{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.8,
  "eval_steps": 500,
  "global_step": 320,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "Batch Mean": -3.411376953125,
      "accuracy": 0.40625,
      "epoch": 0,
      "step": 0
    },
    {
      "Batch Mean": -3.425048828125,
      "accuracy": 0.4375,
      "epoch": 0,
      "step": 0
    },
    {
      "Batch Mean": -3.405029296875,
      "accuracy": 0.53125,
      "epoch": 0,
      "step": 0
    },
    {
      "Batch Mean": -3.482421875,
      "accuracy": 0.3125,
      "epoch": 0,
      "step": 0
    },
    {
      "epoch": 0.0025,
      "grad_norm": 7.395800590515137,
      "learning_rate": 1.5000000000000002e-07,
      "loss": 0.7083,
      "step": 1
    },
    {
      "Batch Mean": -3.413330078125,
      "accuracy": 0.46875,
      "epoch": 0.0025,
      "step": 1
    },
    {
      "Batch Mean": -3.407958984375,
      "accuracy": 0.5,
      "epoch": 0.0025,
      "step": 1
    },
    {
      "Batch Mean": -3.4326171875,
      "accuracy": 0.46875,
      "epoch": 0.0025,
      "step": 1
    },
    {
      "Batch Mean": -3.40673828125,
      "accuracy": 0.46875,
      "epoch": 0.0025,
      "step": 1
    },
    {
      "epoch": 0.005,
      "grad_norm": 6.054078102111816,
      "learning_rate": 3.0000000000000004e-07,
      "loss": 0.7038,
      "step": 2
    },
    {
      "Batch Mean": -3.418212890625,
      "accuracy": 0.46875,
      "epoch": 0.005,
      "step": 2
    },
    {
      "Batch Mean": -3.4130859375,
      "accuracy": 0.5625,
      "epoch": 0.005,
      "step": 2
    },
    {
      "Batch Mean": -3.4619140625,
      "accuracy": 0.375,
      "epoch": 0.005,
      "step": 2
    },
    {
      "Batch Mean": -3.51220703125,
      "accuracy": 0.5,
      "epoch": 0.005,
      "step": 2
    },
    {
      "epoch": 0.0075,
      "grad_norm": 6.486208915710449,
      "learning_rate": 4.5e-07,
      "loss": 0.6902,
      "step": 3
    },
    {
      "Batch Mean": -3.40234375,
      "accuracy": 0.5,
      "epoch": 0.0075,
      "step": 3
    },
    {
      "Batch Mean": -3.492431640625,
      "accuracy": 0.46875,
      "epoch": 0.0075,
      "step": 3
    },
    {
      "Batch Mean": -3.45166015625,
      "accuracy": 0.46875,
      "epoch": 0.0075,
      "step": 3
    },
    {
      "Batch Mean": -3.4267578125,
      "accuracy": 0.5625,
      "epoch": 0.0075,
      "step": 3
    },
    {
      "epoch": 0.01,
      "grad_norm": 6.4884724617004395,
      "learning_rate": 6.000000000000001e-07,
      "loss": 0.6921,
      "step": 4
    },
    {
      "Batch Mean": -3.444580078125,
      "accuracy": 0.59375,
      "epoch": 0.01,
      "step": 4
    },
    {
      "Batch Mean": -3.358154296875,
      "accuracy": 0.65625,
      "epoch": 0.01,
      "step": 4
    },
    {
      "Batch Mean": -3.40966796875,
      "accuracy": 0.5625,
      "epoch": 0.01,
      "step": 4
    },
    {
      "Batch Mean": -3.458251953125,
      "accuracy": 0.40625,
      "epoch": 0.01,
      "step": 4
    },
    {
      "epoch": 0.0125,
      "grad_norm": 6.3613505363464355,
      "learning_rate": 7.5e-07,
      "loss": 0.6776,
      "step": 5
    },
    {
      "Batch Mean": -3.421875,
      "accuracy": 0.5,
      "epoch": 0.0125,
      "step": 5
    },
    {
      "Batch Mean": -3.458740234375,
      "accuracy": 0.375,
      "epoch": 0.0125,
      "step": 5
    },
    {
      "Batch Mean": -3.420166015625,
      "accuracy": 0.5,
      "epoch": 0.0125,
      "step": 5
    },
    {
      "Batch Mean": -3.45556640625,
      "accuracy": 0.59375,
      "epoch": 0.0125,
      "step": 5
    },
