{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 400,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "Batch Mean": 1.9420166015625,
      "accuracy": 0.46875,
      "epoch": 0,
      "step": 0
    },
    {
      "epoch": 0.0025,
      "grad_norm": 9.666593551635742,
      "learning_rate": 2.5000000000000004e-07,
      "loss": 0.7076,
      "step": 1
    },
    {
      "Batch Mean": 1.9312744140625,
      "accuracy": 0.4375,
      "epoch": 0.0025,
      "step": 1
    },
    {
      "epoch": 0.005,
      "grad_norm": 7.076696395874023,
      "learning_rate": 5.000000000000001e-07,
      "loss": 0.7073,
      "step": 2
    },
    {
      "Batch Mean": 1.9347686767578125,
      "accuracy": 0.5,
      "epoch": 0.005,
      "step": 2
    },
    {
      "epoch": 0.0075,
      "grad_norm": 6.997435092926025,
      "learning_rate": 7.5e-07,
      "loss": 0.6918,
      "step": 3
    },
    {
      "Batch Mean": 1.94329833984375,
      "accuracy": 0.53125,
      "epoch": 0.0075,
      "step": 3
    },
    {
      "epoch": 0.01,
      "grad_norm": 7.863523006439209,
      "learning_rate": 1.0000000000000002e-06,
      "loss": 0.6877,
      "step": 4
    },
    {
      "Batch Mean": 1.94287109375,
      "accuracy": 0.5703125,
      "epoch": 0.01,
      "step": 4
    },
    {
      "epoch": 0.0125,
      "grad_norm": 6.760534286499023,
      "learning_rate": 1.25e-06,
      "loss": 0.6861,
      "step": 5
    },
    {
      "Batch Mean": 1.97332763671875,
      "accuracy": 0.53125,
      "epoch": 0.0125,
      "step": 5
    },
    {
      "epoch": 0.015,
      "grad_norm": 8.08924674987793,
      "learning_rate": 1.5e-06,
      "loss": 0.6869,
      "step": 6
    },
    {
      "Batch Mean": 2.018096923828125,
      "accuracy": 0.46875,
      "epoch": 0.015,
      "step": 6
    },
    {
      "epoch": 0.0175,
      "grad_norm": 6.236896991729736,
      "learning_rate": 1.75e-06,
      "loss": 0.7053,
      "step": 7
    },
    {
      "Batch Mean": 2.14398193359375,
      "accuracy": 0.5390625,
      "epoch": 0.0175,
      "step": 7
    },
    {
      "epoch": 0.02,
      "grad_norm": 6.7978620529174805,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 0.6925,
      "step": 8
    },
    {
      "Batch Mean": 2.214202880859375,
      "accuracy": 0.4453125,
      "epoch": 0.02,
      "step": 8
    },
    {
      "epoch": 0.0225,
      "grad_norm": 7.25023889541626,
      "learning_rate": 2.25e-06,
      "loss": 0.7048,
      "step": 9
    },
    {
      "Batch Mean": 2.3272705078125,
      "accuracy": 0.5859375,
      "epoch": 0.0225,
      "step": 9
    },
    {
      "epoch": 0.025,
      "grad_norm": 6.383249759674072,
      "learning_rate": 2.5e-06,
      "loss": 0.6815,
      "step": 10
    },
    {
      "Batch Mean": 2.49114990234375,
      "accuracy": 0.6640625,
      "epoch": 0.025,
      "step": 10
    },
    {
      "epoch": 0.0275,
      "grad_norm": 9.219279289245605,
      "learning_rate": 2.7500000000000004e-06,
      "loss": 0.6547,
      "step": 11
    },
    {
      "Batch Mean": 2.550567626953125,
      "accuracy": 0.6328125,
      "epoch": 0.0275,
      "step": 11
    },
    {
      "epoch": 0.03,
      "grad_norm": 7.704563617706299,
      "learning_rate": 3e-06,
      "loss": 0.6432,
      "step": 12
    },
    {
      "Batch Mean": 2.645782470703125,
      "accuracy": 0.671875,
      "epoch": 0.03,
      "step": 12
    },
    {
      "epoch": 0.0325,
      "grad_norm": 7.639222145080566,
      "learning_rate": 3.2500000000000002e-06,
      "loss": 0.6365,
      "step": 13
    },
    {
      "Batch Mean": 2.8451995849609375,
      "accuracy": 0.6171875,
      "epoch": 0.0325,
      "step": 13
    },
    {
      "epoch": 0.035,
      "grad_norm": 13.006075859069824,
      "learning_rate": 3.5e-06,
      "loss": 0.659,
      "step": 14
    },
    {
      "Batch Mean": 3.0160064697265625,
      "accuracy": 0.609375,
      "epoch": 0.035,
      "step": 14
    },
    {
      "epoch": 0.0375,
      "grad_norm": 11.644806861877441,
      "learning_rate": 3.7500000000000005e-06,
      "loss": 0.6574,
      "step": 15
    },
    {
      "Batch Mean": 3.0444812774658203,
      "accuracy": 0.671875,
      "epoch": 0.0375,
      "step": 15
    },
    {
      "epoch": 0.04,
      "grad_norm": 12.809576034545898,
      "learning_rate": 4.000000000000001e-06,
      "loss": 0.6138,
      "step": 16
    },
    {
      "Batch Mean": 3.1425399780273438,
      "accuracy": 0.6875,
      "epoch": 0.04,
      "step": 16
    },
    {
      "epoch": 0.0425,
      "grad_norm": 10.398422241210938,
      "learning_rate": 4.25e-06,
      "loss": 0.5373,
      "step": 17
    },
    {
      "Batch Mean": 2.997763156890869,
      "accuracy": 0.6015625,
      "epoch": 0.0425,
      "step": 17
    },
    {
      "epoch": 0.045,
      "grad_norm": 9.611076354980469,
      "learning_rate": 4.5e-06,
      "loss": 0.5884,
      "step": 18
    },
    {
      "Batch Mean": 2.822673797607422,
      "accuracy": 0.6953125,
      "epoch": 0.045,
      "step": 18
    },
    {
      "epoch": 0.0475,
      "grad_norm": 11.101300239562988,
      "learning_rate": 4.75e-06,
      "loss": 0.5819,
      "step": 19
    },
    {
      "Batch Mean": 2.9509963989257812,
      "accuracy": 0.6796875,
      "epoch": 0.0475,
      "step": 19
    },
    {
      "epoch": 0.05,
      "grad_norm": 11.16808032989502,
      "learning_rate": 5e-06,
      "loss": 0.5849,
      "step": 20
    },
    {
      "Batch Mean": 2.7339065074920654,
      "accuracy": 0.609375,
      "epoch": 0.05,
      "step": 20
    },
    {
      "epoch": 0.0525,
      "grad_norm": 13.900754928588867,
      "learning_rate": 4.986842105263158e-06,
      "loss": 0.6193,
      "step": 21
    },
    {
      "Batch Mean": 2.5777530670166016,
      "accuracy": 0.71875,
      "epoch": 0.0525,
      "step": 21
    },
    {
      "epoch": 0.055,
      "grad_norm": 9.619379997253418,
      "learning_rate": 4.973684210526316e-06,
      "loss": 0.599,
      "step": 22
    },
    {
      "Batch Mean": 2.6994237899780273,
      "accuracy": 0.65625,
      "epoch": 0.055,
      "step": 22
    },
    {
      "epoch": 0.0575,
      "grad_norm": 9.464532852172852,
      "learning_rate": 4.960526315789474e-06,
      "loss": 0.6118,
      "step": 23
    },
    {
      "Batch Mean": 2.70928955078125,
      "accuracy": 0.6484375,
      "epoch": 0.0575,
      "step": 23
    },
    {
      "epoch": 0.06,
      "grad_norm": 8.05068588256836,
      "learning_rate": 4.947368421052632e-06,
      "loss": 0.6214,
      "step": 24
    },
    {
      "Batch Mean": 2.8170089721679688,
      "accuracy": 0.6796875,
      "epoch": 0.06,
      "step": 24
    },
    {
      "epoch": 0.0625,
      "grad_norm": 9.21177864074707,
      "learning_rate": 4.9342105263157895e-06,
      "loss": 0.6165,
      "step": 25
    },
    {
      "Batch Mean": 2.7590713500976562,
      "accuracy": 0.6171875,
      "epoch": 0.0625,
      "step": 25
    },
    {
      "epoch": 0.065,
      "grad_norm": 6.078210353851318,
      "learning_rate": 4.921052631578948e-06,
      "loss": 0.6027,
      "step": 26
    },
    {
      "Batch Mean": 2.8484420776367188,
      "accuracy": 0.6796875,
      "epoch": 0.065,
      "step": 26
    },
    {
      "epoch": 0.0675,
      "grad_norm": 5.933135032653809,
      "learning_rate": 4.907894736842106e-06,
      "loss": 0.5721,
      "step": 27
    },
    {
      "Batch Mean": 2.9572601318359375,
      "accuracy": 0.6796875,
      "epoch": 0.0675,
      "step": 27
    },
    {
      "epoch": 0.07,
      "grad_norm": 6.4003143310546875,
      "learning_rate": 4.894736842105264e-06,
      "loss": 0.5626,
      "step": 28
    },
    {
      "Batch Mean": 2.9868011474609375,
      "accuracy": 0.6796875,
      "epoch": 0.07,
      "step": 28
    },
    {
      "epoch": 0.0725,
      "grad_norm": 4.7936859130859375,
      "learning_rate": 4.881578947368422e-06,
      "loss": 0.5809,
      "step": 29
    },
    {
      "Batch Mean": 3.106109619140625,
      "accuracy": 0.65625,
      "epoch": 0.0725,
      "step": 29
    },
    {
      "epoch": 0.075,
      "grad_norm": 5.339725017547607,
      "learning_rate": 4.8684210526315795e-06,
      "loss": 0.6132,
      "step": 30
    },
    {
      "Batch Mean": 3.29296875,
      "accuracy": 0.734375,
      "epoch": 0.075,
      "step": 30
    },
    {
      "epoch": 0.0775,
      "grad_norm": 5.850327491760254,
      "learning_rate": 4.855263157894737e-06,
      "loss": 0.584,
      "step": 31
    },
    {
      "Batch Mean": 3.51275634765625,
      "accuracy": 0.65625,
      "epoch": 0.0775,
      "step": 31
    },
    {
      "epoch": 0.08,
      "grad_norm": 6.158403396606445,
      "learning_rate": 4.842105263157895e-06,
      "loss": 0.5911,
      "step": 32
    },
    {
      "Batch Mean": 3.4871044158935547,
      "accuracy": 0.71875,
      "epoch": 0.08,
      "step": 32
    },
    {
      "epoch": 0.0825,
      "grad_norm": 5.687869071960449,
      "learning_rate": 4.828947368421053e-06,
      "loss": 0.5243,
      "step": 33
    },
    {
      "Batch Mean": 3.6311264038085938,
      "accuracy": 0.8203125,
      "epoch": 0.0825,
      "step": 33
    },
    {
      "epoch": 0.085,
      "grad_norm": 6.883860111236572,
      "learning_rate": 4.815789473684211e-06,
      "loss": 0.4525,
      "step": 34
    },
    {
      "Batch Mean": 3.747730255126953,
      "accuracy": 0.6875,
      "epoch": 0.085,
      "step": 34
    },
    {
      "epoch": 0.0875,
      "grad_norm": 8.108345031738281,
      "learning_rate": 4.802631578947369e-06,
      "loss": 0.5392,
      "step": 35
    },
    {
      "Batch Mean": 4.052231788635254,
      "accuracy": 0.671875,
      "epoch": 0.0875,
      "step": 35
    },
    {
      "epoch": 0.09,
      "grad_norm": 8.877079963684082,
      "learning_rate": 4.789473684210527e-06,
      "loss": 0.5759,
      "step": 36
    },
    {
      "Batch Mean": 4.003006458282471,
      "accuracy": 0.7421875,
      "epoch": 0.09,
      "step": 36
    },
    {
      "epoch": 0.0925,
      "grad_norm": 9.152173042297363,
      "learning_rate": 4.7763157894736844e-06,
      "loss": 0.5118,
      "step": 37
    },
    {
      "Batch Mean": 4.063720703125,
      "accuracy": 0.6171875,
      "epoch": 0.0925,
      "step": 37
    },
    {
      "epoch": 0.095,
      "grad_norm": 10.134416580200195,
      "learning_rate": 4.763157894736842e-06,
      "loss": 0.6653,
      "step": 38
    },
    {
      "Batch Mean": 4.406402587890625,
      "accuracy": 0.703125,
      "epoch": 0.095,
      "step": 38
    },
    {
      "epoch": 0.0975,
      "grad_norm": 7.997857570648193,
      "learning_rate": 4.75e-06,
      "loss": 0.525,
      "step": 39
    },
    {
      "Batch Mean": 4.049163818359375,
      "accuracy": 0.7578125,
      "epoch": 0.0975,
      "step": 39
    },
    {
      "epoch": 0.1,
      "grad_norm": 8.24782657623291,
      "learning_rate": 4.736842105263158e-06,
      "loss": 0.4568,
      "step": 40
    },
    {
      "Batch Mean": 4.293998718261719,
      "accuracy": 0.7890625,
      "epoch": 0.1,
      "step": 40
    },
    {
      "epoch": 0.1025,
      "grad_norm": 7.328802108764648,
      "learning_rate": 4.723684210526316e-06,
      "loss": 0.4618,
      "step": 41
    },
    {
      "Batch Mean": 4.420158386230469,
      "accuracy": 0.75,
      "epoch": 0.1025,
      "step": 41
    },
    {
      "epoch": 0.105,
      "grad_norm": 7.795598030090332,
      "learning_rate": 4.710526315789474e-06,
      "loss": 0.5121,
      "step": 42
    },
    {
      "Batch Mean": 4.374538421630859,
      "accuracy": 0.6953125,
      "epoch": 0.105,
      "step": 42
    },
    {
      "epoch": 0.1075,
      "grad_norm": 9.923321723937988,
      "learning_rate": 4.697368421052632e-06,
      "loss": 0.5919,
      "step": 43
    },
    {
      "Batch Mean": 4.3918304443359375,
      "accuracy": 0.78125,
      "epoch": 0.1075,
      "step": 43
    },
    {
      "epoch": 0.11,
      "grad_norm": 6.950074672698975,
      "learning_rate": 4.68421052631579e-06,
      "loss": 0.499,
      "step": 44
    },
    {
      "Batch Mean": 4.343505859375,
      "accuracy": 0.7578125,
      "epoch": 0.11,
      "step": 44
    },
    {
      "epoch": 0.1125,
      "grad_norm": 7.051249980926514,
      "learning_rate": 4.671052631578948e-06,
      "loss": 0.5362,
      "step": 45
    },
    {
      "Batch Mean": 4.268529891967773,
      "accuracy": 0.7265625,
      "epoch": 0.1125,
      "step": 45
    },
    {
      "epoch": 0.115,
      "grad_norm": 7.27433443069458,
      "learning_rate": 4.657894736842106e-06,
      "loss": 0.527,
      "step": 46
    },
    {
      "Batch Mean": 4.0211334228515625,
      "accuracy": 0.796875,
      "epoch": 0.115,
      "step": 46
    },
    {
      "epoch": 0.1175,
      "grad_norm": 7.680591106414795,
      "learning_rate": 4.6447368421052635e-06,
      "loss": 0.5004,
      "step": 47
    },
    {
      "Batch Mean": 4.125762939453125,
      "accuracy": 0.7578125,
      "epoch": 0.1175,
      "step": 47
    },
    {
      "epoch": 0.12,
      "grad_norm": 6.570550441741943,
      "learning_rate": 4.631578947368421e-06,
      "loss": 0.4887,
      "step": 48
    },
    {
      "Batch Mean": 3.82733154296875,
      "accuracy": 0.6953125,
      "epoch": 0.12,
      "step": 48
    },
    {
      "epoch": 0.1225,
      "grad_norm": 6.966435432434082,
      "learning_rate": 4.618421052631579e-06,
      "loss": 0.5351,
      "step": 49
    },
    {
      "Batch Mean": 3.548530101776123,
      "accuracy": 0.6796875,
      "epoch": 0.1225,
      "step": 49
    },
    {
      "epoch": 0.125,
      "grad_norm": 9.178398132324219,
      "learning_rate": 4.605263157894737e-06,
      "loss": 0.6348,
      "step": 50
    },
    {
      "Batch Mean": 3.5271453857421875,
      "accuracy": 0.7421875,
      "epoch": 0.125,
      "step": 50
    },
    {
      "epoch": 0.1275,
      "grad_norm": 7.21920108795166,
      "learning_rate": 4.592105263157895e-06,
