {
      "epoch": 174.50980392156862,
      "grad_norm": 0.8585467338562012,
      "learning_rate": 5.92e-06,
      "loss": 0.4017,
      "step": 4450
    },
    {
      "epoch": 176.47058823529412,
      "grad_norm": 0.967064380645752,
      "learning_rate": 5.836666666666667e-06,
      "loss": 0.4028,
      "step": 4500
    },
    {
      "epoch": 176.47058823529412,
      "eval_loss": 0.395079106092453,
      "eval_runtime": 6.3984,
      "eval_samples_per_second": 28.132,
      "eval_steps_per_second": 3.595,
      "step": 4500
    },
    {
      "epoch": 178.4313725490196,
      "grad_norm": 1.2672348022460938,
      "learning_rate": 5.753333333333334e-06,
      "loss": 0.4068,
      "step": 4550
    },
    {
      "epoch": 180.3921568627451,
      "grad_norm": 0.8008630275726318,
      "learning_rate": 5.67e-06,
      "loss": 0.4112,
      "step": 4600
    },
    {
      "epoch": 182.35294117647058,
      "grad_norm": 1.0430930852890015,
      "learning_rate": 5.586666666666667e-06,
      "loss": 0.4145,
      "step": 4650
    },
    {
      "epoch": 184.31372549019608,
      "grad_norm": 0.767160952091217,
      "learning_rate": 5.503333333333334e-06,
      "loss": 0.3992,
      "step": 4700
    },
    {
      "epoch": 186.27450980392157,
      "grad_norm": 1.0731803178787231,
      "learning_rate": 5.420000000000001e-06,
      "loss": 0.4025,
      "step": 4750
    },
    {
      "epoch": 188.23529411764707,
      "grad_norm": 0.7984374761581421,
      "learning_rate": 5.336666666666667e-06,
      "loss": 0.4002,
      "step": 4800
    },
    {
      "epoch": 190.19607843137254,
      "grad_norm": 0.8159019947052002,
      "learning_rate": 5.2533333333333336e-06,
      "loss": 0.4023,
      "step": 4850
    },
    {
      "epoch": 192.15686274509804,
      "grad_norm": 1.1482937335968018,
      "learning_rate": 5.1700000000000005e-06,
      "loss": 0.4021,
      "step": 4900
    },
    {
      "epoch": 194.11764705882354,
      "grad_norm": 0.9159001708030701,
      "learning_rate": 5.086666666666667e-06,
      "loss": 0.4058,
      "step": 4950
    },
    {
      "epoch": 196.07843137254903,
      "grad_norm": 0.8861828446388245,
      "learning_rate": 5.0033333333333334e-06,
      "loss": 0.3982,
      "step": 5000
    },
    {
      "epoch": 196.07843137254903,
      "eval_loss": 0.3931240439414978,
      "eval_runtime": 6.4434,
      "eval_samples_per_second": 27.936,
      "eval_steps_per_second": 3.57,
      "step": 5000
    },
    {
      "epoch": 198.0392156862745,
      "grad_norm": 0.9328787922859192,
      "learning_rate": 4.92e-06,
      "loss": 0.405,
      "step": 5050
    },
    {
      "epoch": 200.0,
      "grad_norm": 1.2074756622314453,
      "learning_rate": 4.836666666666667e-06,
      "loss": 0.3996,
      "step": 5100
    },
    {
      "epoch": 201.9607843137255,
      "grad_norm": 0.7930981516838074,
      "learning_rate": 4.753333333333333e-06,
      "loss": 0.3994,
      "step": 5150
    },
    {
      "epoch": 203.92156862745097,
      "grad_norm": 0.8685852885246277,
      "learning_rate": 4.670000000000001e-06,
      "loss": 0.3983,
      "step": 5200
    },
    {
      "epoch": 205.88235294117646,
      "grad_norm": 0.8023931980133057,
      "learning_rate": 4.586666666666667e-06,
      "loss": 0.4041,
      "step": 5250
    },
    {
      "epoch": 207.84313725490196,
      "grad_norm": 0.7825555801391602,
      "learning_rate": 4.503333333333333e-06,
      "loss": 0.4012,
      "step": 5300
    },
    {