    {
      "epoch": 0.015,
      "grad_norm": 8.138617515563965,
      "learning_rate": 9e-07,
      "loss": 0.6939,
      "step": 6
    },
    {
      "Batch Mean": -3.35302734375,
      "accuracy": 0.40625,
      "epoch": 0.015,
      "step": 6
    },
    {
      "Batch Mean": -3.459228515625,
      "accuracy": 0.5,
      "epoch": 0.015,
      "step": 6
    },
    {
      "Batch Mean": -3.4326171875,
      "accuracy": 0.71875,
      "epoch": 0.015,
      "step": 6
    },
    {
      "Batch Mean": -3.396240234375,
      "accuracy": 0.4375,
      "epoch": 0.015,
      "step": 6
    },
    {
      "epoch": 0.0175,
      "grad_norm": 6.334463596343994,
      "learning_rate": 1.05e-06,
      "loss": 0.6898,
      "step": 7
    },
    {
      "Batch Mean": -3.45947265625,
      "accuracy": 0.4375,
      "epoch": 0.0175,
      "step": 7
    },
    {
      "Batch Mean": -3.437744140625,
      "accuracy": 0.4375,
      "epoch": 0.0175,
      "step": 7
    },
    {
      "Batch Mean": -3.44287109375,
      "accuracy": 0.59375,
      "epoch": 0.0175,
      "step": 7
    },
    {
      "Batch Mean": -3.466552734375,
      "accuracy": 0.65625,
      "epoch": 0.0175,
      "step": 7
    },
    {
      "epoch": 0.02,
      "grad_norm": 6.167725563049316,
      "learning_rate": 1.2000000000000002e-06,
      "loss": 0.6892,
      "step": 8
    },
    {
      "Batch Mean": -3.371826171875,
      "accuracy": 0.4375,
      "epoch": 0.02,
      "step": 8
    },
    {
      "Batch Mean": -3.379150390625,
      "accuracy": 0.5,
      "epoch": 0.02,
      "step": 8
    },
    {
      "Batch Mean": -3.42626953125,
      "accuracy": 0.59375,
      "epoch": 0.02,
      "step": 8
    },
    {
      "Batch Mean": -3.397216796875,
      "accuracy": 0.3125,
      "epoch": 0.02,
      "step": 8
    },
    {
      "epoch": 0.0225,
      "grad_norm": 6.235499382019043,
      "learning_rate": 1.35e-06,
      "loss": 0.699,
      "step": 9
    },
    {
      "Batch Mean": -3.39697265625,
      "accuracy": 0.5,
      "epoch": 0.0225,
      "step": 9
    },
    {
      "Batch Mean": -3.365966796875,
      "accuracy": 0.53125,
      "epoch": 0.0225,
      "step": 9
    },
    {
      "Batch Mean": -3.386474609375,
      "accuracy": 0.53125,
      "epoch": 0.0225,
      "step": 9
    },
    {
      "Batch Mean": -3.41552734375,
      "accuracy": 0.375,
      "epoch": 0.0225,
      "step": 9
    },
    {
      "epoch": 0.025,
      "grad_norm": 7.100496292114258,
      "learning_rate": 1.5e-06,
      "loss": 0.696,
      "step": 10
    },
    {
      "Batch Mean": -3.426025390625,
      "accuracy": 0.65625,
      "epoch": 0.025,
      "step": 10
    },
    {
      "Batch Mean": -3.349853515625,
      "accuracy": 0.53125,
      "epoch": 0.025,
      "step": 10
    },
    {
      "Batch Mean": -3.46533203125,
      "accuracy": 0.5625,
      "epoch": 0.025,
      "step": 10
    },
    {
      "Batch Mean": -3.40234375,
      "accuracy": 0.5625,
      "epoch": 0.025,
      "step": 10
    },
    {
      "epoch": 0.0275,
      "grad_norm": 6.040953636169434,
      "learning_rate": 1.65e-06,
      "loss": 0.6655,
      "step": 11
    },
    {
      "Batch Mean": -3.449462890625,
      "accuracy": 0.5625,
      "epoch": 0.0275,
      "step": 11
    },
    {
      "Batch Mean": -3.42236328125,
      "accuracy": 0.59375,
      "epoch": 0.0275,
      "step": 11
    },
    {
      "Batch Mean": -3.34033203125,
      "accuracy": 0.53125,
      "epoch": 0.0275,
      "step": 11
    },
    {
      "Batch Mean": -3.392578125,
      "accuracy": 0.53125,
      "epoch": 0.0275,
      "step": 11
