{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 10.0,
  "eval_steps": 500,
  "global_step": 510,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0196078431372549,
      "grad_norm": 2.359375,
      "learning_rate": 3.92156862745098e-06,
      "loss": 2.7922,
      "step": 1
    },
    {
      "epoch": 0.09803921568627451,
      "grad_norm": 1.8984375,
      "learning_rate": 1.9607843137254903e-05,
      "loss": 2.8181,
      "step": 5
    },
    {
      "epoch": 0.19607843137254902,
      "grad_norm": 1.5,
      "learning_rate": 3.9215686274509805e-05,
      "loss": 2.7665,
      "step": 10
    },
    {
      "epoch": 0.29411764705882354,
      "grad_norm": 1.953125,
      "learning_rate": 5.882352941176471e-05,
      "loss": 2.6609,
      "step": 15
    },
    {
      "epoch": 0.39215686274509803,
      "grad_norm": 1.6796875,
      "learning_rate": 7.843137254901961e-05,
      "loss": 2.4693,
      "step": 20
    },
    {
      "epoch": 0.49019607843137253,
      "grad_norm": 1.8125,
      "learning_rate": 9.80392156862745e-05,
      "loss": 2.2357,
      "step": 25
    },
    {
      "epoch": 0.5882352941176471,
      "grad_norm": 2.3125,
      "learning_rate": 0.00011764705882352942,
      "loss": 2.0763,
      "step": 30
    },
    {
      "epoch": 0.6862745098039216,
      "grad_norm": 0.98046875,
      "learning_rate": 0.0001372549019607843,
      "loss": 1.8956,
      "step": 35
    },
    {
      "epoch": 0.7843137254901961,
      "grad_norm": 0.8515625,
      "learning_rate": 0.00015686274509803922,
      "loss": 1.7407,
      "step": 40
    },
    {
      "epoch": 0.8823529411764706,
      "grad_norm": 0.6328125,
      "learning_rate": 0.00017647058823529413,
      "loss": 1.5856,
      "step": 45
    },
    {
      "epoch": 0.9803921568627451,
      "grad_norm": 1.6171875,
      "learning_rate": 0.000196078431372549,
      "loss": 1.4889,
      "step": 50
    },
    {
      "epoch": 1.0,
      "eval_loss": 2.5875163078308105,
      "eval_runtime": 0.5475,
      "eval_samples_per_second": 18.266,
      "eval_steps_per_second": 1.827,
      "step": 51
    },
    {
      "epoch": 1.0784313725490196,
      "grad_norm": 0.52734375,
      "learning_rate": 0.00019996252538028507,
      "loss": 1.3908,
      "step": 55
    },
    {
      "epoch": 1.1764705882352942,
      "grad_norm": 0.412109375,
      "learning_rate": 0.00019981033287370443,
      "loss": 1.336,
      "step": 60
    },
    {
      "epoch": 1.2745098039215685,
      "grad_norm": 0.5078125,
      "learning_rate": 0.00019954125840299165,
      "loss": 1.3043,
      "step": 65
    },
    {
      "epoch": 1.3725490196078431,
      "grad_norm": 0.69140625,
      "learning_rate": 0.00019915561706530883,
      "loss": 1.2637,
      "step": 70
    },
    {
      "epoch": 1.4705882352941178,
      "grad_norm": 0.3203125,
      "learning_rate": 0.00019865386046236596,
      "loss": 1.2476,
      "step": 75
    },
    {
      "epoch": 1.5686274509803921,
      "grad_norm": 0.56640625,
      "learning_rate": 0.0001980365761715769,
      "loss": 1.2254,
      "step": 80
    },
    {
      "epoch": 1.6666666666666665,
      "grad_norm": 0.26953125,
      "learning_rate": 0.00019730448705798239,
      "loss": 1.2052,
      "step": 85
    },
    {
      "epoch": 1.7647058823529411,
      "grad_norm": 0.345703125,
      "learning_rate": 0.00019645845042774553,
      "loss": 1.1999,
      "step": 90
    },
    {
      "epoch": 1.8627450980392157,
      "grad_norm": 0.291015625,
      "learning_rate": 0.00019549945702421144,
      "loss": 1.1781,
      "step": 95
    },
    {
      "epoch": 1.9607843137254903,
      "grad_norm": 0.427734375,
      "learning_rate": 0.00019442862986770646,
      "loss": 1.1695,
      "step": 100
    },
    {
      "epoch": 2.0,
      "eval_loss": 2.5043954849243164,
      "eval_runtime": 0.539,
      "eval_samples_per_second": 18.554,