      "loss": 0.5138,
      "step": 51
    },
    {
      "Batch Mean": 3.3803443908691406,
      "accuracy": 0.75,
      "epoch": 0.1275,
      "step": 51
    },
    {
      "epoch": 0.13,
      "grad_norm": 6.866484642028809,
      "learning_rate": 4.578947368421053e-06,
      "loss": 0.5163,
      "step": 52
    },
    {
      "Batch Mean": 3.42218017578125,
      "accuracy": 0.734375,
      "epoch": 0.13,
      "step": 52
    },
    {
      "epoch": 0.1325,
      "grad_norm": 5.488794803619385,
      "learning_rate": 4.565789473684211e-06,
      "loss": 0.5135,
      "step": 53
    },
    {
      "Batch Mean": 3.271286964416504,
      "accuracy": 0.6484375,
      "epoch": 0.1325,
      "step": 53
    },
    {
      "epoch": 0.135,
      "grad_norm": 6.306389808654785,
      "learning_rate": 4.552631578947369e-06,
      "loss": 0.5649,
      "step": 54
    },
    {
      "Batch Mean": 3.0698814392089844,
      "accuracy": 0.7578125,
      "epoch": 0.135,
      "step": 54
    },
    {
      "epoch": 0.1375,
      "grad_norm": 6.536716461181641,
      "learning_rate": 4.539473684210527e-06,
      "loss": 0.5415,
      "step": 55
    },
    {
      "Batch Mean": 3.088502883911133,
      "accuracy": 0.75,
      "epoch": 0.1375,
      "step": 55
    },
    {
      "epoch": 0.14,
      "grad_norm": 7.343382358551025,
      "learning_rate": 4.526315789473685e-06,
      "loss": 0.5206,
      "step": 56
    },
    {
      "Batch Mean": 3.2023324966430664,
      "accuracy": 0.6796875,
      "epoch": 0.14,
      "step": 56
    },
    {
      "epoch": 0.1425,
      "grad_norm": 7.171923637390137,
      "learning_rate": 4.513157894736843e-06,
      "loss": 0.5765,
      "step": 57
    },
    {
      "Batch Mean": 2.9620323181152344,
      "accuracy": 0.75,
      "epoch": 0.1425,
      "step": 57
    },
    {
      "epoch": 0.145,
      "grad_norm": 7.703383922576904,
      "learning_rate": 4.5e-06,
      "loss": 0.5144,
      "step": 58
    },
    {
      "Batch Mean": 2.995695114135742,
      "accuracy": 0.734375,
      "epoch": 0.145,
      "step": 58
    },
    {
      "epoch": 0.1475,
      "grad_norm": 8.559469223022461,
      "learning_rate": 4.4868421052631584e-06,
      "loss": 0.4926,
      "step": 59
    },
    {
      "Batch Mean": 3.2689762115478516,
      "accuracy": 0.7421875,
      "epoch": 0.1475,
      "step": 59
    },
    {
      "epoch": 0.15,
      "grad_norm": 6.663382530212402,
      "learning_rate": 4.473684210526316e-06,
      "loss": 0.4705,
      "step": 60
    },
    {
      "Batch Mean": 2.9887285232543945,
      "accuracy": 0.7265625,
      "epoch": 0.15,
      "step": 60
    },
    {
      "epoch": 0.1525,
      "grad_norm": 7.883590221405029,
      "learning_rate": 4.460526315789474e-06,
      "loss": 0.5447,
      "step": 61
    },
    {
      "Batch Mean": 2.901198387145996,
      "accuracy": 0.78125,
      "epoch": 0.1525,
      "step": 61
    },
    {
      "epoch": 0.155,
      "grad_norm": 7.087621688842773,
      "learning_rate": 4.447368421052632e-06,
      "loss": 0.4382,
      "step": 62
    },
    {
      "Batch Mean": 2.978053569793701,
      "accuracy": 0.7265625,
      "epoch": 0.155,
      "step": 62
    },
    {
      "epoch": 0.1575,
      "grad_norm": 8.361074447631836,
      "learning_rate": 4.43421052631579e-06,
      "loss": 0.5379,
      "step": 63
    },
    {
      "Batch Mean": 2.9929721355438232,
      "accuracy": 0.765625,
      "epoch": 0.1575,
      "step": 63
    },
    {
      "epoch": 0.16,
      "grad_norm": 10.397054672241211,
      "learning_rate": 4.4210526315789476e-06,
      "loss": 0.489,
      "step": 64
    },
    {
      "Batch Mean": 2.9623823165893555,
      "accuracy": 0.734375,
      "epoch": 0.16,
      "step": 64
    },
    {
      "epoch": 0.1625,
      "grad_norm": 11.552237510681152,
      "learning_rate": 4.407894736842105e-06,
      "loss": 0.5394,
      "step": 65
    },
    {
      "Batch Mean": 2.897951602935791,
      "accuracy": 0.734375,
      "epoch": 0.1625,
      "step": 65
    },
    {
      "epoch": 0.165,
      "grad_norm": 9.772311210632324,
      "learning_rate": 4.394736842105263e-06,
      "loss": 0.5274,
      "step": 66
    },
    {
      "Batch Mean": 2.8228421211242676,
      "accuracy": 0.75,
      "epoch": 0.165,
      "step": 66
    },
    {
      "epoch": 0.1675,
      "grad_norm": 8.786587715148926,
      "learning_rate": 4.381578947368421e-06,
      "loss": 0.5151,
      "step": 67
    },
    {
      "Batch Mean": 2.93172550201416,
      "accuracy": 0.8125,
      "epoch": 0.1675,
      "step": 67
    },
    {
      "epoch": 0.17,
      "grad_norm": 7.272209167480469,
      "learning_rate": 4.368421052631579e-06,
      "loss": 0.4259,
      "step": 68
    },
    {
      "Batch Mean": 3.0047380924224854,
      "accuracy": 0.6953125,
      "epoch": 0.17,
      "step": 68
    },
    {
      "epoch": 0.1725,
      "grad_norm": 10.469968795776367,
      "learning_rate": 4.3552631578947375e-06,
      "loss": 0.5409,
      "step": 69
    },
    {
      "Batch Mean": 3.2963790893554688,
      "accuracy": 0.75,
      "epoch": 0.1725,
      "step": 69
    },
    {
      "epoch": 0.175,
      "grad_norm": 8.683908462524414,
      "learning_rate": 4.342105263157895e-06,
      "loss": 0.4782,
      "step": 70
    },
    {
      "Batch Mean": 3.195692300796509,
      "accuracy": 0.7734375,
      "epoch": 0.175,
      "step": 70
    },
    {
      "epoch": 0.1775,
      "grad_norm": 7.932145118713379,
      "learning_rate": 4.328947368421053e-06,
      "loss": 0.4693,
      "step": 71
    },
    {
      "Batch Mean": 3.188863754272461,
      "accuracy": 0.8125,
      "epoch": 0.1775,
      "step": 71
    },
    {
      "epoch": 0.18,
      "grad_norm": 6.82938814163208,
      "learning_rate": 4.315789473684211e-06,
      "loss": 0.3917,
      "step": 72
    },
    {
      "Batch Mean": 3.1435837745666504,
      "accuracy": 0.7421875,
      "epoch": 0.18,
      "step": 72
    },
    {
      "epoch": 0.1825,
      "grad_norm": 6.974360942840576,
      "learning_rate": 4.302631578947369e-06,
      "loss": 0.49,
      "step": 73
    },
    {
      "Batch Mean": 3.5599288940429688,
      "accuracy": 0.78125,
      "epoch": 0.1825,
      "step": 73
    },
    {
      "epoch": 0.185,
      "grad_norm": 7.391234397888184,
      "learning_rate": 4.289473684210527e-06,
      "loss": 0.507,
      "step": 74
    },
    {
      "Batch Mean": 3.1217479705810547,
      "accuracy": 0.7421875,
      "epoch": 0.185,
      "step": 74
    },
    {
      "epoch": 0.1875,
      "grad_norm": 6.925199508666992,
      "learning_rate": 4.276315789473684e-06,
      "loss": 0.5203,
      "step": 75
    },
    {
      "Batch Mean": 3.3439712524414062,
      "accuracy": 0.7109375,
      "epoch": 0.1875,
      "step": 75
    },
    {
      "epoch": 0.19,
      "grad_norm": 8.055448532104492,
      "learning_rate": 4.2631578947368425e-06,
      "loss": 0.5173,
      "step": 76
    },
    {
      "Batch Mean": 3.372138023376465,
      "accuracy": 0.7734375,
      "epoch": 0.19,
      "step": 76
    },
    {
      "epoch": 0.1925,
      "grad_norm": 7.226714611053467,
      "learning_rate": 4.25e-06,
      "loss": 0.487,
      "step": 77
    },
    {
      "Batch Mean": 3.52044677734375,
      "accuracy": 0.8359375,
      "epoch": 0.1925,
      "step": 77
    },
    {
      "epoch": 0.195,
      "grad_norm": 7.280280113220215,
      "learning_rate": 4.236842105263158e-06,
      "loss": 0.4553,
      "step": 78
    },
    {
      "Batch Mean": 3.2147364616394043,
      "accuracy": 0.8359375,
      "epoch": 0.195,
      "step": 78
    },
    {
      "epoch": 0.1975,
      "grad_norm": 6.3655781745910645,
      "learning_rate": 4.223684210526316e-06,
      "loss": 0.4328,
      "step": 79
    },
    {
      "Batch Mean": 3.4875216484069824,
      "accuracy": 0.8203125,
      "epoch": 0.1975,
      "step": 79
    },
    {
      "epoch": 0.2,
      "grad_norm": 6.6272759437561035,
      "learning_rate": 4.210526315789474e-06,
      "loss": 0.4119,
      "step": 80
    },
    {
      "Batch Mean": 3.427578926086426,
      "accuracy": 0.78125,
      "epoch": 0.2,
      "step": 80
    },
    {
      "epoch": 0.2025,
      "grad_norm": 8.57567024230957,
      "learning_rate": 4.197368421052632e-06,
      "loss": 0.4891,
      "step": 81
    },
    {
      "Batch Mean": 3.7187185287475586,
      "accuracy": 0.7578125,
      "epoch": 0.2025,
      "step": 81
    },
    {
      "epoch": 0.205,
      "grad_norm": 7.171870708465576,
      "learning_rate": 4.18421052631579e-06,
      "loss": 0.439,
      "step": 82
    },
    {
      "Batch Mean": 3.990264892578125,
      "accuracy": 0.7734375,
      "epoch": 0.205,
      "step": 82
    },
    {
      "epoch": 0.2075,
      "grad_norm": 8.19045639038086,
      "learning_rate": 4.171052631578948e-06,
      "loss": 0.4541,
      "step": 83
    },
    {
      "Batch Mean": 4.509843826293945,
      "accuracy": 0.7265625,
      "epoch": 0.2075,
      "step": 83
    },
    {
      "epoch": 0.21,
      "grad_norm": 9.934694290161133,
      "learning_rate": 4.157894736842106e-06,
      "loss": 0.5258,
      "step": 84
    },
    {
      "Batch Mean": 4.694008827209473,
      "accuracy": 0.8125,
      "epoch": 0.21,
      "step": 84
    },
    {
      "epoch": 0.2125,
      "grad_norm": 7.037469863891602,
      "learning_rate": 4.144736842105263e-06,
      "loss": 0.4123,
      "step": 85
    },
    {
      "Batch Mean": 4.4774580001831055,
      "accuracy": 0.71875,
      "epoch": 0.2125,
      "step": 85
    },
    {
      "epoch": 0.215,
      "grad_norm": 9.059208869934082,
      "learning_rate": 4.1315789473684216e-06,
      "loss": 0.541,
      "step": 86
    },
    {
      "Batch Mean": 4.323337078094482,
      "accuracy": 0.7734375,
      "epoch": 0.215,
      "step": 86
    },
    {
      "epoch": 0.2175,
      "grad_norm": 7.794407367706299,
      "learning_rate": 4.118421052631579e-06,
      "loss": 0.4797,
      "step": 87
    },
    {
      "Batch Mean": 4.874683380126953,
      "accuracy": 0.7578125,
      "epoch": 0.2175,
      "step": 87
    },
    {
      "epoch": 0.22,
      "grad_norm": 9.554245948791504,
      "learning_rate": 4.105263157894737e-06,
      "loss": 0.5024,
      "step": 88
    },
    {
      "Batch Mean": 4.5128173828125,
      "accuracy": 0.8046875,
      "epoch": 0.22,
      "step": 88
    },
    {
      "epoch": 0.2225,
      "grad_norm": 7.580309867858887,
      "learning_rate": 4.092105263157895e-06,
      "loss": 0.4127,
      "step": 89
    },
    {
      "Batch Mean": 4.122764587402344,
      "accuracy": 0.7890625,
      "epoch": 0.2225,
      "step": 89
    },
    {
      "epoch": 0.225,
      "grad_norm": 7.205588340759277,
      "learning_rate": 4.078947368421053e-06,
      "loss": 0.4207,
      "step": 90
    },
    {
      "Batch Mean": 4.157896041870117,
      "accuracy": 0.765625,
      "epoch": 0.225,
      "step": 90
    },
    {
      "epoch": 0.2275,
      "grad_norm": 9.179079055786133,
      "learning_rate": 4.065789473684211e-06,
      "loss": 0.505,
      "step": 91
    },
    {
      "Batch Mean": 4.137635231018066,
      "accuracy": 0.7890625,
      "epoch": 0.2275,
      "step": 91
    },
    {
      "epoch": 0.23,
      "grad_norm": 6.633444309234619,
      "learning_rate": 4.052631578947368e-06,
      "loss": 0.3927,
      "step": 92
    },
    {
      "Batch Mean": 3.7938804626464844,
      "accuracy": 0.828125,
      "epoch": 0.23,
      "step": 92
    },
    {
      "epoch": 0.2325,
      "grad_norm": 6.795446395874023,
      "learning_rate": 4.0394736842105265e-06,
      "loss": 0.4159,
      "step": 93
    },
    {
      "Batch Mean": 3.357052803039551,
      "accuracy": 0.8046875,
      "epoch": 0.2325,
      "step": 93
    },
    {
      "epoch": 0.235,
      "grad_norm": 7.158699989318848,
      "learning_rate": 4.026315789473684e-06,
      "loss": 0.4018,
      "step": 94
    },
    {
      "Batch Mean": 3.3779516220092773,
      "accuracy": 0.75,
      "epoch": 0.235,
      "step": 94
    },
    {
      "epoch": 0.2375,
      "grad_norm": 7.249587059020996,
      "learning_rate": 4.013157894736842e-06,
      "loss": 0.4374,
      "step": 95
    },
    {
      "Batch Mean": 3.2399396896362305,
      "accuracy": 0.8046875,
      "epoch": 0.2375,
      "step": 95
    },
    {
      "epoch": 0.24,
      "grad_norm": 7.230793476104736,
      "learning_rate": 4.000000000000001e-06,
      "loss": 0.4345,
      "step": 96
    },
    {
      "Batch Mean": 3.1136474609375,
      "accuracy": 0.8046875,
      "epoch": 0.24,
      "step": 96
    },
    {
      "epoch": 0.2425,
      "grad_norm": 8.101882934570312,
      "learning_rate": 3.986842105263158e-06,
      "loss": 0.4576,
      "step": 97
    },
    {
      "Batch Mean": 3.214869260787964,
      "accuracy": 0.8125,
      "epoch": 0.2425,
      "step": 97
    },
    {
      "epoch": 0.245,
      "grad_norm": 8.142618179321289,
      "learning_rate": 3.9736842105263165e-06,
      "loss": 0.4481,
      "step": 98
    },
    {
      "Batch Mean": 3.2699995040893555,
      "accuracy": 0.8046875,
      "epoch": 0.245,
      "step": 98
    },
    {
      "epoch": 0.2475,
      "grad_norm": 8.672248840332031,
      "learning_rate": 3.960526315789474e-06,
      "loss": 0.4518,
      "step": 99
    },
    {
      "Batch Mean": 3.287128448486328,
      "accuracy": 0.7578125,
      "epoch": 0.2475,
      "step": 99
    },
    {
      "epoch": 0.25,
      "grad_norm": 9.054008483886719,
      "learning_rate": 3.947368421052632e-06,
      "loss": 0.4934,
      "step": 100
    },
    {
      "Batch Mean": 2.6553401947021484,
      "accuracy": 0.7734375,
      "epoch": 0.25,
      "step": 100
    },
    {
      "epoch": 0.2525,
      "grad_norm": 9.374197006225586,
      "learning_rate": 3.93421052631579e-06,
      "loss": 0.5046,
      "step": 101
    },
    {
      "Batch Mean": 2.436018943786621,