      "epoch": 209.80392156862746,
      "grad_norm": 1.1338242292404175,
      "learning_rate": 4.42e-06,
      "loss": 0.3996,
      "step": 5350
    },
    {
      "epoch": 211.76470588235293,
      "grad_norm": 1.3594533205032349,
      "learning_rate": 4.336666666666667e-06,
      "loss": 0.4022,
      "step": 5400
    },
    {
      "epoch": 213.72549019607843,
      "grad_norm": 1.0148471593856812,
      "learning_rate": 4.253333333333334e-06,
      "loss": 0.4003,
      "step": 5450
    },
    {
      "epoch": 215.68627450980392,
      "grad_norm": 0.932323157787323,
      "learning_rate": 4.17e-06,
      "loss": 0.4055,
      "step": 5500
    },
    {
      "epoch": 215.68627450980392,
      "eval_loss": 0.3945559561252594,
      "eval_runtime": 6.6419,
      "eval_samples_per_second": 27.101,
      "eval_steps_per_second": 3.463,
      "step": 5500
    },
    {
      "epoch": 217.64705882352942,
      "grad_norm": 0.7886359095573425,
      "learning_rate": 4.086666666666667e-06,
      "loss": 0.3938,
      "step": 5550
    },
    {
      "epoch": 219.6078431372549,
      "grad_norm": 1.1809077262878418,
      "learning_rate": 4.003333333333334e-06,
      "loss": 0.4033,
      "step": 5600
    },
    {
      "epoch": 221.5686274509804,
      "grad_norm": 0.8167886137962341,
      "learning_rate": 3.920000000000001e-06,
      "loss": 0.4002,
      "step": 5650
    },
    {
      "epoch": 223.52941176470588,
      "grad_norm": 1.1182901859283447,
      "learning_rate": 3.836666666666667e-06,
      "loss": 0.3982,
      "step": 5700
    },
    {
      "epoch": 225.49019607843138,
      "grad_norm": 1.0442403554916382,
      "learning_rate": 3.753333333333334e-06,
      "loss": 0.3973,
      "step": 5750
    },
    {
      "epoch": 227.45098039215685,
      "grad_norm": 1.0428043603897095,
      "learning_rate": 3.6700000000000004e-06,
      "loss": 0.4001,
      "step": 5800
    },
    {
      "epoch": 229.41176470588235,
      "grad_norm": 0.9441725611686707,
      "learning_rate": 3.5866666666666673e-06,
      "loss": 0.3976,
      "step": 5850
    },
    {
      "epoch": 231.37254901960785,
      "grad_norm": 0.8371317386627197,
      "learning_rate": 3.5033333333333334e-06,
      "loss": 0.3973,
      "step": 5900
    },
    {
      "epoch": 233.33333333333334,
      "grad_norm": 0.7603981494903564,
      "learning_rate": 3.4200000000000007e-06,
      "loss": 0.3943,
      "step": 5950
    },
    {
      "epoch": 235.2941176470588,
      "grad_norm": 1.1027510166168213,
      "learning_rate": 3.3366666666666668e-06,
      "loss": 0.4019,
      "step": 6000
    },
    {
      "epoch": 235.2941176470588,
      "eval_loss": 0.39249518513679504,
      "eval_runtime": 6.4221,
      "eval_samples_per_second": 28.028,
      "eval_steps_per_second": 3.581,
      "step": 6000
    },
    {
      "epoch": 237.2549019607843,
      "grad_norm": 1.0101209878921509,
      "learning_rate": 3.2533333333333332e-06,
      "loss": 0.396,
      "step": 6050
    },
    {
      "epoch": 239.2156862745098,
      "grad_norm": 1.2180397510528564,
      "learning_rate": 3.17e-06,
      "loss": 0.3986,
      "step": 6100
    },
    {
      "epoch": 241.1764705882353,
      "grad_norm": 0.9254348874092102,
      "learning_rate": 3.0866666666666666e-06,
      "loss": 0.397,
      "step": 6150
    },
    {
      "epoch": 243.13725490196077,
      "grad_norm": 1.0118534564971924,
      "learning_rate": 3.0033333333333335e-06,
      "loss": 0.3974,
      "step": 6200
    },
    {