      "eval_steps_per_second": 1.855,
      "step": 102
    },
    {
      "epoch": 2.0588235294117645,
      "grad_norm": 0.52734375,
      "learning_rate": 0.00019324722294043558,
      "loss": 1.1426,
      "step": 105
    },
    {
      "epoch": 2.156862745098039,
      "grad_norm": 0.37109375,
      "learning_rate": 0.00019195661971801827,
      "loss": 1.1312,
      "step": 110
    },
    {
      "epoch": 2.2549019607843137,
      "grad_norm": 0.52734375,
      "learning_rate": 0.00019055833154938207,
      "loss": 1.1323,
      "step": 115
    },
    {
      "epoch": 2.3529411764705883,
      "grad_norm": 0.5078125,
      "learning_rate": 0.00018905399588691163,
      "loss": 1.1302,
      "step": 120
    },
    {
      "epoch": 2.450980392156863,
      "grad_norm": 0.390625,
      "learning_rate": 0.00018744537436892516,
      "loss": 1.1181,
      "step": 125
    },
    {
      "epoch": 2.549019607843137,
      "grad_norm": 0.365234375,
      "learning_rate": 0.00018573435075672424,
      "loss": 1.1019,
      "step": 130
    },
    {
      "epoch": 2.6470588235294117,
      "grad_norm": 0.375,
      "learning_rate": 0.00018392292872863267,
      "loss": 1.119,
      "step": 135
    },
    {
      "epoch": 2.7450980392156863,
      "grad_norm": 0.458984375,
      "learning_rate": 0.00018201322953360758,
      "loss": 1.095,
      "step": 140
    },
    {
      "epoch": 2.843137254901961,
      "grad_norm": 0.421875,
      "learning_rate": 0.00018000748950717038,
      "loss": 1.0959,
      "step": 145
    },
    {
      "epoch": 2.9411764705882355,
      "grad_norm": 0.45703125,
      "learning_rate": 0.00017790805745256704,
      "loss": 1.1003,
      "step": 150
    },
    {
      "epoch": 3.0,
      "eval_loss": 2.5000319480895996,
      "eval_runtime": 0.5409,
      "eval_samples_per_second": 18.489,
      "eval_steps_per_second": 1.849,
      "step": 153
    },
    {
      "epoch": 3.0392156862745097,
      "grad_norm": 0.609375,
      "learning_rate": 0.00017571739189022365,
      "loss": 1.0867,
      "step": 155
    },
    {
      "epoch": 3.1372549019607843,
      "grad_norm": 0.6171875,
      "learning_rate": 0.00017343805817871886,
      "loss": 1.0606,
      "step": 160
    },
    {
      "epoch": 3.235294117647059,
      "grad_norm": 0.58984375,
      "learning_rate": 0.00017107272551064473,
      "loss": 1.0521,
      "step": 165
    },
    {
      "epoch": 3.3333333333333335,
      "grad_norm": 0.51171875,
      "learning_rate": 0.0001686241637868734,
      "loss": 1.0691,
      "step": 170
    },
    {
      "epoch": 3.431372549019608,
      "grad_norm": 0.65625,
      "learning_rate": 0.00016609524037289019,
      "loss": 1.0573,
      "step": 175
    },
    {
      "epoch": 3.5294117647058822,
      "grad_norm": 0.470703125,
      "learning_rate": 0.0001634889167409923,
      "loss": 1.063,
      "step": 180
    },
    {
      "epoch": 3.627450980392157,
      "grad_norm": 0.44921875,
      "learning_rate": 0.00016080824500228367,
      "loss": 1.0503,
      "step": 185
    },
    {
      "epoch": 3.7254901960784315,
      "grad_norm": 0.5234375,
      "learning_rate": 0.00015805636433252891,
      "loss": 1.051,
      "step": 190
    },
    {
      "epoch": 3.8235294117647056,
      "grad_norm": 0.431640625,
      "learning_rate": 0.0001552364972960506,
      "loss": 1.0651,
      "step": 195
    },
    {
      "epoch": 3.9215686274509802,
      "grad_norm": 0.5,
      "learning_rate": 0.00015235194607197508,
      "loss": 1.0537,
      "step": 200
    },
    {
      "epoch": 4.0,
      "eval_loss": 2.503964900970459,
      "eval_runtime": 0.5385,
      "eval_samples_per_second": 18.568,
      "eval_steps_per_second": 1.857,
      "step": 204
    },
    {
      "epoch": 4.019607843137255,
      "grad_norm": 0.41796875,
      "learning_rate": 0.0001494060885872464,