      "accuracy": 0.78125,
      "epoch": 0.2525,
      "step": 101
    },
    {
      "epoch": 0.255,
      "grad_norm": 9.066594123840332,
      "learning_rate": 3.921052631578947e-06,
      "loss": 0.5216,
      "step": 102
    },
    {
      "Batch Mean": 2.2069954872131348,
      "accuracy": 0.7734375,
      "epoch": 0.255,
      "step": 102
    },
    {
      "epoch": 0.2575,
      "grad_norm": 6.94660758972168,
      "learning_rate": 3.907894736842106e-06,
      "loss": 0.4278,
      "step": 103
    },
    {
      "Batch Mean": 2.225872755050659,
      "accuracy": 0.7734375,
      "epoch": 0.2575,
      "step": 103
    },
    {
      "epoch": 0.26,
      "grad_norm": 6.518649101257324,
      "learning_rate": 3.894736842105263e-06,
      "loss": 0.4473,
      "step": 104
    },
    {
      "Batch Mean": 2.217318058013916,
      "accuracy": 0.796875,
      "epoch": 0.26,
      "step": 104
    },
    {
      "epoch": 0.2625,
      "grad_norm": 6.582048416137695,
      "learning_rate": 3.8815789473684214e-06,
      "loss": 0.4356,
      "step": 105
    },
    {
      "Batch Mean": 1.917119026184082,
      "accuracy": 0.828125,
      "epoch": 0.2625,
      "step": 105
    },
    {
      "epoch": 0.265,
      "grad_norm": 7.497337818145752,
      "learning_rate": 3.868421052631579e-06,
      "loss": 0.3825,
      "step": 106
    },
    {
      "Batch Mean": 1.5780539512634277,
      "accuracy": 0.7578125,
      "epoch": 0.265,
      "step": 106
    },
    {
      "epoch": 0.2675,
      "grad_norm": 8.559232711791992,
      "learning_rate": 3.855263157894737e-06,
      "loss": 0.4984,
      "step": 107
    },
    {
      "Batch Mean": 1.9931042194366455,
      "accuracy": 0.8203125,
      "epoch": 0.2675,
      "step": 107
    },
    {
      "epoch": 0.27,
      "grad_norm": 6.91691780090332,
      "learning_rate": 3.842105263157895e-06,
      "loss": 0.401,
      "step": 108
    },
    {
      "Batch Mean": 2.131106376647949,
      "accuracy": 0.8046875,
      "epoch": 0.27,
      "step": 108
    },
    {
      "epoch": 0.2725,
      "grad_norm": 6.739565849304199,
      "learning_rate": 3.828947368421053e-06,
      "loss": 0.3582,
      "step": 109
    },
    {
      "Batch Mean": 2.5897467136383057,
      "accuracy": 0.8671875,
      "epoch": 0.2725,
      "step": 109
    },
    {
      "epoch": 0.275,
      "grad_norm": 8.209200859069824,
      "learning_rate": 3.815789473684211e-06,
      "loss": 0.3216,
      "step": 110
    },
    {
      "Batch Mean": 2.6740541458129883,
      "accuracy": 0.75,
      "epoch": 0.275,
      "step": 110
    },
    {
      "epoch": 0.2775,
      "grad_norm": 10.885510444641113,
      "learning_rate": 3.802631578947369e-06,
      "loss": 0.5196,
      "step": 111
    },
    {
      "Batch Mean": 2.9480109214782715,
      "accuracy": 0.703125,
      "epoch": 0.2775,
      "step": 111
    },
    {
      "epoch": 0.28,
      "grad_norm": 13.55850887298584,
      "learning_rate": 3.789473684210527e-06,
      "loss": 0.5931,
      "step": 112
    },
    {
      "Batch Mean": 3.3308558464050293,
      "accuracy": 0.8515625,
      "epoch": 0.28,
      "step": 112
    },
    {
      "epoch": 0.2825,
      "grad_norm": 9.246851921081543,
      "learning_rate": 3.7763157894736847e-06,
      "loss": 0.3901,
      "step": 113
    },
    {
      "Batch Mean": 2.8297948837280273,
      "accuracy": 0.7890625,
      "epoch": 0.2825,
      "step": 113
    },
    {
      "epoch": 0.285,
      "grad_norm": 9.644515037536621,
      "learning_rate": 3.7631578947368426e-06,
      "loss": 0.4573,
      "step": 114
    },
    {
      "Batch Mean": 3.7451376914978027,
      "accuracy": 0.7265625,
      "epoch": 0.285,
      "step": 114
    },
    {
      "epoch": 0.2875,
      "grad_norm": 12.58398151397705,
      "learning_rate": 3.7500000000000005e-06,
      "loss": 0.5545,
      "step": 115
    },
    {
      "Batch Mean": 3.6171021461486816,
      "accuracy": 0.7890625,
      "epoch": 0.2875,
      "step": 115
    },
    {
      "epoch": 0.29,
      "grad_norm": 9.254029273986816,
      "learning_rate": 3.736842105263158e-06,
      "loss": 0.4333,
      "step": 116
    },
    {
      "Batch Mean": 2.7716548442840576,
      "accuracy": 0.8046875,
      "epoch": 0.29,
      "step": 116
    },
    {
      "epoch": 0.2925,
      "grad_norm": 10.669205665588379,
      "learning_rate": 3.723684210526316e-06,
      "loss": 0.4602,
      "step": 117
    },
    {
      "Batch Mean": 2.781661033630371,
      "accuracy": 0.7734375,
      "epoch": 0.2925,
      "step": 117
    },
    {
      "epoch": 0.295,
      "grad_norm": 10.094223976135254,
      "learning_rate": 3.710526315789474e-06,
      "loss": 0.528,
      "step": 118
    },
    {
      "Batch Mean": 2.198807716369629,
      "accuracy": 0.7578125,
      "epoch": 0.295,
      "step": 118
    },
    {
      "epoch": 0.2975,
      "grad_norm": 7.635989189147949,
      "learning_rate": 3.6973684210526317e-06,
      "loss": 0.4268,
      "step": 119
    },
    {
      "Batch Mean": 2.0033488273620605,
      "accuracy": 0.7734375,
      "epoch": 0.2975,
      "step": 119
    },
    {
      "epoch": 0.3,
      "grad_norm": 7.30623197555542,
      "learning_rate": 3.6842105263157896e-06,
      "loss": 0.4528,
      "step": 120
    },
    {
      "Batch Mean": 1.8473689556121826,
      "accuracy": 0.859375,
      "epoch": 0.3,
      "step": 120
    },
    {
      "epoch": 0.3025,
      "grad_norm": 6.51129674911499,
      "learning_rate": 3.6710526315789476e-06,
      "loss": 0.4022,
      "step": 121
    },
    {
      "Batch Mean": 1.650212287902832,
      "accuracy": 0.78125,
      "epoch": 0.3025,
      "step": 121
    },
    {
      "epoch": 0.305,
      "grad_norm": 7.191035270690918,
      "learning_rate": 3.657894736842106e-06,
      "loss": 0.4487,
      "step": 122
    },
    {
      "Batch Mean": 1.7975679636001587,
      "accuracy": 0.7890625,
      "epoch": 0.305,
      "step": 122
    },
    {
      "epoch": 0.3075,
      "grad_norm": 6.7035651206970215,
      "learning_rate": 3.644736842105264e-06,
      "loss": 0.4338,
      "step": 123
    },
    {
      "Batch Mean": 1.5073213577270508,
      "accuracy": 0.8125,
      "epoch": 0.3075,
      "step": 123
    },
    {
      "epoch": 0.31,
      "grad_norm": 6.955989837646484,
      "learning_rate": 3.6315789473684217e-06,
      "loss": 0.4144,
      "step": 124
    },
    {
      "Batch Mean": 1.7353585958480835,
      "accuracy": 0.75,
      "epoch": 0.31,
      "step": 124
    },
    {
      "epoch": 0.3125,
      "grad_norm": 8.312121391296387,
      "learning_rate": 3.618421052631579e-06,
      "loss": 0.5017,
      "step": 125
    },
    {
      "Batch Mean": 1.718948245048523,
      "accuracy": 0.734375,
      "epoch": 0.3125,
      "step": 125
    },
    {
      "epoch": 0.315,
      "grad_norm": 8.334877967834473,
      "learning_rate": 3.605263157894737e-06,
      "loss": 0.4731,
      "step": 126
    },
    {
      "Batch Mean": 1.7114816904067993,
      "accuracy": 0.8359375,
      "epoch": 0.315,
      "step": 126
    },
    {
      "epoch": 0.3175,
      "grad_norm": 7.221029281616211,
      "learning_rate": 3.592105263157895e-06,
      "loss": 0.3843,
      "step": 127
    },
    {
      "Batch Mean": 1.466775894165039,
      "accuracy": 0.7578125,
      "epoch": 0.3175,
      "step": 127
    },
    {
      "epoch": 0.32,
      "grad_norm": 9.671870231628418,
      "learning_rate": 3.578947368421053e-06,
      "loss": 0.517,
      "step": 128
    },
    {
      "Batch Mean": 2.0674805641174316,
      "accuracy": 0.8125,
      "epoch": 0.32,
      "step": 128
    },
    {
      "epoch": 0.3225,
      "grad_norm": 7.535254955291748,
      "learning_rate": 3.565789473684211e-06,
      "loss": 0.4002,
      "step": 129
    },
    {
      "Batch Mean": 1.9499692916870117,
      "accuracy": 0.8046875,
      "epoch": 0.3225,
      "step": 129
    },
    {
      "epoch": 0.325,
      "grad_norm": 8.19841480255127,
      "learning_rate": 3.5526315789473687e-06,
      "loss": 0.4101,
      "step": 130
    },
    {
      "Batch Mean": 1.8141679763793945,
      "accuracy": 0.7890625,
      "epoch": 0.325,
      "step": 130
    },
    {
      "epoch": 0.3275,
      "grad_norm": 8.877778053283691,
      "learning_rate": 3.5394736842105266e-06,
      "loss": 0.458,
      "step": 131
    },
    {
      "Batch Mean": 2.349834680557251,
      "accuracy": 0.8046875,
      "epoch": 0.3275,
      "step": 131
    },
    {
      "epoch": 0.33,
      "grad_norm": 7.818430423736572,
      "learning_rate": 3.5263157894736846e-06,
      "loss": 0.3948,
      "step": 132
    },
    {
      "Batch Mean": 2.4295411109924316,
      "accuracy": 0.765625,
      "epoch": 0.33,
      "step": 132
    },
    {
      "epoch": 0.3325,
      "grad_norm": 10.082527160644531,
      "learning_rate": 3.513157894736842e-06,
      "loss": 0.4616,
      "step": 133
    },
    {
      "Batch Mean": 2.609513282775879,
      "accuracy": 0.8359375,
      "epoch": 0.3325,
      "step": 133
    },
    {
      "epoch": 0.335,
      "grad_norm": 8.784564971923828,
      "learning_rate": 3.5e-06,
      "loss": 0.3703,
      "step": 134
    },
    {
      "Batch Mean": 2.7128183841705322,
      "accuracy": 0.7890625,
      "epoch": 0.335,
      "step": 134
    },
    {
      "epoch": 0.3375,
      "grad_norm": 9.766796112060547,
      "learning_rate": 3.486842105263158e-06,
      "loss": 0.3847,
      "step": 135
    },
    {
      "Batch Mean": 3.1417746543884277,
      "accuracy": 0.7265625,
      "epoch": 0.3375,
      "step": 135
    },
    {
      "epoch": 0.34,
      "grad_norm": 11.59825325012207,
      "learning_rate": 3.473684210526316e-06,
      "loss": 0.5143,
      "step": 136
    },
    {
      "Batch Mean": 2.894944190979004,
      "accuracy": 0.7421875,
      "epoch": 0.34,
      "step": 136
    },
    {
      "epoch": 0.3425,
      "grad_norm": 12.137927055358887,
      "learning_rate": 3.460526315789474e-06,
      "loss": 0.5355,
      "step": 137
    },
    {
      "Batch Mean": 2.7786073684692383,
      "accuracy": 0.7890625,
      "epoch": 0.3425,
      "step": 137
    },
    {
      "epoch": 0.345,
      "grad_norm": 9.075066566467285,
      "learning_rate": 3.447368421052632e-06,
      "loss": 0.4115,
      "step": 138
    },
    {
      "Batch Mean": 3.0002739429473877,
      "accuracy": 0.734375,
      "epoch": 0.345,
      "step": 138
    },
    {
      "epoch": 0.3475,
      "grad_norm": 10.028230667114258,
      "learning_rate": 3.43421052631579e-06,
      "loss": 0.4722,
      "step": 139
    },
    {
      "Batch Mean": 2.4786624908447266,
      "accuracy": 0.765625,
      "epoch": 0.3475,
      "step": 139
    },
    {
      "epoch": 0.35,
      "grad_norm": 9.266149520874023,
      "learning_rate": 3.421052631578948e-06,
      "loss": 0.4656,
      "step": 140
    },
    {
      "Batch Mean": 2.1704983711242676,
      "accuracy": 0.8125,
      "epoch": 0.35,
      "step": 140
    },
    {
      "epoch": 0.3525,
      "grad_norm": 6.836582660675049,
      "learning_rate": 3.4078947368421057e-06,
      "loss": 0.3686,
      "step": 141
    },
    {
      "Batch Mean": 2.178023338317871,
      "accuracy": 0.828125,
      "epoch": 0.3525,
      "step": 141
    },
    {
      "epoch": 0.355,
      "grad_norm": 7.482564449310303,
      "learning_rate": 3.3947368421052636e-06,
      "loss": 0.4434,
      "step": 142
    },
    {
      "Batch Mean": 2.207672595977783,
      "accuracy": 0.8359375,
      "epoch": 0.355,
      "step": 142
    },
    {
      "epoch": 0.3575,
      "grad_norm": 9.649080276489258,
      "learning_rate": 3.381578947368421e-06,
      "loss": 0.4042,
      "step": 143
    },
    {
      "Batch Mean": 2.477532386779785,
      "accuracy": 0.8046875,
      "epoch": 0.3575,
      "step": 143
    },
    {
      "epoch": 0.36,
      "grad_norm": 6.769596576690674,
      "learning_rate": 3.368421052631579e-06,
      "loss": 0.3742,
      "step": 144
    },
    {
      "Batch Mean": 2.3647336959838867,
      "accuracy": 0.7578125,
      "epoch": 0.36,
      "step": 144
    },
    {
      "epoch": 0.3625,
      "grad_norm": 9.769651412963867,
      "learning_rate": 3.355263157894737e-06,
      "loss": 0.4972,
      "step": 145
    },
    {
      "Batch Mean": 2.282806396484375,
      "accuracy": 0.765625,
      "epoch": 0.3625,
      "step": 145
    },
    {
      "epoch": 0.365,
      "grad_norm": 9.153217315673828,
      "learning_rate": 3.342105263157895e-06,
      "loss": 0.4581,
      "step": 146
    },
    {
      "Batch Mean": 2.2191848754882812,
      "accuracy": 0.7421875,
      "epoch": 0.365,
      "step": 146
    },
    {
      "epoch": 0.3675,
      "grad_norm": 7.448163032531738,
      "learning_rate": 3.3289473684210528e-06,
      "loss": 0.4751,
      "step": 147
    },
    {
      "Batch Mean": 2.034508228302002,
      "accuracy": 0.7109375,
      "epoch": 0.3675,
      "step": 147
    },
    {
      "epoch": 0.37,
      "grad_norm": 8.64769458770752,
      "learning_rate": 3.3157894736842107e-06,
      "loss": 0.55,
      "step": 148
    },
    {
      "Batch Mean": 1.7862319946289062,
      "accuracy": 0.859375,
      "epoch": 0.37,
      "step": 148
    },
    {
      "epoch": 0.3725,
      "grad_norm": 6.058819770812988,
      "learning_rate": 3.302631578947369e-06,
      "loss": 0.3536,
      "step": 149
    },
    {
      "Batch Mean": 1.8278677463531494,
      "accuracy": 0.8046875,
      "epoch": 0.3725,
      "step": 149
    },
    {
      "epoch": 0.375,
      "grad_norm": 7.812012672424316,
      "learning_rate": 3.289473684210527e-06,
      "loss": 0.4398,
      "step": 150
    },
    {
      "Batch Mean": 1.9891557693481445,
      "accuracy": 0.8203125,
      "epoch": 0.375,
      "step": 150
    },
    {
      "epoch": 0.3775,
      "grad_norm": 7.331231117248535,
      "learning_rate": 3.276315789473685e-06,
      "loss": 0.4227,
      "step": 151
    },