      "epoch": 245.09803921568627,
      "grad_norm": 0.8282018899917603,
      "learning_rate": 2.92e-06,
      "loss": 0.3909,
      "step": 6250
    },
    {
      "epoch": 247.05882352941177,
      "grad_norm": 0.8930662870407104,
      "learning_rate": 2.836666666666667e-06,
      "loss": 0.3986,
      "step": 6300
    },
    {
      "epoch": 249.01960784313727,
      "grad_norm": 1.106566071510315,
      "learning_rate": 2.7533333333333334e-06,
      "loss": 0.3984,
      "step": 6350
    },
    {
      "epoch": 250.98039215686273,
      "grad_norm": 0.9557139873504639,
      "learning_rate": 2.6700000000000003e-06,
      "loss": 0.3975,
      "step": 6400
    },
    {
      "epoch": 252.94117647058823,
      "grad_norm": 1.1916723251342773,
      "learning_rate": 2.5866666666666667e-06,
      "loss": 0.402,
      "step": 6450
    },
    {
      "epoch": 254.90196078431373,
      "grad_norm": 0.8120290637016296,
      "learning_rate": 2.5033333333333336e-06,
      "loss": 0.4,
      "step": 6500
    },
    {
      "epoch": 254.90196078431373,
      "eval_loss": 0.39403682947158813,
      "eval_runtime": 6.3124,
      "eval_samples_per_second": 28.515,
      "eval_steps_per_second": 3.644,
      "step": 6500
    },
    {
      "epoch": 256.8627450980392,
      "grad_norm": 0.9240416884422302,
      "learning_rate": 2.42e-06,
      "loss": 0.391,
      "step": 6550
    },
    {
      "epoch": 258.8235294117647,
      "grad_norm": 1.0390872955322266,
      "learning_rate": 2.3366666666666666e-06,
      "loss": 0.3985,
      "step": 6600
    },
    {
      "epoch": 260.7843137254902,
      "grad_norm": 0.7079731822013855,
      "learning_rate": 2.2550000000000004e-06,
      "loss": 0.3947,
      "step": 6650
    },
    {
      "epoch": 262.7450980392157,
      "grad_norm": 1.0071649551391602,
      "learning_rate": 2.171666666666667e-06,
      "loss": 0.4006,
      "step": 6700
    },
    {
      "epoch": 264.70588235294116,
      "grad_norm": 0.9264857769012451,
      "learning_rate": 2.088333333333334e-06,
      "loss": 0.3983,
      "step": 6750
    },
    {
      "epoch": 266.6666666666667,
      "grad_norm": 0.7591241002082825,
      "learning_rate": 2.0050000000000003e-06,
      "loss": 0.3995,
      "step": 6800
    },
    {
      "epoch": 268.62745098039215,
      "grad_norm": 0.9408916234970093,
      "learning_rate": 1.9216666666666668e-06,
      "loss": 0.4013,
      "step": 6850
    },
    {
      "epoch": 270.5882352941176,
      "grad_norm": 0.8653966188430786,
      "learning_rate": 1.8383333333333334e-06,
      "loss": 0.3948,
      "step": 6900
    },
    {
      "epoch": 272.54901960784315,
      "grad_norm": 1.010793924331665,
      "learning_rate": 1.7550000000000001e-06,
      "loss": 0.3935,
      "step": 6950
    },
    {
      "epoch": 274.5098039215686,
      "grad_norm": 1.1880955696105957,
      "learning_rate": 1.6716666666666666e-06,
      "loss": 0.4046,
      "step": 7000
    },
    {
      "epoch": 274.5098039215686,
      "eval_loss": 0.3952997326850891,
      "eval_runtime": 6.3393,
      "eval_samples_per_second": 28.394,
      "eval_steps_per_second": 3.628,
      "step": 7000
    },
    {
      "epoch": 276.47058823529414,
      "grad_norm": 0.9707151651382446,
      "learning_rate": 1.5883333333333333e-06,
      "loss": 0.4001,
      "step": 7050
    },
    {
      "epoch": 278.4313725490196,
      "grad_norm": 0.7843708395957947,
      "learning_rate": 1.505e-06,
      "loss": 0.3942,
      "step": 7100
    },
    {