      "loss": 1.0477,
      "step": 205
    },
    {
      "epoch": 4.117647058823529,
      "grad_norm": 0.486328125,
      "learning_rate": 0.00014640237456093634,
      "loss": 1.0205,
      "step": 210
    },
    {
      "epoch": 4.215686274509804,
      "grad_norm": 0.43359375,
      "learning_rate": 0.00014334432146448272,
      "loss": 1.0166,
      "step": 215
    },
    {
      "epoch": 4.313725490196078,
      "grad_norm": 0.4609375,
      "learning_rate": 0.00014023551040258725,
      "loss": 1.0248,
      "step": 220
    },
    {
      "epoch": 4.411764705882353,
      "grad_norm": 0.56640625,
      "learning_rate": 0.00013707958191959608,
      "loss": 1.0165,
      "step": 225
    },
    {
      "epoch": 4.509803921568627,
      "grad_norm": 0.59765625,
      "learning_rate": 0.00013388023173627414,
      "loss": 1.0233,
      "step": 230
    },
    {
      "epoch": 4.607843137254902,
      "grad_norm": 0.5625,
      "learning_rate": 0.00013064120642196548,
      "loss": 1.0261,
      "step": 235
    },
    {
      "epoch": 4.705882352941177,
      "grad_norm": 0.53515625,
      "learning_rate": 0.0001273662990072083,
      "loss": 1.0329,
      "step": 240
    },
    {
      "epoch": 4.803921568627451,
      "grad_norm": 0.484375,
      "learning_rate": 0.00012405934454194146,
      "loss": 1.0197,
      "step": 245
    },
    {
      "epoch": 4.901960784313726,
      "grad_norm": 0.44921875,
      "learning_rate": 0.00012072421560450497,
      "loss": 1.0312,
      "step": 250
    },
    {
      "epoch": 5.0,
      "grad_norm": 0.53515625,
      "learning_rate": 0.00011736481776669306,
      "loss": 1.0131,
      "step": 255
    },
    {
      "epoch": 5.0,
      "eval_loss": 2.513137102127075,
      "eval_runtime": 0.537,
      "eval_samples_per_second": 18.622,
      "eval_steps_per_second": 1.862,
      "step": 255
    },
    {
      "epoch": 5.098039215686274,
      "grad_norm": 0.474609375,
      "learning_rate": 0.00011398508502017047,
      "loss": 0.9953,
      "step": 260
    },
    {
      "epoch": 5.196078431372549,
      "grad_norm": 0.482421875,
      "learning_rate": 0.00011058897516960816,
      "loss": 1.0078,
      "step": 265
    },
    {
      "epoch": 5.294117647058823,
      "grad_norm": 0.51171875,
      "learning_rate": 0.00010718046519793276,
      "loss": 1.0003,
      "step": 270
    },
    {
      "epoch": 5.392156862745098,
      "grad_norm": 0.474609375,
      "learning_rate": 0.00010376354660911771,
      "loss": 1.0081,
      "step": 275
    },
    {
      "epoch": 5.490196078431373,
      "grad_norm": 0.4765625,
      "learning_rate": 0.00010034222075396953,
      "loss": 0.9872,
      "step": 280
    },
    {
      "epoch": 5.588235294117647,
      "grad_norm": 0.466796875,
      "learning_rate": 9.692049414438299e-05,
      "loss": 0.988,
      "step": 285
    },
    {
      "epoch": 5.686274509803922,
      "grad_norm": 0.451171875,
      "learning_rate": 9.350237376155269e-05,
      "loss": 0.9944,
      "step": 290
    },
    {
      "epoch": 5.784313725490196,
      "grad_norm": 0.4921875,
      "learning_rate": 9.00918623636349e-05,
      "loss": 0.985,
      "step": 295
    },
    {
      "epoch": 5.882352941176471,
      "grad_norm": 0.4921875,
      "learning_rate": 8.669295379835467e-05,
      "loss": 1.0003,
      "step": 300
    },
    {
      "epoch": 5.980392156862745,
      "grad_norm": 0.46875,
      "learning_rate": 8.330962832604747e-05,
      "loss": 0.995,
      "step": 305
    },
    {
      "epoch": 6.0,
      "eval_loss": 2.5207793712615967,
      "eval_runtime": 0.5386,
      "eval_samples_per_second": 18.566,
      "eval_steps_per_second": 1.857,
      "step": 306
    },
    {
      "epoch": 6.078431372549019,
      "grad_norm": 0.56640625,
      "learning_rate": 7.994584795861249e-05,
      "loss": 0.9828,
      "step": 310
    },
    {