    {
      "Batch Mean": 1.683842658996582,
      "accuracy": 0.7578125,
      "epoch": 0.3775,
      "step": 151
    },
    {
      "epoch": 0.38,
      "grad_norm": 6.605899333953857,
      "learning_rate": 3.2631578947368423e-06,
      "loss": 0.4956,
      "step": 152
    },
    {
      "Batch Mean": 1.8491768836975098,
      "accuracy": 0.8203125,
      "epoch": 0.38,
      "step": 152
    },
    {
      "epoch": 0.3825,
      "grad_norm": 6.61679744720459,
      "learning_rate": 3.2500000000000002e-06,
      "loss": 0.3815,
      "step": 153
    },
    {
      "Batch Mean": 1.5500710010528564,
      "accuracy": 0.8203125,
      "epoch": 0.3825,
      "step": 153
    },
    {
      "epoch": 0.385,
      "grad_norm": 7.707108497619629,
      "learning_rate": 3.236842105263158e-06,
      "loss": 0.4151,
      "step": 154
    },
    {
      "Batch Mean": 1.4066014289855957,
      "accuracy": 0.7890625,
      "epoch": 0.385,
      "step": 154
    },
    {
      "epoch": 0.3875,
      "grad_norm": 7.169893741607666,
      "learning_rate": 3.223684210526316e-06,
      "loss": 0.4474,
      "step": 155
    },
    {
      "Batch Mean": 1.7346997261047363,
      "accuracy": 0.7734375,
      "epoch": 0.3875,
      "step": 155
    },
    {
      "epoch": 0.39,
      "grad_norm": 10.779695510864258,
      "learning_rate": 3.210526315789474e-06,
      "loss": 0.4299,
      "step": 156
    },
    {
      "Batch Mean": 1.810314655303955,
      "accuracy": 0.796875,
      "epoch": 0.39,
      "step": 156
    },
    {
      "epoch": 0.3925,
      "grad_norm": 7.573057174682617,
      "learning_rate": 3.197368421052632e-06,
      "loss": 0.4399,
      "step": 157
    },
    {
      "Batch Mean": 1.7872644662857056,
      "accuracy": 0.7734375,
      "epoch": 0.3925,
      "step": 157
    },
    {
      "epoch": 0.395,
      "grad_norm": 7.816164493560791,
      "learning_rate": 3.1842105263157898e-06,
      "loss": 0.4131,
      "step": 158
    },
    {
      "Batch Mean": 1.804029941558838,
      "accuracy": 0.765625,
      "epoch": 0.395,
      "step": 158
    },
    {
      "epoch": 0.3975,
      "grad_norm": 8.655269622802734,
      "learning_rate": 3.1710526315789477e-06,
      "loss": 0.4737,
      "step": 159
    },
    {
      "Batch Mean": 1.565956711769104,
      "accuracy": 0.796875,
      "epoch": 0.3975,
      "step": 159
    },
    {
      "epoch": 0.4,
      "grad_norm": 7.699905872344971,
      "learning_rate": 3.157894736842105e-06,
      "loss": 0.4022,
      "step": 160
    },
    {
      "Batch Mean": 1.9926738739013672,
      "accuracy": 0.7734375,
      "epoch": 0.4,
      "step": 160
    },
    {
      "epoch": 0.4025,
      "grad_norm": 8.933215141296387,
      "learning_rate": 3.144736842105263e-06,
      "loss": 0.4674,
      "step": 161
    },
    {
      "Batch Mean": 1.7887498140335083,
      "accuracy": 0.7421875,
      "epoch": 0.4025,
      "step": 161
    },
    {
      "epoch": 0.405,
      "grad_norm": 9.933899879455566,
      "learning_rate": 3.131578947368421e-06,
      "loss": 0.459,
      "step": 162
    },
    {
      "Batch Mean": 1.8328676223754883,
      "accuracy": 0.828125,
      "epoch": 0.405,
      "step": 162
    },
    {
      "epoch": 0.4075,
      "grad_norm": 7.8370561599731445,
      "learning_rate": 3.1184210526315793e-06,
      "loss": 0.413,
      "step": 163
    },
    {
      "Batch Mean": 1.4974079132080078,
      "accuracy": 0.828125,
      "epoch": 0.4075,
      "step": 163
    },
    {
      "epoch": 0.41,
      "grad_norm": 9.378034591674805,
      "learning_rate": 3.1052631578947372e-06,
      "loss": 0.3895,
      "step": 164
    },
    {
      "Batch Mean": 1.8256797790527344,
      "accuracy": 0.75,
      "epoch": 0.41,
      "step": 164
    },
    {
      "epoch": 0.4125,
      "grad_norm": 9.905291557312012,
      "learning_rate": 3.092105263157895e-06,
      "loss": 0.4858,
      "step": 165
    },
    {
      "Batch Mean": 1.5948219299316406,
      "accuracy": 0.8046875,
      "epoch": 0.4125,
      "step": 165
    },
    {
      "epoch": 0.415,
      "grad_norm": 7.982060432434082,
      "learning_rate": 3.078947368421053e-06,
      "loss": 0.3709,
      "step": 166
    },
    {
      "Batch Mean": 1.8082324266433716,
      "accuracy": 0.8125,
      "epoch": 0.415,
      "step": 166
    },
    {
      "epoch": 0.4175,
      "grad_norm": 10.082322120666504,
      "learning_rate": 3.065789473684211e-06,
      "loss": 0.4202,
      "step": 167
    },
    {
      "Batch Mean": 1.3404994010925293,
      "accuracy": 0.8046875,
      "epoch": 0.4175,
      "step": 167
    },
    {
      "epoch": 0.42,
      "grad_norm": 10.225940704345703,
      "learning_rate": 3.052631578947369e-06,
      "loss": 0.5083,
      "step": 168
    },
    {
      "Batch Mean": 1.620276927947998,
      "accuracy": 0.796875,
      "epoch": 0.42,
      "step": 168
    },
    {
      "epoch": 0.4225,
      "grad_norm": 9.665302276611328,
      "learning_rate": 3.0394736842105268e-06,
      "loss": 0.4014,
      "step": 169
    },
    {
      "Batch Mean": 1.4625792503356934,
      "accuracy": 0.7890625,
      "epoch": 0.4225,
      "step": 169
    },
    {
      "epoch": 0.425,
      "grad_norm": 10.56104850769043,
      "learning_rate": 3.0263157894736843e-06,
      "loss": 0.5039,
      "step": 170
    },
    {
      "Batch Mean": 1.3835270404815674,
      "accuracy": 0.8671875,
      "epoch": 0.425,
      "step": 170
    },
    {
      "epoch": 0.4275,
      "grad_norm": 6.6405487060546875,
      "learning_rate": 3.013157894736842e-06,
      "loss": 0.2855,
      "step": 171
    },
    {
      "Batch Mean": 1.3700923919677734,
      "accuracy": 0.8125,
      "epoch": 0.4275,
      "step": 171
    },
    {
      "epoch": 0.43,
      "grad_norm": 7.546975612640381,
      "learning_rate": 3e-06,
      "loss": 0.3914,
      "step": 172
    },
    {
      "Batch Mean": 1.0351073741912842,
      "accuracy": 0.75,
      "epoch": 0.43,
      "step": 172
    },
    {
      "epoch": 0.4325,
      "grad_norm": 7.8717851638793945,
      "learning_rate": 2.986842105263158e-06,
      "loss": 0.4362,
      "step": 173
    },
    {
      "Batch Mean": 1.3376412391662598,
      "accuracy": 0.8359375,
      "epoch": 0.4325,
      "step": 173
    },
    {
      "epoch": 0.435,
      "grad_norm": 8.62287425994873,
      "learning_rate": 2.973684210526316e-06,
      "loss": 0.4212,
      "step": 174
    },
    {
      "Batch Mean": 1.219254493713379,
      "accuracy": 0.71875,
      "epoch": 0.435,
      "step": 174
    },
    {
      "epoch": 0.4375,
      "grad_norm": 11.215203285217285,
      "learning_rate": 2.960526315789474e-06,
      "loss": 0.5043,
      "step": 175
    },
    {
      "Batch Mean": 0.7928104400634766,
      "accuracy": 0.7578125,
      "epoch": 0.4375,
      "step": 175
    },
    {
      "epoch": 0.44,
      "grad_norm": 8.945316314697266,
      "learning_rate": 2.9473684210526317e-06,
      "loss": 0.4956,
      "step": 176
    },
    {
      "Batch Mean": 0.7014296650886536,
      "accuracy": 0.7578125,
      "epoch": 0.44,
      "step": 176
    },
    {
      "epoch": 0.4425,
      "grad_norm": 9.637972831726074,
      "learning_rate": 2.93421052631579e-06,
      "loss": 0.4785,
      "step": 177
    },
    {
      "Batch Mean": 1.184436321258545,
      "accuracy": 0.8046875,
      "epoch": 0.4425,
      "step": 177
    },
    {
      "epoch": 0.445,
      "grad_norm": 7.308509826660156,
      "learning_rate": 2.921052631578948e-06,
      "loss": 0.419,
      "step": 178
    },
    {
      "Batch Mean": 0.6965446472167969,
      "accuracy": 0.8125,
      "epoch": 0.445,
      "step": 178
    },
    {
      "epoch": 0.4475,
      "grad_norm": 8.029718399047852,
      "learning_rate": 2.907894736842106e-06,
      "loss": 0.4325,
      "step": 179
    },
    {
      "Batch Mean": 0.39546746015548706,
      "accuracy": 0.796875,
      "epoch": 0.4475,
      "step": 179
    },
    {
      "epoch": 0.45,
      "grad_norm": 7.985748291015625,
      "learning_rate": 2.8947368421052634e-06,
      "loss": 0.4038,
      "step": 180
    },
    {
      "Batch Mean": 0.5008649826049805,
      "accuracy": 0.8203125,
      "epoch": 0.45,
      "step": 180
    },
    {
      "epoch": 0.4525,
      "grad_norm": 7.97329044342041,
      "learning_rate": 2.8815789473684213e-06,
      "loss": 0.3997,
      "step": 181
    },
    {
      "Batch Mean": 0.5764102935791016,
      "accuracy": 0.8125,
      "epoch": 0.4525,
      "step": 181
    },
    {
      "epoch": 0.455,
      "grad_norm": 8.872234344482422,
      "learning_rate": 2.868421052631579e-06,
      "loss": 0.454,
      "step": 182
    },
    {
      "Batch Mean": 1.0936000347137451,
      "accuracy": 0.828125,
      "epoch": 0.455,
      "step": 182
    },
    {
      "epoch": 0.4575,
      "grad_norm": 8.255775451660156,
      "learning_rate": 2.855263157894737e-06,
      "loss": 0.4011,
      "step": 183
    },
    {
      "Batch Mean": 1.0273089408874512,
      "accuracy": 0.78125,
      "epoch": 0.4575,
      "step": 183
    },
    {
      "epoch": 0.46,
      "grad_norm": 9.933390617370605,
      "learning_rate": 2.842105263157895e-06,
      "loss": 0.4856,
      "step": 184
    },
    {
      "Batch Mean": 1.2797164916992188,
      "accuracy": 0.8359375,
      "epoch": 0.46,
      "step": 184
    },
    {
      "epoch": 0.4625,
      "grad_norm": 8.756768226623535,
      "learning_rate": 2.828947368421053e-06,
      "loss": 0.3929,
      "step": 185
    },
    {
      "Batch Mean": 1.3405977487564087,
      "accuracy": 0.8046875,
      "epoch": 0.4625,
      "step": 185
    },
    {
      "epoch": 0.465,
      "grad_norm": 8.689950942993164,
      "learning_rate": 2.815789473684211e-06,
      "loss": 0.4513,
      "step": 186
    },
    {
      "Batch Mean": 1.369067668914795,
      "accuracy": 0.7890625,
      "epoch": 0.465,
      "step": 186
    },
    {
      "epoch": 0.4675,
      "grad_norm": 8.023719787597656,
      "learning_rate": 2.8026315789473683e-06,
      "loss": 0.4026,
      "step": 187
    },
    {
      "Batch Mean": 2.006643295288086,
      "accuracy": 0.7578125,
      "epoch": 0.4675,
      "step": 187
    },
    {
      "epoch": 0.47,
      "grad_norm": 10.34017276763916,
      "learning_rate": 2.789473684210526e-06,
      "loss": 0.4653,
      "step": 188
    },
    {
      "Batch Mean": 2.0173683166503906,
      "accuracy": 0.8203125,
      "epoch": 0.47,
      "step": 188
    },
    {
      "epoch": 0.4725,
      "grad_norm": 9.120015144348145,
      "learning_rate": 2.776315789473684e-06,
      "loss": 0.4107,
      "step": 189
    },
    {
      "Batch Mean": 1.916172981262207,
      "accuracy": 0.8515625,
      "epoch": 0.4725,
      "step": 189
    },
    {
      "epoch": 0.475,
      "grad_norm": 7.870621681213379,
      "learning_rate": 2.7631578947368424e-06,
      "loss": 0.4053,
      "step": 190
    },
    {
      "Batch Mean": 2.2058982849121094,
      "accuracy": 0.765625,
      "epoch": 0.475,
      "step": 190
    },
    {
      "epoch": 0.4775,
      "grad_norm": 11.424184799194336,
      "learning_rate": 2.7500000000000004e-06,
      "loss": 0.512,
      "step": 191
    },
    {
      "Batch Mean": 2.4413208961486816,
      "accuracy": 0.84375,
      "epoch": 0.4775,
      "step": 191
    },
    {
      "epoch": 0.48,
      "grad_norm": 7.611173152923584,
      "learning_rate": 2.7368421052631583e-06,
      "loss": 0.3401,
      "step": 192
    },
    {
      "Batch Mean": 2.9837169647216797,
      "accuracy": 0.8125,
      "epoch": 0.48,
      "step": 192
    },
    {
      "epoch": 0.4825,
      "grad_norm": 7.9044413566589355,
      "learning_rate": 2.723684210526316e-06,
      "loss": 0.3885,
      "step": 193
    },
    {
      "Batch Mean": 2.2724387645721436,
      "accuracy": 0.8046875,
      "epoch": 0.4825,
      "step": 193
    },
    {
      "epoch": 0.485,
      "grad_norm": 7.795435905456543,
      "learning_rate": 2.710526315789474e-06,
      "loss": 0.4036,
      "step": 194
    },
    {
      "Batch Mean": 2.2937026023864746,
      "accuracy": 0.71875,
      "epoch": 0.485,
      "step": 194
    },
    {
      "epoch": 0.4875,
      "grad_norm": 8.691864013671875,
      "learning_rate": 2.697368421052632e-06,
      "loss": 0.4933,
      "step": 195
    },
    {
      "Batch Mean": 2.3943445682525635,
      "accuracy": 0.8359375,
      "epoch": 0.4875,
      "step": 195
    },
    {
      "epoch": 0.49,
      "grad_norm": 7.490312576293945,
      "learning_rate": 2.68421052631579e-06,
      "loss": 0.4102,
      "step": 196
    },
    {
      "Batch Mean": 1.93037748336792,
      "accuracy": 0.7578125,
      "epoch": 0.49,
      "step": 196
    },
    {
      "epoch": 0.4925,
      "grad_norm": 7.2265729904174805,
      "learning_rate": 2.6710526315789474e-06,
      "loss": 0.4356,
      "step": 197
    },
    {
      "Batch Mean": 2.346465587615967,
      "accuracy": 0.7578125,
      "epoch": 0.4925,
      "step": 197
    },
    {
      "epoch": 0.495,
      "grad_norm": 8.016980171203613,
      "learning_rate": 2.6578947368421053e-06,
      "loss": 0.4563,
      "step": 198
    },
    {
      "Batch Mean": 2.213311195373535,
      "accuracy": 0.7421875,
      "epoch": 0.495,
      "step": 198
    },
    {
      "epoch": 0.4975,
      "grad_norm": 7.577280521392822,
      "learning_rate": 2.644736842105263e-06,
      "loss": 0.4736,
      "step": 199
    },
    {
      "Batch Mean": 1.7689447402954102,
      "accuracy": 0.75,
      "epoch": 0.4975,
      "step": 199
    },
    {
      "epoch": 0.5,
      "grad_norm": 6.788050651550293,
      "learning_rate": 2.631578947368421e-06,
      "loss": 0.4655,
      "step": 200
    },
    {
      "Batch Mean": 1.7899178266525269,
      "accuracy": 0.8203125,
      "epoch": 0.5,
      "step": 200
    },
    {
      "epoch": 0.5025,
      "grad_norm": 6.991423606872559,
      "learning_rate": 2.618421052631579e-06,