      "epoch": 280.3921568627451,
      "grad_norm": 0.900497555732727,
      "learning_rate": 1.4216666666666667e-06,
      "loss": 0.3981,
      "step": 7150
    },
    {
      "epoch": 282.3529411764706,
      "grad_norm": 0.7594972848892212,
      "learning_rate": 1.3383333333333334e-06,
      "loss": 0.3961,
      "step": 7200
    },
    {
      "epoch": 284.3137254901961,
      "grad_norm": 0.7281601428985596,
      "learning_rate": 1.255e-06,
      "loss": 0.394,
      "step": 7250
    },
    {
      "epoch": 286.27450980392155,
      "grad_norm": 0.7863117456436157,
      "learning_rate": 1.1716666666666667e-06,
      "loss": 0.3899,
      "step": 7300
    },
    {
      "epoch": 288.2352941176471,
      "grad_norm": 0.7224944233894348,
      "learning_rate": 1.0883333333333334e-06,
      "loss": 0.3993,
      "step": 7350
    },
    {
      "epoch": 290.19607843137254,
      "grad_norm": 0.7235053181648254,
      "learning_rate": 1.0050000000000001e-06,
      "loss": 0.3928,
      "step": 7400
    },
    {
      "epoch": 292.15686274509807,
      "grad_norm": 0.7260425686836243,
      "learning_rate": 9.216666666666667e-07,
      "loss": 0.3904,
      "step": 7450
    },
    {
      "epoch": 294.11764705882354,
      "grad_norm": 0.7799311876296997,
      "learning_rate": 8.383333333333334e-07,
      "loss": 0.3955,
      "step": 7500
    },
    {
      "epoch": 294.11764705882354,
      "eval_loss": 0.3945452570915222,
      "eval_runtime": 6.6937,
      "eval_samples_per_second": 26.891,
      "eval_steps_per_second": 3.436,
      "step": 7500
    },
    {
      "epoch": 296.078431372549,
      "grad_norm": 0.7973820567131042,
      "learning_rate": 7.550000000000001e-07,
      "loss": 0.3954,
      "step": 7550
    },
    {
      "epoch": 298.03921568627453,
      "grad_norm": 0.7256256341934204,
      "learning_rate": 6.716666666666668e-07,
      "loss": 0.3934,
      "step": 7600
    },
    {
      "epoch": 300.0,
      "grad_norm": 1.1149753332138062,
      "learning_rate": 5.883333333333333e-07,
      "loss": 0.3913,
      "step": 7650
    },
    {
      "epoch": 301.96078431372547,
      "grad_norm": 0.6526250243186951,
      "learning_rate": 5.05e-07,
      "loss": 0.391,
      "step": 7700
    },
    {
      "epoch": 303.921568627451,
      "grad_norm": 0.8052578568458557,
      "learning_rate": 4.2166666666666667e-07,
      "loss": 0.3889,
      "step": 7750
    },
    {
      "epoch": 305.88235294117646,
      "grad_norm": 0.7933726906776428,
      "learning_rate": 3.3833333333333336e-07,
      "loss": 0.3918,
      "step": 7800
    },
    {
      "epoch": 307.84313725490193,
      "grad_norm": 0.7680366635322571,
      "learning_rate": 2.55e-07,
      "loss": 0.3948,
      "step": 7850
    },
    {
      "epoch": 309.80392156862746,
      "grad_norm": 0.8762979507446289,
      "learning_rate": 1.7166666666666668e-07,
      "loss": 0.3937,
      "step": 7900
    },
    {
      "epoch": 311.7647058823529,
      "grad_norm": 0.7918867468833923,
      "learning_rate": 8.833333333333334e-08,
      "loss": 0.3889,
      "step": 7950
    },
    {
      "epoch": 313.72549019607845,
      "grad_norm": 0.7141278982162476,
      "learning_rate": 5e-09,
      "loss": 0.3944,
      "step": 8000
    },
    {
      "epoch": 313.72549019607845,
      "eval_loss": 0.3929772675037384,
      "eval_runtime": 6.3752,
      "eval_samples_per_second": 28.234,
      "eval_steps_per_second": 3.608,
      "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
}