      "epoch": 6.176470588235294,
      "grad_norm": 0.48828125,
      "learning_rate": 7.660555181983518e-05,
      "loss": 0.9815,
      "step": 315
    },
    {
      "epoch": 6.2745098039215685,
      "grad_norm": 0.515625,
      "learning_rate": 7.329265153251285e-05,
      "loss": 0.9688,
      "step": 320
    },
    {
      "epoch": 6.372549019607844,
      "grad_norm": 0.48828125,
      "learning_rate": 7.001102663778533e-05,
      "loss": 0.9786,
      "step": 325
    },
    {
      "epoch": 6.470588235294118,
      "grad_norm": 0.50390625,
      "learning_rate": 6.676452005203406e-05,
      "loss": 0.9787,
      "step": 330
    },
    {
      "epoch": 6.568627450980392,
      "grad_norm": 0.5546875,
      "learning_rate": 6.355693356667065e-05,
      "loss": 0.9836,
      "step": 335
    },
    {
      "epoch": 6.666666666666667,
      "grad_norm": 0.490234375,
      "learning_rate": 6.039202339608432e-05,
      "loss": 0.9728,
      "step": 340
    },
    {
      "epoch": 6.764705882352941,
      "grad_norm": 0.486328125,
      "learning_rate": 5.727349577896194e-05,
      "loss": 0.9746,
      "step": 345
    },
    {
      "epoch": 6.862745098039216,
      "grad_norm": 0.4921875,
      "learning_rate": 5.4205002638131406e-05,
      "loss": 0.9733,
      "step": 350
    },
    {
      "epoch": 6.96078431372549,
      "grad_norm": 0.4453125,
      "learning_rate": 5.119013730401152e-05,
      "loss": 0.9739,
      "step": 355
    },
    {
      "epoch": 7.0,
      "eval_loss": 2.5337109565734863,
      "eval_runtime": 0.5399,
      "eval_samples_per_second": 18.521,
      "eval_steps_per_second": 1.852,
      "step": 357
    },
    {
      "epoch": 7.0588235294117645,
      "grad_norm": 0.451171875,
      "learning_rate": 4.823243030667576e-05,
      "loss": 0.9639,
      "step": 360
    },
    {
      "epoch": 7.1568627450980395,
      "grad_norm": 0.484375,
      "learning_rate": 4.533534524145756e-05,
      "loss": 0.9543,
      "step": 365
    },
    {
      "epoch": 7.254901960784314,
      "grad_norm": 0.4140625,
      "learning_rate": 4.250227471293935e-05,
      "loss": 0.9674,
      "step": 370
    },
    {
      "epoch": 7.352941176470588,
      "grad_norm": 0.44921875,
      "learning_rate": 3.973653636207437e-05,
      "loss": 0.9617,
      "step": 375
    },
    {
      "epoch": 7.450980392156863,
      "grad_norm": 0.439453125,
      "learning_rate": 3.7041368981094024e-05,
      "loss": 0.9616,
      "step": 380
    },
    {
      "epoch": 7.549019607843137,
      "grad_norm": 0.431640625,
      "learning_rate": 3.441992872075027e-05,
      "loss": 0.9685,
      "step": 385
    },
    {
      "epoch": 7.647058823529412,
      "grad_norm": 0.458984375,
      "learning_rate": 3.187528539433458e-05,
      "loss": 0.975,
      "step": 390
    },
    {
      "epoch": 7.745098039215686,
      "grad_norm": 0.466796875,
      "learning_rate": 2.9410418882801683e-05,
      "loss": 0.9707,
      "step": 395
    },
    {
      "epoch": 7.8431372549019605,
      "grad_norm": 0.443359375,
      "learning_rate": 2.702821564520732e-05,
      "loss": 0.9593,
      "step": 400
    },
    {
      "epoch": 7.9411764705882355,
      "grad_norm": 0.408203125,
      "learning_rate": 2.4731465338547556e-05,
      "loss": 0.9626,
      "step": 405
    },
    {
      "epoch": 8.0,
      "eval_loss": 2.5445075035095215,
      "eval_runtime": 0.5398,
      "eval_samples_per_second": 18.525,
      "eval_steps_per_second": 1.852,
      "step": 408
    },
    {
      "epoch": 8.03921568627451,
      "grad_norm": 0.40625,
      "learning_rate": 2.252285755095652e-05,
      "loss": 0.9591,
      "step": 410
    },
    {
      "epoch": 8.137254901960784,
      "grad_norm": 0.40625,
      "learning_rate": 2.0404978652089323e-05,
      "loss": 0.9516,
      "step": 415
    },
    {
      "epoch": 8.235294117647058,