      "loss": 0.4277,
      "step": 201
    },
    {
      "Batch Mean": 1.6013970375061035,
      "accuracy": 0.7734375,
      "epoch": 0.5025,
      "step": 201
    },
    {
      "epoch": 0.505,
      "grad_norm": 6.626009464263916,
      "learning_rate": 2.605263157894737e-06,
      "loss": 0.4654,
      "step": 202
    },
    {
      "Batch Mean": 1.734081745147705,
      "accuracy": 0.828125,
      "epoch": 0.505,
      "step": 202
    },
    {
      "epoch": 0.5075,
      "grad_norm": 6.0885725021362305,
      "learning_rate": 2.592105263157895e-06,
      "loss": 0.3811,
      "step": 203
    },
    {
      "Batch Mean": 1.5769119262695312,
      "accuracy": 0.765625,
      "epoch": 0.5075,
      "step": 203
    },
    {
      "epoch": 0.51,
      "grad_norm": 7.119544982910156,
      "learning_rate": 2.578947368421053e-06,
      "loss": 0.4682,
      "step": 204
    },
    {
      "Batch Mean": 1.2697408199310303,
      "accuracy": 0.7890625,
      "epoch": 0.51,
      "step": 204
    },
    {
      "epoch": 0.5125,
      "grad_norm": 7.171177387237549,
      "learning_rate": 2.565789473684211e-06,
      "loss": 0.4633,
      "step": 205
    },
    {
      "Batch Mean": 1.5697135925292969,
      "accuracy": 0.8203125,
      "epoch": 0.5125,
      "step": 205
    },
    {
      "epoch": 0.515,
      "grad_norm": 7.186983585357666,
      "learning_rate": 2.552631578947369e-06,
      "loss": 0.4177,
      "step": 206
    },
    {
      "Batch Mean": 1.5726741552352905,
      "accuracy": 0.78125,
      "epoch": 0.515,
      "step": 206
    },
    {
      "epoch": 0.5175,
      "grad_norm": 7.522491931915283,
      "learning_rate": 2.5394736842105265e-06,
      "loss": 0.4629,
      "step": 207
    },
    {
      "Batch Mean": 1.6065094470977783,
      "accuracy": 0.828125,
      "epoch": 0.5175,
      "step": 207
    },
    {
      "epoch": 0.52,
      "grad_norm": 7.612408638000488,
      "learning_rate": 2.5263157894736844e-06,
      "loss": 0.3536,
      "step": 208
    },
    {
      "Batch Mean": 2.1261379718780518,
      "accuracy": 0.8125,
      "epoch": 0.52,
      "step": 208
    },
    {
      "epoch": 0.5225,
      "grad_norm": 7.919214725494385,
      "learning_rate": 2.5131578947368423e-06,
      "loss": 0.4265,
      "step": 209
    },
    {
      "Batch Mean": 2.403383493423462,
      "accuracy": 0.7578125,
      "epoch": 0.5225,
      "step": 209
    },
    {
      "epoch": 0.525,
      "grad_norm": 8.452061653137207,
      "learning_rate": 2.5e-06,
      "loss": 0.4476,
      "step": 210
    },
    {
      "Batch Mean": 2.177506446838379,
      "accuracy": 0.7734375,
      "epoch": 0.525,
      "step": 210
    },
    {
      "epoch": 0.5275,
      "grad_norm": 10.394681930541992,
      "learning_rate": 2.486842105263158e-06,
      "loss": 0.5057,
      "step": 211
    },
    {
      "Batch Mean": 2.585557222366333,
      "accuracy": 0.8125,
      "epoch": 0.5275,
      "step": 211
    },
    {
      "epoch": 0.53,
      "grad_norm": 8.910456657409668,
      "learning_rate": 2.473684210526316e-06,
      "loss": 0.4257,
      "step": 212
    },
    {
      "Batch Mean": 2.618654251098633,
      "accuracy": 0.7890625,
      "epoch": 0.53,
      "step": 212
    },
    {
      "epoch": 0.5325,
      "grad_norm": 8.461559295654297,
      "learning_rate": 2.460526315789474e-06,
      "loss": 0.4123,
      "step": 213
    },
    {
      "Batch Mean": 2.872316360473633,
      "accuracy": 0.734375,
      "epoch": 0.5325,
      "step": 213
    },
    {
      "epoch": 0.535,
      "grad_norm": 12.725778579711914,
      "learning_rate": 2.447368421052632e-06,
      "loss": 0.5599,
      "step": 214
    },
    {
      "Batch Mean": 3.0026845932006836,
      "accuracy": 0.8515625,
      "epoch": 0.535,
      "step": 214
    },
    {
      "epoch": 0.5375,
      "grad_norm": 9.596014976501465,
      "learning_rate": 2.4342105263157898e-06,
      "loss": 0.4411,
      "step": 215
    },
    {
      "Batch Mean": 2.897749662399292,
      "accuracy": 0.78125,
      "epoch": 0.5375,
      "step": 215
    },
    {
      "epoch": 0.54,
      "grad_norm": 8.899314880371094,
      "learning_rate": 2.4210526315789477e-06,
      "loss": 0.3853,
      "step": 216
    },
    {
      "Batch Mean": 2.9906558990478516,
      "accuracy": 0.765625,
      "epoch": 0.54,
      "step": 216
    },
    {
      "epoch": 0.5425,
      "grad_norm": 10.041382789611816,
      "learning_rate": 2.4078947368421056e-06,
      "loss": 0.4807,
      "step": 217
    },
    {
      "Batch Mean": 3.049136161804199,
      "accuracy": 0.8359375,
      "epoch": 0.5425,
      "step": 217
    },
    {
      "epoch": 0.545,
      "grad_norm": 9.229241371154785,
      "learning_rate": 2.3947368421052635e-06,
      "loss": 0.3642,
      "step": 218
    },
    {
      "Batch Mean": 2.9170212745666504,
      "accuracy": 0.8046875,
      "epoch": 0.545,
      "step": 218
    },
    {
      "epoch": 0.5475,
      "grad_norm": 9.190288543701172,
      "learning_rate": 2.381578947368421e-06,
      "loss": 0.3924,
      "step": 219
    },
    {
      "Batch Mean": 2.6925058364868164,
      "accuracy": 0.7890625,
      "epoch": 0.5475,
      "step": 219
    },
    {
      "epoch": 0.55,
      "grad_norm": 10.271293640136719,
      "learning_rate": 2.368421052631579e-06,
      "loss": 0.4699,
      "step": 220
    },
    {
      "Batch Mean": 2.789365530014038,
      "accuracy": 0.78125,
      "epoch": 0.55,
      "step": 220
    },
    {
      "epoch": 0.5525,
      "grad_norm": 9.16843032836914,
      "learning_rate": 2.355263157894737e-06,
      "loss": 0.3856,
      "step": 221
    },
    {
      "Batch Mean": 2.4267354011535645,
      "accuracy": 0.8125,
      "epoch": 0.5525,
      "step": 221
    },
    {
      "epoch": 0.555,
      "grad_norm": 9.069971084594727,
      "learning_rate": 2.342105263157895e-06,
      "loss": 0.4526,
      "step": 222
    },
    {
      "Batch Mean": 2.2059507369995117,
      "accuracy": 0.8046875,
      "epoch": 0.555,
      "step": 222
    },
    {
      "epoch": 0.5575,
      "grad_norm": 9.028841018676758,
      "learning_rate": 2.328947368421053e-06,
      "loss": 0.41,
      "step": 223
    },
    {
      "Batch Mean": 1.7158894538879395,
      "accuracy": 0.8125,
      "epoch": 0.5575,
      "step": 223
    },
    {
      "epoch": 0.56,
      "grad_norm": 8.238944053649902,
      "learning_rate": 2.3157894736842105e-06,
      "loss": 0.4087,
      "step": 224
    },
    {
      "Batch Mean": 1.857762336730957,
      "accuracy": 0.796875,
      "epoch": 0.56,
      "step": 224
    },
    {
      "epoch": 0.5625,
      "grad_norm": 7.938877105712891,
      "learning_rate": 2.3026315789473684e-06,
      "loss": 0.418,
      "step": 225
    },
    {
      "Batch Mean": 2.1304166316986084,
      "accuracy": 0.734375,
      "epoch": 0.5625,
      "step": 225
    },
    {
      "epoch": 0.565,
      "grad_norm": 10.606386184692383,
      "learning_rate": 2.2894736842105263e-06,
      "loss": 0.4973,
      "step": 226
    },
    {
      "Batch Mean": 2.0900015830993652,
      "accuracy": 0.734375,
      "epoch": 0.565,
      "step": 226
    },
    {
      "epoch": 0.5675,
      "grad_norm": 9.365285873413086,
      "learning_rate": 2.2763157894736847e-06,
      "loss": 0.49,
      "step": 227
    },
    {
      "Batch Mean": 1.7396864891052246,
      "accuracy": 0.7890625,
      "epoch": 0.5675,
      "step": 227
    },
    {
      "epoch": 0.57,
      "grad_norm": 9.352794647216797,
      "learning_rate": 2.2631578947368426e-06,
      "loss": 0.4665,
      "step": 228
    },
    {
      "Batch Mean": 1.4743752479553223,
      "accuracy": 0.8046875,
      "epoch": 0.57,
      "step": 228
    },
    {
      "epoch": 0.5725,
      "grad_norm": 8.720419883728027,
      "learning_rate": 2.25e-06,
      "loss": 0.433,
      "step": 229
    },
    {
      "Batch Mean": 1.5897538661956787,
      "accuracy": 0.8046875,
      "epoch": 0.5725,
      "step": 229
    },
    {
      "epoch": 0.575,
      "grad_norm": 8.863330841064453,
      "learning_rate": 2.236842105263158e-06,
      "loss": 0.4247,
      "step": 230
    },
    {
      "Batch Mean": 1.2378873825073242,
      "accuracy": 0.7578125,
      "epoch": 0.575,
      "step": 230
    },
    {
      "epoch": 0.5775,
      "grad_norm": 8.975724220275879,
      "learning_rate": 2.223684210526316e-06,
      "loss": 0.4741,
      "step": 231
    },
    {
      "Batch Mean": 1.3916802406311035,
      "accuracy": 0.7578125,
      "epoch": 0.5775,
      "step": 231
    },
    {
      "epoch": 0.58,
      "grad_norm": 8.523025512695312,
      "learning_rate": 2.2105263157894738e-06,
      "loss": 0.4634,
      "step": 232
    },
    {
      "Batch Mean": 1.4983201026916504,
      "accuracy": 0.7421875,
      "epoch": 0.58,
      "step": 232
    },
    {
      "epoch": 0.5825,
      "grad_norm": 9.57751178741455,
      "learning_rate": 2.1973684210526317e-06,
      "loss": 0.5449,
      "step": 233
    },
    {
      "Batch Mean": 1.2628620862960815,
      "accuracy": 0.765625,
      "epoch": 0.5825,
      "step": 233
    },
    {
      "epoch": 0.585,
      "grad_norm": 8.061960220336914,
      "learning_rate": 2.1842105263157896e-06,
      "loss": 0.4441,
      "step": 234
    },
    {
      "Batch Mean": 1.1907455921173096,
      "accuracy": 0.8515625,
      "epoch": 0.585,
      "step": 234
    },
    {
      "epoch": 0.5875,
      "grad_norm": 6.615776538848877,
      "learning_rate": 2.1710526315789475e-06,
      "loss": 0.3421,
      "step": 235
    },
    {
      "Batch Mean": 1.1369013786315918,
      "accuracy": 0.8515625,
      "epoch": 0.5875,
      "step": 235
    },
    {
      "epoch": 0.59,
      "grad_norm": 7.37549352645874,
      "learning_rate": 2.1578947368421054e-06,
      "loss": 0.3825,
      "step": 236
    },
    {
      "Batch Mean": 1.1984944343566895,
      "accuracy": 0.7890625,
      "epoch": 0.59,
      "step": 236
    },
    {
      "epoch": 0.5925,
      "grad_norm": 7.898306846618652,
      "learning_rate": 2.1447368421052633e-06,
      "loss": 0.4196,
      "step": 237
    },
    {
      "Batch Mean": 0.9305081367492676,
      "accuracy": 0.796875,
      "epoch": 0.5925,
      "step": 237
    },
    {
      "epoch": 0.595,
      "grad_norm": 7.740181922912598,
      "learning_rate": 2.1315789473684212e-06,
      "loss": 0.4401,
      "step": 238
    },
    {
      "Batch Mean": 1.1609792709350586,
      "accuracy": 0.7578125,
      "epoch": 0.595,
      "step": 238
    },
    {
      "epoch": 0.5975,
      "grad_norm": 7.328685283660889,
      "learning_rate": 2.118421052631579e-06,
      "loss": 0.3894,
      "step": 239
    },
    {
      "Batch Mean": 1.2955622673034668,
      "accuracy": 0.7890625,
      "epoch": 0.5975,
      "step": 239
    },
    {
      "epoch": 0.6,
      "grad_norm": 7.802572250366211,
      "learning_rate": 2.105263157894737e-06,
      "loss": 0.3676,
      "step": 240
    },
    {
      "Batch Mean": 1.6187195777893066,
      "accuracy": 0.8046875,
      "epoch": 0.6,
      "step": 240
    },
    {
      "epoch": 0.6025,
      "grad_norm": 9.184107780456543,
      "learning_rate": 2.092105263157895e-06,
      "loss": 0.4378,
      "step": 241
    },
    {
      "Batch Mean": 1.7035493850708008,
      "accuracy": 0.765625,
      "epoch": 0.6025,
      "step": 241
    },
    {
      "epoch": 0.605,
      "grad_norm": 8.614889144897461,
      "learning_rate": 2.078947368421053e-06,
      "loss": 0.4155,
      "step": 242
    },
    {
      "Batch Mean": 1.7622737884521484,
      "accuracy": 0.78125,
      "epoch": 0.605,
      "step": 242
    },
    {
      "epoch": 0.6075,
      "grad_norm": 9.022790908813477,
      "learning_rate": 2.0657894736842108e-06,
      "loss": 0.435,
      "step": 243
    },
    {
      "Batch Mean": 1.6522586345672607,
      "accuracy": 0.7578125,
      "epoch": 0.6075,
      "step": 243
    },
    {
      "epoch": 0.61,
      "grad_norm": 9.872020721435547,
      "learning_rate": 2.0526315789473687e-06,
      "loss": 0.4509,
      "step": 244
    },
    {
      "Batch Mean": 1.6742737293243408,
      "accuracy": 0.796875,
      "epoch": 0.61,
      "step": 244
    },
    {
      "epoch": 0.6125,
      "grad_norm": 9.214981079101562,
      "learning_rate": 2.0394736842105266e-06,
      "loss": 0.4094,
      "step": 245
    },
    {
      "Batch Mean": 1.6322050094604492,
      "accuracy": 0.7890625,
      "epoch": 0.6125,
      "step": 245
    },
    {
      "epoch": 0.615,
      "grad_norm": 9.1509428024292,
      "learning_rate": 2.026315789473684e-06,
      "loss": 0.4213,
      "step": 246
    },
    {
      "Batch Mean": 1.2047111988067627,
      "accuracy": 0.8046875,
      "epoch": 0.615,
      "step": 246
    },
    {
      "epoch": 0.6175,
      "grad_norm": 10.241280555725098,
      "learning_rate": 2.013157894736842e-06,
      "loss": 0.4428,
      "step": 247
    },
    {
      "Batch Mean": 1.6580332517623901,
      "accuracy": 0.78125,
      "epoch": 0.6175,
      "step": 247
    },
    {
      "epoch": 0.62,
      "grad_norm": 9.183330535888672,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 0.4128,
      "step": 248
    },
    {
      "Batch Mean": 1.211564064025879,
      "accuracy": 0.859375,
      "epoch": 0.62,
      "step": 248
    },
    {
      "epoch": 0.6225,
      "grad_norm": 7.7444915771484375,
      "learning_rate": 1.9868421052631582e-06,
      "loss": 0.3432,
      "step": 249
    },
    {
      "Batch Mean": 1.4931328296661377,
      "accuracy": 0.734375,
      "epoch": 0.6225,
      "step": 249
    },
    {
      "epoch": 0.625,
      "grad_norm": 9.798284530639648,
      "learning_rate": 1.973684210526316e-06,
      "loss": 0.467,
      "step": 250
    },
    {
      "Batch Mean": 1.1091797351837158,
      "accuracy": 0.8125,
      "epoch": 0.625,
      "step": 250
    },
    {
      "epoch": 0.6275,
      "grad_norm": 8.74545955657959,
      "learning_rate": 1.9605263157894736e-06,
      "loss": 0.4159,
      "step": 251
    },
    {
      "Batch Mean": 1.0245903730392456,
      "accuracy": 0.8046875,
      "epoch": 0.6275,
      "step": 251
    },
    {
      "epoch": 0.63,
      "grad_norm": 8.791230201721191,
      "learning_rate": 1.9473684210526315e-06,
      "loss": 0.4036,
      "step": 252
    },
    {
      "Batch Mean": 1.796776294708252,
      "accuracy": 0.828125,
      "epoch": 0.63,
      "step": 252
    },
    {
      "epoch": 0.6325,
      "grad_norm": 9.18313980102539,
      "learning_rate": 1.9342105263157895e-06,
      "loss": 0.4212,
      "step": 253
    },
    {
      "Batch Mean": 1.1153364181518555,
      "accuracy": 0.796875,
      "epoch": 0.6325,
      "step": 253
    },
    {
      "epoch": 0.635,
      "grad_norm": 9.925728797912598,
      "learning_rate": 1.9210526315789474e-06,
      "loss": 0.4159,
      "step": 254
    },
    {
      "Batch Mean": 1.2809149026870728,
      "accuracy": 0.796875,
      "epoch": 0.635,
      "step": 254
    },
    {
      "epoch": 0.6375,
      "grad_norm": 8.616195678710938,
      "learning_rate": 1.9078947368421057e-06,
      "loss": 0.4015,
      "step": 255
    },
    {
      "Batch Mean": 1.6480159759521484,
      "accuracy": 0.765625,
      "epoch": 0.6375,
      "step": 255
    },
    {
      "epoch": 0.64,
      "grad_norm": 10.048429489135742,
      "learning_rate": 1.8947368421052634e-06,
      "loss": 0.4418,
      "step": 256
    },
    {
      "Batch Mean": 1.6273455619812012,
      "accuracy": 0.7265625,
      "epoch": 0.64,
      "step": 256
    },
    {
      "epoch": 0.6425,
      "grad_norm": 9.716386795043945,
      "learning_rate": 1.8815789473684213e-06,
      "loss": 0.4678,
      "step": 257
    },
    {
      "Batch Mean": 1.7655820846557617,
      "accuracy": 0.7734375,
      "epoch": 0.6425,
      "step": 257
    },
    {
      "epoch": 0.645,
      "grad_norm": 8.57561206817627,
      "learning_rate": 1.868421052631579e-06,
      "loss": 0.4057,
      "step": 258
    },
    {
      "Batch Mean": 2.1368942260742188,
      "accuracy": 0.828125,
      "epoch": 0.645,
      "step": 258
    },
    {
      "epoch": 0.6475,
      "grad_norm": 9.588972091674805,
      "learning_rate": 1.855263157894737e-06,
      "loss": 0.3979,
      "step": 259
    },
    {
      "Batch Mean": 2.052062511444092,
      "accuracy": 0.8125,
      "epoch": 0.6475,
      "step": 259
    },
    {
      "epoch": 0.65,
      "grad_norm": 9.272086143493652,
      "learning_rate": 1.8421052631578948e-06,
      "loss": 0.4115,
      "step": 260
    },
    {
      "Batch Mean": 1.610414981842041,
      "accuracy": 0.796875,
      "epoch": 0.65,
      "step": 260
    },
    {
      "epoch": 0.6525,
      "grad_norm": 8.309432029724121,
      "learning_rate": 1.828947368421053e-06,
      "loss": 0.3825,
      "step": 261
    },
    {
      "Batch Mean": 1.932809829711914,
      "accuracy": 0.7578125,
      "epoch": 0.6525,
      "step": 261
    },
    {
      "epoch": 0.655,
      "grad_norm": 9.389504432678223,
      "learning_rate": 1.8157894736842109e-06,
      "loss": 0.4142,
      "step": 262
    },
    {
      "Batch Mean": 1.9246654510498047,
      "accuracy": 0.7734375,
      "epoch": 0.655,
      "step": 262
    },
    {
      "epoch": 0.6575,
      "grad_norm": 9.920475006103516,
      "learning_rate": 1.8026315789473685e-06,
      "loss": 0.4758,
      "step": 263
    },
    {
      "Batch Mean": 1.93519926071167,
      "accuracy": 0.796875,
      "epoch": 0.6575,
      "step": 263
    },
    {
      "epoch": 0.66,
      "grad_norm": 9.552947044372559,
      "learning_rate": 1.7894736842105265e-06,
      "loss": 0.4303,
      "step": 264
    },
    {
      "Batch Mean": 2.076547861099243,
      "accuracy": 0.8203125,
      "epoch": 0.66,
      "step": 264
    },
    {
      "epoch": 0.6625,
      "grad_norm": 9.472382545471191,
      "learning_rate": 1.7763157894736844e-06,
      "loss": 0.3981,
      "step": 265
    },
    {
      "Batch Mean": 1.9459846019744873,
      "accuracy": 0.90625,
      "epoch": 0.6625,
      "step": 265
    },
    {
      "epoch": 0.665,
      "grad_norm": 7.733583450317383,
      "learning_rate": 1.7631578947368423e-06,
      "loss": 0.3092,
      "step": 266
    },
    {
      "Batch Mean": 1.9703218936920166,
      "accuracy": 0.8515625,
      "epoch": 0.665,
      "step": 266
    },
    {
      "epoch": 0.6675,
      "grad_norm": 9.029424667358398,
      "learning_rate": 1.75e-06,
      "loss": 0.3787,
      "step": 267
    },
    {
      "Batch Mean": 2.1942596435546875,
      "accuracy": 0.8046875,
      "epoch": 0.6675,
      "step": 267
    },
    {
      "epoch": 0.67,
      "grad_norm": 9.263646125793457,
      "learning_rate": 1.736842105263158e-06,
      "loss": 0.4368,
      "step": 268
    },
    {
      "Batch Mean": 2.0240073204040527,
      "accuracy": 0.828125,
      "epoch": 0.67,
      "step": 268
    },
    {
      "epoch": 0.6725,
      "grad_norm": 8.806449890136719,
      "learning_rate": 1.723684210526316e-06,
      "loss": 0.379,
      "step": 269
    },
    {
      "Batch Mean": 2.138075828552246,
      "accuracy": 0.7734375,
      "epoch": 0.6725,
      "step": 269
    },
    {
      "epoch": 0.675,
      "grad_norm": 10.799676895141602,
      "learning_rate": 1.710526315789474e-06,
      "loss": 0.4921,
      "step": 270
    },
    {
      "Batch Mean": 1.9924249649047852,
      "accuracy": 0.8203125,
      "epoch": 0.675,
      "step": 270
    },
    {
      "epoch": 0.6775,
      "grad_norm": 10.256522178649902,
      "learning_rate": 1.6973684210526318e-06,
      "loss": 0.4736,
      "step": 271
    },
    {
      "Batch Mean": 2.115017890930176,
      "accuracy": 0.796875,
      "epoch": 0.6775,
      "step": 271
    },
    {
      "epoch": 0.68,
      "grad_norm": 8.95787239074707,
      "learning_rate": 1.6842105263157895e-06,
      "loss": 0.3887,
      "step": 272
    },
    {
      "Batch Mean": 2.272857666015625,
      "accuracy": 0.765625,
      "epoch": 0.68,
      "step": 272
    },
    {
      "epoch": 0.6825,
      "grad_norm": 10.41603946685791,
      "learning_rate": 1.6710526315789474e-06,
      "loss": 0.452,
      "step": 273
    },
    {
      "Batch Mean": 2.3034958839416504,
      "accuracy": 0.8125,
      "epoch": 0.6825,
      "step": 273
    },
    {
      "epoch": 0.685,
      "grad_norm": 9.792698860168457,
      "learning_rate": 1.6578947368421053e-06,
      "loss": 0.4109,
      "step": 274
    },
    {
      "Batch Mean": 2.2534396648406982,
      "accuracy": 0.734375,
      "epoch": 0.685,
      "step": 274
    },
    {
      "epoch": 0.6875,
      "grad_norm": 10.466201782226562,
      "learning_rate": 1.6447368421052635e-06,
      "loss": 0.52,
      "step": 275
    },
    {
      "Batch Mean": 2.1817493438720703,
      "accuracy": 0.78125,
      "epoch": 0.6875,
      "step": 275
    },
    {
      "epoch": 0.69,
      "grad_norm": 8.804483413696289,
      "learning_rate": 1.6315789473684212e-06,
      "loss": 0.4599,
      "step": 276
    },
    {
      "Batch Mean": 2.727586507797241,
      "accuracy": 0.84375,
      "epoch": 0.69,
      "step": 276
    },
    {
      "epoch": 0.6925,
      "grad_norm": 9.147492408752441,
      "learning_rate": 1.618421052631579e-06,
      "loss": 0.4186,
      "step": 277
    },
    {
      "Batch Mean": 2.453937530517578,
      "accuracy": 0.828125,
      "epoch": 0.6925,
      "step": 277
    },
    {
      "epoch": 0.695,
      "grad_norm": 8.768532752990723,
      "learning_rate": 1.605263157894737e-06,
      "loss": 0.4019,
      "step": 278
    },
    {
      "Batch Mean": 2.7397398948669434,
      "accuracy": 0.7734375,
      "epoch": 0.695,
      "step": 278
    },
    {
      "epoch": 0.6975,
      "grad_norm": 8.36785888671875,
      "learning_rate": 1.5921052631578949e-06,
      "loss": 0.4471,
      "step": 279
    },
    {
      "Batch Mean": 2.2289133071899414,
      "accuracy": 0.75,
      "epoch": 0.6975,
      "step": 279
    },
    {
      "epoch": 0.7,
      "grad_norm": 12.092652320861816,
      "learning_rate": 1.5789473684210526e-06,
      "loss": 0.5102,
      "step": 280
    },
    {
      "Batch Mean": 2.356043577194214,
      "accuracy": 0.8359375,
      "epoch": 0.7,
      "step": 280
    },
    {
      "epoch": 0.7025,
      "grad_norm": 8.004008293151855,
      "learning_rate": 1.5657894736842105e-06,
      "loss": 0.3634,
      "step": 281
    },
    {
      "Batch Mean": 2.659639596939087,
      "accuracy": 0.7734375,
      "epoch": 0.7025,
      "step": 281
    },
    {
      "epoch": 0.705,
      "grad_norm": 8.374983787536621,
      "learning_rate": 1.5526315789473686e-06,
      "loss": 0.4304,
      "step": 282
    },
    {
      "Batch Mean": 2.38856840133667,
      "accuracy": 0.7421875,
      "epoch": 0.705,
      "step": 282
    },
    {
      "epoch": 0.7075,
      "grad_norm": 9.618534088134766,
      "learning_rate": 1.5394736842105265e-06,
      "loss": 0.4379,
      "step": 283
    },
    {
      "Batch Mean": 2.196455955505371,
      "accuracy": 0.796875,
      "epoch": 0.7075,
      "step": 283
    },
    {
      "epoch": 0.71,
      "grad_norm": 8.226493835449219,
      "learning_rate": 1.5263157894736844e-06,
      "loss": 0.4122,
      "step": 284
    },
    {
      "Batch Mean": 2.7598533630371094,
      "accuracy": 0.765625,
      "epoch": 0.71,
      "step": 284
    },
    {
      "epoch": 0.7125,
      "grad_norm": 8.447372436523438,
      "learning_rate": 1.5131578947368421e-06,
      "loss": 0.451,
      "step": 285
    },
    {
      "Batch Mean": 2.280998468399048,
      "accuracy": 0.75,
      "epoch": 0.7125,
      "step": 285
    },
    {
      "epoch": 0.715,
      "grad_norm": 9.170207977294922,
      "learning_rate": 1.5e-06,
      "loss": 0.4512,
      "step": 286
    },
    {
      "Batch Mean": 2.208192825317383,
      "accuracy": 0.8203125,
      "epoch": 0.715,
      "step": 286
    },
    {
      "epoch": 0.7175,
      "grad_norm": 8.343831062316895,
      "learning_rate": 1.486842105263158e-06,
      "loss": 0.3791,
      "step": 287
    },
    {
      "Batch Mean": 2.5024871826171875,
      "accuracy": 0.8515625,
      "epoch": 0.7175,
      "step": 287
    },
    {
      "epoch": 0.72,
      "grad_norm": 7.829848289489746,
      "learning_rate": 1.4736842105263159e-06,
      "loss": 0.3538,
      "step": 288
    },
    {
      "Batch Mean": 2.4838662147521973,
      "accuracy": 0.7734375,
      "epoch": 0.72,
      "step": 288
    },
    {
      "epoch": 0.7225,
      "grad_norm": 8.380528450012207,
      "learning_rate": 1.460526315789474e-06,
      "loss": 0.4281,
      "step": 289
    },
    {
      "Batch Mean": 2.4715757369995117,
      "accuracy": 0.8046875,
      "epoch": 0.7225,
      "step": 289
    },
    {
      "epoch": 0.725,
      "grad_norm": 9.04012680053711,
      "learning_rate": 1.4473684210526317e-06,
      "loss": 0.4141,
      "step": 290
    },
    {
      "Batch Mean": 2.498483657836914,
      "accuracy": 0.84375,
      "epoch": 0.725,
      "step": 290
    },
    {
      "epoch": 0.7275,
      "grad_norm": 7.590081691741943,
      "learning_rate": 1.4342105263157896e-06,
      "loss": 0.33,
      "step": 291
    },
    {
      "Batch Mean": 2.728640556335449,
      "accuracy": 0.875,
      "epoch": 0.7275,
      "step": 291
    },
    {
      "epoch": 0.73,
      "grad_norm": 7.328402996063232,
      "learning_rate": 1.4210526315789475e-06,
      "loss": 0.3089,
      "step": 292
    },
    {
      "Batch Mean": 2.8817148208618164,
      "accuracy": 0.796875,
      "epoch": 0.73,
      "step": 292
    },
    {
      "epoch": 0.7325,
      "grad_norm": 9.085417747497559,
      "learning_rate": 1.4078947368421054e-06,
      "loss": 0.438,
      "step": 293
    },
    {
      "Batch Mean": 3.2919604778289795,
      "accuracy": 0.7890625,
      "epoch": 0.7325,
      "step": 293
    },
    {
      "epoch": 0.735,
      "grad_norm": 9.366758346557617,
      "learning_rate": 1.394736842105263e-06,
      "loss": 0.3997,
      "step": 294
    },
    {
      "Batch Mean": 3.028531551361084,
      "accuracy": 0.828125,
      "epoch": 0.735,
      "step": 294
    },
    {
      "epoch": 0.7375,
      "grad_norm": 9.361329078674316,
      "learning_rate": 1.3815789473684212e-06,
      "loss": 0.4117,
      "step": 295
    },
    {
      "Batch Mean": 3.166205406188965,
      "accuracy": 0.7890625,
      "epoch": 0.7375,
      "step": 295
    },
    {
      "epoch": 0.74,
      "grad_norm": 8.724553108215332,
      "learning_rate": 1.3684210526315791e-06,
      "loss": 0.3743,
      "step": 296
    },
    {
      "Batch Mean": 3.3602304458618164,
      "accuracy": 0.8125,
      "epoch": 0.74,
      "step": 296
    },
    {
      "epoch": 0.7425,
      "grad_norm": 9.30712604522705,
      "learning_rate": 1.355263157894737e-06,
      "loss": 0.4148,
      "step": 297
    },
    {
      "Batch Mean": 3.169215202331543,
      "accuracy": 0.828125,
      "epoch": 0.7425,
      "step": 297
    },
    {
      "epoch": 0.745,
      "grad_norm": 9.28420352935791,
      "learning_rate": 1.342105263157895e-06,
      "loss": 0.3247,
      "step": 298
    },
    {
      "Batch Mean": 3.769045829772949,
      "accuracy": 0.8125,
      "epoch": 0.745,
      "step": 298
    },
    {
      "epoch": 0.7475,
      "grad_norm": 9.424715995788574,
      "learning_rate": 1.3289473684210526e-06,
      "loss": 0.3602,
      "step": 299
    },
    {
      "Batch Mean": 3.6478939056396484,
      "accuracy": 0.859375,
      "epoch": 0.7475,
      "step": 299
    },
    {
      "epoch": 0.75,
      "grad_norm": 9.039104461669922,
      "learning_rate": 1.3157894736842106e-06,
      "loss": 0.3256,
      "step": 300
    },
    {
      "Batch Mean": 3.6405322551727295,
      "accuracy": 0.8125,
      "epoch": 0.75,
      "step": 300
    },
    {
      "epoch": 0.7525,
      "grad_norm": 9.519824981689453,