      "grad_norm": 0.412109375,
      "learning_rate": 1.8380308764377842e-05,
      "loss": 0.951,
      "step": 420
    },
    {
      "epoch": 8.333333333333334,
      "grad_norm": 0.447265625,
      "learning_rate": 1.6451218858706374e-05,
      "loss": 0.9559,
      "step": 425
    },
    {
      "epoch": 8.431372549019608,
      "grad_norm": 0.427734375,
      "learning_rate": 1.4619967977908156e-05,
      "loss": 0.9495,
      "step": 430
    },
    {
      "epoch": 8.529411764705882,
      "grad_norm": 0.404296875,
      "learning_rate": 1.2888700591334223e-05,
      "loss": 0.9549,
      "step": 435
    },
    {
      "epoch": 8.627450980392156,
      "grad_norm": 0.41796875,
      "learning_rate": 1.1259444083592585e-05,
      "loss": 0.9636,
      "step": 440
    },
    {
      "epoch": 8.72549019607843,
      "grad_norm": 0.427734375,
      "learning_rate": 9.734106380398023e-06,
      "loss": 0.9547,
      "step": 445
    },
    {
      "epoch": 8.823529411764707,
      "grad_norm": 0.40625,
      "learning_rate": 8.314473714313719e-06,
      "loss": 0.9643,
      "step": 450
    },
    {
      "epoch": 8.92156862745098,
      "grad_norm": 0.4296875,
      "learning_rate": 7.002208532999932e-06,
      "loss": 0.9572,
      "step": 455
    },
    {
      "epoch": 9.0,
      "eval_loss": 2.544271469116211,
      "eval_runtime": 0.5384,
      "eval_samples_per_second": 18.575,
      "eval_steps_per_second": 1.857,
      "step": 459
    },
    {
      "epoch": 9.019607843137255,
      "grad_norm": 0.412109375,
      "learning_rate": 5.798847552420183e-06,
      "loss": 0.9625,
      "step": 460
    },
    {
      "epoch": 9.117647058823529,
      "grad_norm": 0.421875,
      "learning_rate": 4.705799957284351e-06,
      "loss": 0.9482,
      "step": 465
    },
    {
      "epoch": 9.215686274509803,
      "grad_norm": 0.408203125,
      "learning_rate": 3.724345750835878e-06,
      "loss": 0.943,
      "step": 470
    },
    {
      "epoch": 9.313725490196079,
      "grad_norm": 0.41796875,
      "learning_rate": 2.855634255915951e-06,
      "loss": 0.9568,
      "step": 475
    },
    {
      "epoch": 9.411764705882353,
      "grad_norm": 0.421875,
      "learning_rate": 2.100682769059548e-06,
      "loss": 0.9561,
      "step": 480
    },
    {
      "epoch": 9.509803921568627,
      "grad_norm": 0.404296875,
      "learning_rate": 1.4603753691998735e-06,
      "loss": 0.9634,
      "step": 485
    },
    {
      "epoch": 9.607843137254902,
      "grad_norm": 0.400390625,
      "learning_rate": 9.354618823758655e-07,
      "loss": 0.9588,
      "step": 490
    },
    {
      "epoch": 9.705882352941176,
      "grad_norm": 0.408203125,
      "learning_rate": 5.265570036553813e-07,
      "loss": 0.9482,
      "step": 495
    },
    {
      "epoch": 9.803921568627452,
      "grad_norm": 0.427734375,
      "learning_rate": 2.3413957730226144e-07,
      "loss": 0.9663,
      "step": 500
    },
    {
      "epoch": 9.901960784313726,
      "grad_norm": 0.41796875,
      "learning_rate": 5.8552036030179445e-08,
      "loss": 0.9606,
      "step": 505
    },
    {
      "epoch": 10.0,
      "grad_norm": 0.421875,
      "learning_rate": 0.0,
      "loss": 0.9554,
      "step": 510
    },
    {
      "epoch": 10.0,
      "eval_loss": 2.545398235321045,
      "eval_runtime": 0.5358,
      "eval_samples_per_second": 18.663,
      "eval_steps_per_second": 1.866,
      "step": 510
    },
    {
      "epoch": 10.0,
      "step": 510,
      "total_flos": 2.9956040277203354e+17,
      "train_loss": 1.144676014956306,
      "train_runtime": 1752.8789,
      "train_samples_per_second": 13.857,
      "train_steps_per_second": 0.291
    }
  ],
  "logging_steps": 5,
  "max_steps": 510,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 100,
  "total_flos": 2.9956040277203354e+17,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}