      "learning_rate": 1.3026315789473685e-06,
      "loss": 0.4135,
      "step": 301
    },
    {
      "Batch Mean": 3.9676380157470703,
      "accuracy": 0.828125,
      "epoch": 0.7525,
      "step": 301
    },
    {
      "epoch": 0.755,
      "grad_norm": 9.68959903717041,
      "learning_rate": 1.2894736842105266e-06,
      "loss": 0.3233,
      "step": 302
    },
    {
      "Batch Mean": 4.056262969970703,
      "accuracy": 0.8203125,
      "epoch": 0.755,
      "step": 302
    },
    {
      "epoch": 0.7575,
      "grad_norm": 11.665163040161133,
      "learning_rate": 1.2763157894736845e-06,
      "loss": 0.4197,
      "step": 303
    },
    {
      "Batch Mean": 3.99052095413208,
      "accuracy": 0.84375,
      "epoch": 0.7575,
      "step": 303
    },
    {
      "epoch": 0.76,
      "grad_norm": 9.797762870788574,
      "learning_rate": 1.2631578947368422e-06,
      "loss": 0.3341,
      "step": 304
    },
    {
      "Batch Mean": 4.704170227050781,
      "accuracy": 0.890625,
      "epoch": 0.76,
      "step": 304
    },
    {
      "epoch": 0.7625,
      "grad_norm": 10.098402976989746,
      "learning_rate": 1.25e-06,
      "loss": 0.3156,
      "step": 305
    },
    {
      "Batch Mean": 4.484651565551758,
      "accuracy": 0.796875,
      "epoch": 0.7625,
      "step": 305
    },
    {
      "epoch": 0.765,
      "grad_norm": 13.691995620727539,
      "learning_rate": 1.236842105263158e-06,
      "loss": 0.4064,
      "step": 306
    },
    {
      "Batch Mean": 3.8555450439453125,
      "accuracy": 0.8203125,
      "epoch": 0.765,
      "step": 306
    },
    {
      "epoch": 0.7675,
      "grad_norm": 11.670147895812988,
      "learning_rate": 1.223684210526316e-06,
      "loss": 0.4056,
      "step": 307
    },
    {
      "Batch Mean": 4.468677043914795,
      "accuracy": 0.84375,
      "epoch": 0.7675,
      "step": 307
    },
    {
      "epoch": 0.77,
      "grad_norm": 11.534550666809082,
      "learning_rate": 1.2105263157894738e-06,
      "loss": 0.4006,
      "step": 308
    },
    {
      "Batch Mean": 3.834693431854248,
      "accuracy": 0.7734375,
      "epoch": 0.77,
      "step": 308
    },
    {
      "epoch": 0.7725,
      "grad_norm": 11.872078895568848,
      "learning_rate": 1.1973684210526317e-06,
      "loss": 0.4734,
      "step": 309
    },
    {
      "Batch Mean": 3.850661277770996,
      "accuracy": 0.7734375,
      "epoch": 0.7725,
      "step": 309
    },
    {
      "epoch": 0.775,
      "grad_norm": 12.335707664489746,
      "learning_rate": 1.1842105263157894e-06,
      "loss": 0.4272,
      "step": 310
    },
    {
      "Batch Mean": 3.788878917694092,
      "accuracy": 0.734375,
      "epoch": 0.775,
      "step": 310
    },
    {
      "epoch": 0.7775,
      "grad_norm": 13.264825820922852,
      "learning_rate": 1.1710526315789476e-06,
      "loss": 0.4977,
      "step": 311
    },
    {
      "Batch Mean": 3.859971046447754,
      "accuracy": 0.78125,
      "epoch": 0.7775,
      "step": 311
    },
    {
      "epoch": 0.78,
      "grad_norm": 10.746715545654297,
      "learning_rate": 1.1578947368421053e-06,
      "loss": 0.4457,
      "step": 312
    },
    {
      "Batch Mean": 3.103464126586914,
      "accuracy": 0.7890625,
      "epoch": 0.78,
      "step": 312
    },
    {
      "epoch": 0.7825,
      "grad_norm": 10.573457717895508,
      "learning_rate": 1.1447368421052632e-06,
      "loss": 0.4286,
      "step": 313
    },
    {
      "Batch Mean": 3.299260139465332,
      "accuracy": 0.7890625,
      "epoch": 0.7825,
      "step": 313
    },
    {
      "epoch": 0.785,
      "grad_norm": 10.1765775680542,
      "learning_rate": 1.1315789473684213e-06,
      "loss": 0.4275,
      "step": 314
    },
    {
      "Batch Mean": 2.910822868347168,
      "accuracy": 0.734375,
      "epoch": 0.785,
      "step": 314
    },
    {
      "epoch": 0.7875,
      "grad_norm": 11.196432113647461,
      "learning_rate": 1.118421052631579e-06,
      "loss": 0.5264,
      "step": 315
    },
    {
      "Batch Mean": 3.1886849403381348,
      "accuracy": 0.8203125,
      "epoch": 0.7875,
      "step": 315
    },
    {
      "epoch": 0.79,
      "grad_norm": 9.3234281539917,
      "learning_rate": 1.1052631578947369e-06,
      "loss": 0.3845,
      "step": 316
    },
    {
      "Batch Mean": 2.9300460815429688,
      "accuracy": 0.7578125,
      "epoch": 0.79,
      "step": 316
    },
    {
      "epoch": 0.7925,
      "grad_norm": 10.788275718688965,
      "learning_rate": 1.0921052631578948e-06,
      "loss": 0.4715,
      "step": 317
    },
    {
      "Batch Mean": 2.96604061126709,
      "accuracy": 0.8203125,
      "epoch": 0.7925,
      "step": 317
    },
    {
      "epoch": 0.795,
      "grad_norm": 9.273138999938965,
      "learning_rate": 1.0789473684210527e-06,
      "loss": 0.4013,
      "step": 318
    },
    {
      "Batch Mean": 2.7190191745758057,
      "accuracy": 0.8359375,
      "epoch": 0.795,
      "step": 318
    },
    {
      "epoch": 0.7975,
      "grad_norm": 8.655777931213379,
      "learning_rate": 1.0657894736842106e-06,
      "loss": 0.3833,
      "step": 319
    },
    {
      "Batch Mean": 2.6934633255004883,
      "accuracy": 0.7734375,
      "epoch": 0.7975,
      "step": 319
    },
    {
      "epoch": 0.8,
      "grad_norm": 8.541173934936523,
      "learning_rate": 1.0526315789473685e-06,
      "loss": 0.4374,
      "step": 320
    },
    {
      "Batch Mean": 2.920466423034668,
      "accuracy": 0.734375,
      "epoch": 0.8,
      "step": 320
    },
    {
      "epoch": 0.8025,
      "grad_norm": 10.408062934875488,
      "learning_rate": 1.0394736842105264e-06,
      "loss": 0.5146,
      "step": 321
    },
    {
      "Batch Mean": 3.0105621814727783,
      "accuracy": 0.828125,
      "epoch": 0.8025,
      "step": 321
    },
    {
      "epoch": 0.805,
      "grad_norm": 8.558656692504883,
      "learning_rate": 1.0263157894736843e-06,
      "loss": 0.4025,
      "step": 322
    },
    {
      "Batch Mean": 2.5406768321990967,
      "accuracy": 0.7109375,
      "epoch": 0.805,
      "step": 322
    },
    {
      "epoch": 0.8075,
      "grad_norm": 10.365202903747559,
      "learning_rate": 1.013157894736842e-06,
      "loss": 0.5076,
      "step": 323
    },
    {
      "Batch Mean": 2.724036693572998,
      "accuracy": 0.8125,
      "epoch": 0.8075,
      "step": 323
    },
    {
      "epoch": 0.81,
      "grad_norm": 9.693260192871094,
      "learning_rate": 1.0000000000000002e-06,
      "loss": 0.4517,
      "step": 324
    },
    {
      "Batch Mean": 2.643817901611328,
      "accuracy": 0.7734375,
      "epoch": 0.81,
      "step": 324
    },
    {
      "epoch": 0.8125,
      "grad_norm": 9.147430419921875,
      "learning_rate": 9.86842105263158e-07,
      "loss": 0.4092,
      "step": 325
    },
    {
      "Batch Mean": 2.660895347595215,
      "accuracy": 0.8125,
      "epoch": 0.8125,
      "step": 325
    },
    {
      "epoch": 0.815,
      "grad_norm": 8.100640296936035,
      "learning_rate": 9.736842105263158e-07,
      "loss": 0.4067,
      "step": 326
    },
    {
      "Batch Mean": 2.5972213745117188,
      "accuracy": 0.7734375,
      "epoch": 0.815,
      "step": 326
    },
    {
      "epoch": 0.8175,
      "grad_norm": 8.55834674835205,
      "learning_rate": 9.605263157894737e-07,
      "loss": 0.4489,
      "step": 327
    },
    {
      "Batch Mean": 2.7025046348571777,
      "accuracy": 0.8125,
      "epoch": 0.8175,
      "step": 327
    },
    {
      "epoch": 0.82,
      "grad_norm": 7.632991313934326,
      "learning_rate": 9.473684210526317e-07,
      "loss": 0.4292,
      "step": 328
    },
    {
      "Batch Mean": 2.485902786254883,
      "accuracy": 0.8046875,
      "epoch": 0.82,
      "step": 328
    },
    {
      "epoch": 0.8225,
      "grad_norm": 8.363178253173828,
      "learning_rate": 9.342105263157895e-07,
      "loss": 0.3919,
      "step": 329
    },
    {
      "Batch Mean": 2.6269283294677734,
      "accuracy": 0.8203125,
      "epoch": 0.8225,
      "step": 329
    },
    {
      "epoch": 0.825,
      "grad_norm": 8.078348159790039,
      "learning_rate": 9.210526315789474e-07,
      "loss": 0.4515,
      "step": 330
    },
    {
      "Batch Mean": 2.464303970336914,
      "accuracy": 0.8203125,
      "epoch": 0.825,
      "step": 330
    },
    {
      "epoch": 0.8275,
      "grad_norm": 8.487800598144531,
      "learning_rate": 9.078947368421054e-07,
      "loss": 0.4312,
      "step": 331
    },
    {
      "Batch Mean": 2.7176826000213623,
      "accuracy": 0.84375,
      "epoch": 0.8275,
      "step": 331
    },
    {
      "epoch": 0.83,
      "grad_norm": 7.366917133331299,
      "learning_rate": 8.947368421052632e-07,
      "loss": 0.3685,
      "step": 332
    },
    {
      "Batch Mean": 2.4911954402923584,
      "accuracy": 0.828125,
      "epoch": 0.83,
      "step": 332
    },
    {
      "epoch": 0.8325,
      "grad_norm": 7.673061370849609,
      "learning_rate": 8.815789473684211e-07,
      "loss": 0.3904,
      "step": 333
    },
    {
      "Batch Mean": 2.8516788482666016,
      "accuracy": 0.8125,
      "epoch": 0.8325,
      "step": 333
    },
    {
      "epoch": 0.835,
      "grad_norm": 7.1761088371276855,
      "learning_rate": 8.68421052631579e-07,
      "loss": 0.3968,
      "step": 334
    },
    {
      "Batch Mean": 3.2786450386047363,
      "accuracy": 0.8125,
      "epoch": 0.835,
      "step": 334
    },
    {
      "epoch": 0.8375,
      "grad_norm": 6.7936177253723145,
      "learning_rate": 8.55263157894737e-07,
      "loss": 0.3717,
      "step": 335
    },
    {
      "Batch Mean": 3.0723023414611816,
      "accuracy": 0.828125,
      "epoch": 0.8375,
      "step": 335
    },
    {
      "epoch": 0.84,
      "grad_norm": 6.839074611663818,
      "learning_rate": 8.421052631578948e-07,
      "loss": 0.3472,
      "step": 336
    },
    {
      "Batch Mean": 2.7969651222229004,
      "accuracy": 0.890625,
      "epoch": 0.84,
      "step": 336
    },
    {
      "epoch": 0.8425,
      "grad_norm": 7.81821870803833,
      "learning_rate": 8.289473684210527e-07,
      "loss": 0.3287,
      "step": 337
    },
    {
      "Batch Mean": 3.1232268810272217,
      "accuracy": 0.7890625,
      "epoch": 0.8425,
      "step": 337
    },
    {
      "epoch": 0.845,
      "grad_norm": 9.474512100219727,
      "learning_rate": 8.157894736842106e-07,
      "loss": 0.4474,
      "step": 338
    },
    {
      "Batch Mean": 3.809293746948242,
      "accuracy": 0.78125,
      "epoch": 0.845,
      "step": 338
    },
    {
      "epoch": 0.8475,
      "grad_norm": 8.809016227722168,
      "learning_rate": 8.026315789473685e-07,
      "loss": 0.4153,
      "step": 339
    },
    {
      "Batch Mean": 3.514373302459717,
      "accuracy": 0.8203125,
      "epoch": 0.8475,
      "step": 339
    },
    {
      "epoch": 0.85,
      "grad_norm": 8.725719451904297,
      "learning_rate": 7.894736842105263e-07,
      "loss": 0.3813,
      "step": 340
    },
    {
      "Batch Mean": 3.5422677993774414,
      "accuracy": 0.78125,
      "epoch": 0.85,
      "step": 340
    },
    {
      "epoch": 0.8525,
      "grad_norm": 8.447900772094727,
      "learning_rate": 7.763157894736843e-07,
      "loss": 0.4199,
      "step": 341
    },
    {
      "Batch Mean": 3.5171260833740234,
      "accuracy": 0.8359375,
      "epoch": 0.8525,
      "step": 341
    },
    {
      "epoch": 0.855,
      "grad_norm": 8.496635437011719,
      "learning_rate": 7.631578947368422e-07,
      "loss": 0.3542,
      "step": 342
    },
    {
      "Batch Mean": 3.378173828125,
      "accuracy": 0.875,
      "epoch": 0.855,
      "step": 342
    },
    {
      "epoch": 0.8575,
      "grad_norm": 7.566522121429443,
      "learning_rate": 7.5e-07,
      "loss": 0.2996,
      "step": 343
    },
    {
      "Batch Mean": 3.8742589950561523,
      "accuracy": 0.8203125,
      "epoch": 0.8575,
      "step": 343
    },
    {
      "epoch": 0.86,
      "grad_norm": 8.366487503051758,
      "learning_rate": 7.368421052631579e-07,
      "loss": 0.3492,
      "step": 344
    },
    {
      "Batch Mean": 4.238893508911133,
      "accuracy": 0.7734375,
      "epoch": 0.86,
      "step": 344
    },
    {
      "epoch": 0.8625,
      "grad_norm": 10.260600090026855,
      "learning_rate": 7.236842105263158e-07,
      "loss": 0.4601,
      "step": 345
    },
    {
      "Batch Mean": 3.79160213470459,
      "accuracy": 0.828125,
      "epoch": 0.8625,
      "step": 345
    },
    {
      "epoch": 0.865,
      "grad_norm": 9.80223274230957,
      "learning_rate": 7.105263157894737e-07,
      "loss": 0.407,
      "step": 346
    },
    {
      "Batch Mean": 3.4316647052764893,
      "accuracy": 0.78125,
      "epoch": 0.865,
      "step": 346
    },
    {
      "epoch": 0.8675,
      "grad_norm": 9.012909889221191,
      "learning_rate": 6.973684210526316e-07,
      "loss": 0.4111,
      "step": 347
    },
    {
      "Batch Mean": 3.169463872909546,
      "accuracy": 0.8203125,
      "epoch": 0.8675,
      "step": 347
    },
    {
      "epoch": 0.87,
      "grad_norm": 7.685094356536865,
      "learning_rate": 6.842105263157896e-07,
      "loss": 0.368,
      "step": 348
    },
    {
      "Batch Mean": 3.5023093223571777,
      "accuracy": 0.8203125,
      "epoch": 0.87,
      "step": 348
    },
    {
      "epoch": 0.8725,
      "grad_norm": 8.767780303955078,
      "learning_rate": 6.710526315789475e-07,
      "loss": 0.3806,
      "step": 349
    },
    {
      "Batch Mean": 3.8760428428649902,
      "accuracy": 0.8046875,
      "epoch": 0.8725,
      "step": 349
    },
    {
      "epoch": 0.875,
      "grad_norm": 10.924077987670898,
      "learning_rate": 6.578947368421053e-07,
      "loss": 0.4436,
      "step": 350
    },
    {
      "Batch Mean": 3.915463447570801,
      "accuracy": 0.828125,
      "epoch": 0.875,
      "step": 350
    },
    {
      "epoch": 0.8775,
      "grad_norm": 11.166847229003906,
      "learning_rate": 6.447368421052633e-07,
      "loss": 0.4323,
      "step": 351
    },
    {
      "Batch Mean": 3.3088676929473877,
      "accuracy": 0.8359375,
      "epoch": 0.8775,
      "step": 351
    },
    {
      "epoch": 0.88,
      "grad_norm": 7.589693546295166,
      "learning_rate": 6.315789473684211e-07,
      "loss": 0.3519,
      "step": 352
    },
    {
      "Batch Mean": 2.881253957748413,
      "accuracy": 0.7890625,
      "epoch": 0.88,
      "step": 352
    },
    {
      "epoch": 0.8825,
      "grad_norm": 9.315062522888184,
      "learning_rate": 6.18421052631579e-07,
      "loss": 0.4301,
      "step": 353
    },
    {
      "Batch Mean": 2.906712055206299,
      "accuracy": 0.765625,
      "epoch": 0.8825,
      "step": 353
    },
    {
      "epoch": 0.885,
      "grad_norm": 11.210281372070312,
      "learning_rate": 6.052631578947369e-07,
      "loss": 0.4919,
      "step": 354
    },
    {
      "Batch Mean": 3.3116188049316406,
      "accuracy": 0.828125,
      "epoch": 0.885,
      "step": 354
    },
    {
      "epoch": 0.8875,
      "grad_norm": 8.310154914855957,
      "learning_rate": 5.921052631578947e-07,
      "loss": 0.3854,
      "step": 355
    },
    {
      "Batch Mean": 3.0483903884887695,
      "accuracy": 0.796875,
      "epoch": 0.8875,
      "step": 355
    },
    {
      "epoch": 0.89,
      "grad_norm": 9.462825775146484,
      "learning_rate": 5.789473684210526e-07,
      "loss": 0.4246,
      "step": 356
    },
    {
      "Batch Mean": 3.21954345703125,
      "accuracy": 0.8046875,
      "epoch": 0.89,
      "step": 356
    },
    {
      "epoch": 0.8925,
      "grad_norm": 8.243721008300781,
      "learning_rate": 5.657894736842106e-07,
      "loss": 0.3305,
      "step": 357
    },
    {
      "Batch Mean": 3.143207311630249,
      "accuracy": 0.7578125,
      "epoch": 0.8925,
      "step": 357
    },
    {
      "epoch": 0.895,
      "grad_norm": 8.959394454956055,
      "learning_rate": 5.526315789473684e-07,
      "loss": 0.4026,
      "step": 358
    },
    {
      "Batch Mean": 3.196836471557617,
      "accuracy": 0.7890625,
      "epoch": 0.895,
      "step": 358
    },
    {
      "epoch": 0.8975,
      "grad_norm": 9.289172172546387,
      "learning_rate": 5.394736842105264e-07,
      "loss": 0.4167,
      "step": 359
    },
    {
      "Batch Mean": 2.7712502479553223,
      "accuracy": 0.8359375,
      "epoch": 0.8975,
      "step": 359
    },
    {
      "epoch": 0.9,
      "grad_norm": 8.765880584716797,
      "learning_rate": 5.263157894736843e-07,
      "loss": 0.3261,
      "step": 360
    },
    {
      "Batch Mean": 3.0266494750976562,
      "accuracy": 0.8125,
      "epoch": 0.9,
      "step": 360
    },
    {
      "epoch": 0.9025,
      "grad_norm": 8.701887130737305,
      "learning_rate": 5.131578947368422e-07,
      "loss": 0.4457,
      "step": 361
    },
    {
      "Batch Mean": 3.071992874145508,
      "accuracy": 0.796875,
      "epoch": 0.9025,
      "step": 361
    },
    {
      "epoch": 0.905,
      "grad_norm": 9.102984428405762,
      "learning_rate": 5.000000000000001e-07,
      "loss": 0.4204,
      "step": 362
    },
    {
      "Batch Mean": 2.5766148567199707,
      "accuracy": 0.84375,
      "epoch": 0.905,
      "step": 362
    },
    {
      "epoch": 0.9075,
      "grad_norm": 7.755275249481201,
      "learning_rate": 4.868421052631579e-07,
      "loss": 0.372,
      "step": 363
    },
    {
      "Batch Mean": 3.2133255004882812,
      "accuracy": 0.78125,
      "epoch": 0.9075,
      "step": 363
    },
    {
      "epoch": 0.91,
      "grad_norm": 8.099285125732422,
      "learning_rate": 4.7368421052631585e-07,
      "loss": 0.4035,
      "step": 364
    },
    {
      "Batch Mean": 2.7393627166748047,
      "accuracy": 0.796875,
      "epoch": 0.91,
      "step": 364
    },
    {
      "epoch": 0.9125,
      "grad_norm": 8.071907043457031,
      "learning_rate": 4.605263157894737e-07,
      "loss": 0.4003,
      "step": 365
    },
    {
      "Batch Mean": 2.6384525299072266,
      "accuracy": 0.7890625,
      "epoch": 0.9125,
      "step": 365
    },
    {
      "epoch": 0.915,
      "grad_norm": 8.659152030944824,
      "learning_rate": 4.473684210526316e-07,
      "loss": 0.4324,
      "step": 366
    },
    {
      "Batch Mean": 3.1158814430236816,
      "accuracy": 0.8359375,
      "epoch": 0.915,
      "step": 366
    },
    {
      "epoch": 0.9175,
      "grad_norm": 7.912168979644775,
      "learning_rate": 4.342105263157895e-07,
      "loss": 0.363,
      "step": 367
    },
    {
      "Batch Mean": 2.7819793224334717,
      "accuracy": 0.8203125,
      "epoch": 0.9175,
      "step": 367
    },
    {
      "epoch": 0.92,
      "grad_norm": 9.396142959594727,
      "learning_rate": 4.210526315789474e-07,
      "loss": 0.436,
      "step": 368
    },
    {
      "Batch Mean": 2.770251750946045,
      "accuracy": 0.8046875,
      "epoch": 0.92,
      "step": 368
    },
    {
      "epoch": 0.9225,
      "grad_norm": 8.482458114624023,
      "learning_rate": 4.078947368421053e-07,
      "loss": 0.4448,
      "step": 369
    },
    {
      "Batch Mean": 2.6800403594970703,
      "accuracy": 0.828125,
      "epoch": 0.9225,
      "step": 369
    },
    {
      "epoch": 0.925,
      "grad_norm": 7.598038673400879,
      "learning_rate": 3.9473684210526315e-07,
      "loss": 0.3877,
      "step": 370
    },
    {
      "Batch Mean": 2.463380813598633,
      "accuracy": 0.7734375,
      "epoch": 0.925,
      "step": 370
    },
    {
      "epoch": 0.9275,
      "grad_norm": 9.041579246520996,
      "learning_rate": 3.815789473684211e-07,
      "loss": 0.4531,
      "step": 371
    },
    {
      "Batch Mean": 2.664177894592285,
      "accuracy": 0.84375,
      "epoch": 0.9275,
      "step": 371
    },
    {
      "epoch": 0.93,
      "grad_norm": 8.277569770812988,
      "learning_rate": 3.6842105263157896e-07,
      "loss": 0.3945,
      "step": 372
    },
    {
      "Batch Mean": 2.9936976432800293,
      "accuracy": 0.859375,
      "epoch": 0.93,
      "step": 372
    },
    {
      "epoch": 0.9325,
      "grad_norm": 8.517739295959473,
      "learning_rate": 3.5526315789473687e-07,
      "loss": 0.3599,
      "step": 373
    },
    {
      "Batch Mean": 2.63564395904541,
      "accuracy": 0.875,
      "epoch": 0.9325,
      "step": 373
    },
    {
      "epoch": 0.935,
      "grad_norm": 7.75486421585083,
      "learning_rate": 3.421052631578948e-07,
      "loss": 0.3049,
      "step": 374
    },
    {
      "Batch Mean": 2.2013025283813477,
      "accuracy": 0.8125,
      "epoch": 0.935,
      "step": 374
    },
    {
      "epoch": 0.9375,
      "grad_norm": 8.340994834899902,
      "learning_rate": 3.2894736842105264e-07,
      "loss": 0.3908,
      "step": 375
    },
    {
      "Batch Mean": 2.306203842163086,
      "accuracy": 0.84375,
      "epoch": 0.9375,
      "step": 375
    },
    {
      "epoch": 0.94,
      "grad_norm": 8.02235221862793,
      "learning_rate": 3.1578947368421055e-07,
      "loss": 0.3408,
      "step": 376
    },
    {
      "Batch Mean": 2.780491590499878,
      "accuracy": 0.75,
      "epoch": 0.94,
      "step": 376
    },
    {
      "epoch": 0.9425,
      "grad_norm": 11.336405754089355,
      "learning_rate": 3.0263157894736846e-07,
      "loss": 0.5006,
      "step": 377
    },
    {
      "Batch Mean": 2.413102388381958,
      "accuracy": 0.828125,
      "epoch": 0.9425,
      "step": 377
    },
    {
      "epoch": 0.945,
      "grad_norm": 8.683170318603516,
      "learning_rate": 2.894736842105263e-07,
      "loss": 0.374,
      "step": 378
    },
    {
      "Batch Mean": 2.5188517570495605,
      "accuracy": 0.8203125,
      "epoch": 0.945,
      "step": 378
    },
    {
      "epoch": 0.9475,
      "grad_norm": 8.190141677856445,
      "learning_rate": 2.763157894736842e-07,
      "loss": 0.4097,
      "step": 379
    },
    {
      "Batch Mean": 2.7684450149536133,
      "accuracy": 0.78125,
      "epoch": 0.9475,
      "step": 379
    },
    {
      "epoch": 0.95,
      "grad_norm": 8.226295471191406,
      "learning_rate": 2.6315789473684213e-07,
      "loss": 0.3831,
      "step": 380
    },
    {
      "Batch Mean": 2.796868324279785,
      "accuracy": 0.796875,
      "epoch": 0.95,
      "step": 380
    },
    {
      "epoch": 0.9525,
      "grad_norm": 9.279502868652344,
      "learning_rate": 2.5000000000000004e-07,
      "loss": 0.4314,
      "step": 381
    },
    {
      "Batch Mean": 2.3730950355529785,
      "accuracy": 0.78125,
      "epoch": 0.9525,
      "step": 381
    },
    {
      "epoch": 0.955,
      "grad_norm": 8.747193336486816,
      "learning_rate": 2.3684210526315792e-07,
      "loss": 0.4125,
      "step": 382
    },
    {
      "Batch Mean": 2.7471389770507812,
      "accuracy": 0.7734375,
      "epoch": 0.955,
      "step": 382
    },
    {
      "epoch": 0.9575,
      "grad_norm": 9.065400123596191,
      "learning_rate": 2.236842105263158e-07,
      "loss": 0.4453,
      "step": 383
    },
    {
      "Batch Mean": 2.758594274520874,
      "accuracy": 0.84375,
      "epoch": 0.9575,
      "step": 383
    },
    {
      "epoch": 0.96,
      "grad_norm": 8.173842430114746,
      "learning_rate": 2.105263157894737e-07,
      "loss": 0.3516,
      "step": 384
    },
    {
      "Batch Mean": 2.677278518676758,
      "accuracy": 0.8515625,
      "epoch": 0.96,
      "step": 384
    },
    {
      "epoch": 0.9625,
      "grad_norm": 7.212657451629639,
      "learning_rate": 1.9736842105263157e-07,
      "loss": 0.3345,
      "step": 385
    },
    {
      "Batch Mean": 3.0301685333251953,
      "accuracy": 0.859375,
      "epoch": 0.9625,
      "step": 385
    },
    {
      "epoch": 0.965,
      "grad_norm": 7.157632827758789,
      "learning_rate": 1.8421052631578948e-07,
      "loss": 0.325,
      "step": 386
    },
    {
      "Batch Mean": 2.6034302711486816,
      "accuracy": 0.8046875,
      "epoch": 0.965,
      "step": 386
    },
    {
      "epoch": 0.9675,
      "grad_norm": 8.646699905395508,
      "learning_rate": 1.710526315789474e-07,
      "loss": 0.364,
      "step": 387
    },
    {
      "Batch Mean": 2.835425853729248,
      "accuracy": 0.8359375,
      "epoch": 0.9675,
      "step": 387
    },
    {
      "epoch": 0.97,
      "grad_norm": 8.081703186035156,
      "learning_rate": 1.5789473684210527e-07,
      "loss": 0.3987,
      "step": 388
    },
    {
      "Batch Mean": 2.878167152404785,
      "accuracy": 0.828125,
      "epoch": 0.97,
      "step": 388
    },
    {
      "epoch": 0.9725,
      "grad_norm": 8.71593189239502,
      "learning_rate": 1.4473684210526316e-07,
      "loss": 0.3909,
      "step": 389
    },
    {
      "Batch Mean": 2.9658942222595215,
      "accuracy": 0.8671875,
      "epoch": 0.9725,
      "step": 389
    },
    {
      "epoch": 0.975,
      "grad_norm": 7.501869201660156,
      "learning_rate": 1.3157894736842107e-07,
      "loss": 0.3144,
      "step": 390
    },
    {
      "Batch Mean": 3.0065078735351562,
      "accuracy": 0.8125,
      "epoch": 0.975,
      "step": 390
    },
    {
      "epoch": 0.9775,
      "grad_norm": 9.999802589416504,
      "learning_rate": 1.1842105263157896e-07,
      "loss": 0.4321,
      "step": 391
    },
    {
      "Batch Mean": 3.0289103984832764,
      "accuracy": 0.84375,
      "epoch": 0.9775,
      "step": 391
    },
    {
      "epoch": 0.98,
      "grad_norm": 7.703149318695068,
      "learning_rate": 1.0526315789473685e-07,
      "loss": 0.3455,
      "step": 392
    },
    {
      "Batch Mean": 2.7720205783843994,
      "accuracy": 0.828125,
      "epoch": 0.98,
      "step": 392
    },
    {
      "epoch": 0.9825,
      "grad_norm": 8.7222261428833,
      "learning_rate": 9.210526315789474e-08,
      "loss": 0.3681,
      "step": 393
    },
    {
      "Batch Mean": 2.607419490814209,
      "accuracy": 0.8203125,
      "epoch": 0.9825,
      "step": 393
    },
    {
      "epoch": 0.985,
      "grad_norm": 8.8170747756958,
      "learning_rate": 7.894736842105264e-08,
      "loss": 0.3779,
      "step": 394
    },
    {
      "Batch Mean": 3.094576835632324,
      "accuracy": 0.84375,
      "epoch": 0.985,
      "step": 394
    },
    {
      "epoch": 0.9875,
      "grad_norm": 8.42007827758789,
      "learning_rate": 6.578947368421053e-08,
      "loss": 0.3663,
      "step": 395
    },
    {
      "Batch Mean": 2.809164524078369,
      "accuracy": 0.84375,
      "epoch": 0.9875,
      "step": 395
    },
    {
      "epoch": 0.99,
      "grad_norm": 7.778329849243164,
      "learning_rate": 5.263157894736842e-08,
      "loss": 0.3257,
      "step": 396
    },
    {
      "Batch Mean": 2.802082061767578,
      "accuracy": 0.7734375,
      "epoch": 0.99,
      "step": 396
    },
    {
      "epoch": 0.9925,
      "grad_norm": 10.571059226989746,
      "learning_rate": 3.947368421052632e-08,
      "loss": 0.4216,
      "step": 397
    },
    {
      "Batch Mean": 3.107623815536499,
      "accuracy": 0.8046875,
      "epoch": 0.9925,
      "step": 397
    },
    {
      "epoch": 0.995,
      "grad_norm": 8.333806991577148,
      "learning_rate": 2.631578947368421e-08,
      "loss": 0.4045,
      "step": 398
    },
    {
      "Batch Mean": 2.785451889038086,
      "accuracy": 0.8125,
      "epoch": 0.995,
      "step": 398
    },
    {
      "epoch": 0.9975,
      "grad_norm": 10.335576057434082,
      "learning_rate": 1.3157894736842106e-08,
      "loss": 0.4548,
      "step": 399
    },
    {
      "Batch Mean": 2.530060052871704,
      "accuracy": 0.765625,
      "epoch": 0.9975,
      "step": 399
    },
    {
      "epoch": 1.0,
      "grad_norm": 10.57154369354248,
      "learning_rate": 0.0,
      "loss": 0.4866,
      "step": 400
    }
  ],
  "logging_steps": 1,
  "max_steps": 400,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 80,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}