    },
    {
      "epoch": 0.03,
      "grad_norm": 6.257072448730469,
      "learning_rate": 1.8e-06,
      "loss": 0.6685,
      "step": 12
    },
    {
      "Batch Mean": -3.33984375,
      "accuracy": 0.65625,
      "epoch": 0.03,
      "step": 12
    },
    {
      "Batch Mean": -3.25537109375,
      "accuracy": 0.5625,
      "epoch": 0.03,
      "step": 12
    },
    {
      "Batch Mean": -3.3603515625,
      "accuracy": 0.59375,
      "epoch": 0.03,
      "step": 12
    },
    {
      "Batch Mean": -3.239501953125,
      "accuracy": 0.4375,
      "epoch": 0.03,
      "step": 12
    },
    {
      "epoch": 0.0325,
      "grad_norm": 6.011251449584961,
      "learning_rate": 1.95e-06,
      "loss": 0.6608,
      "step": 13
    },
    {
      "Batch Mean": -3.31787109375,
      "accuracy": 0.75,
      "epoch": 0.0325,
      "step": 13
    },
    {
      "Batch Mean": -3.3154296875,
      "accuracy": 0.71875,
      "epoch": 0.0325,
      "step": 13
    },
    {
      "Batch Mean": -3.2490234375,
      "accuracy": 0.84375,
      "epoch": 0.0325,
      "step": 13
    },
    {
      "Batch Mean": -3.3046875,
      "accuracy": 0.75,
      "epoch": 0.0325,
      "step": 13
    },
    {
      "epoch": 0.035,
      "grad_norm": 6.238981246948242,
      "learning_rate": 2.1e-06,
      "loss": 0.6297,
      "step": 14
    },
    {
      "Batch Mean": -3.24609375,
      "accuracy": 0.75,
      "epoch": 0.035,
      "step": 14
    },
    {
      "Batch Mean": -3.34130859375,
      "accuracy": 0.625,
      "epoch": 0.035,
      "step": 14
    },
    {
      "Batch Mean": -3.279052734375,
      "accuracy": 0.6875,
      "epoch": 0.035,
      "step": 14
    },
    {
      "Batch Mean": -3.28564453125,
      "accuracy": 0.46875,
      "epoch": 0.035,
      "step": 14
    },
    {
      "epoch": 0.0375,
      "grad_norm": 8.44164752960205,
      "learning_rate": 2.25e-06,
      "loss": 0.6589,
      "step": 15
    },
    {
      "Batch Mean": -3.275146484375,
      "accuracy": 0.65625,
      "epoch": 0.0375,
      "step": 15
    },
    {
      "Batch Mean": -3.27294921875,
      "accuracy": 0.5,
      "epoch": 0.0375,
      "step": 15
    },
    {
      "Batch Mean": -3.300048828125,
      "accuracy": 0.71875,
      "epoch": 0.0375,
      "step": 15
    },
    {
      "Batch Mean": -3.251708984375,
      "accuracy": 0.75,
      "epoch": 0.0375,
      "step": 15
    },
    {
      "epoch": 0.04,
      "grad_norm": 5.432738304138184,
      "learning_rate": 2.4000000000000003e-06,
      "loss": 0.6441,
      "step": 16
    },
    {
      "Batch Mean": -3.21826171875,
      "accuracy": 0.6875,
      "epoch": 0.04,
      "step": 16
    },
    {
      "Batch Mean": -3.194580078125,
      "accuracy": 0.65625,
      "epoch": 0.04,
      "step": 16
    },
    {
      "Batch Mean": -3.182373046875,
      "accuracy": 0.59375,
      "epoch": 0.04,
      "step": 16
    },
    {
      "Batch Mean": -3.169677734375,
      "accuracy": 0.84375,
      "epoch": 0.04,
      "step": 16
    },
    {
      "epoch": 0.0425,
      "grad_norm": 5.111855983734131,
      "learning_rate": 2.55e-06,
      "loss": 0.6361,
      "step": 17
    },
    {
      "Batch Mean": -3.256103515625,
      "accuracy": 0.65625,
      "epoch": 0.0425,
      "step": 17
    },
    {
      "Batch Mean": -3.240234375,
      "accuracy": 0.65625,
      "epoch": 0.0425,
      "step": 17
    },
    {
      "Batch Mean": -3.067138671875,
      "accuracy": 0.65625,
      "epoch": 0.0425,
      "step": 17
    },
    {
      "Batch Mean": -3.288330078125,
      "accuracy": 0.59375,
      "epoch": 0.0425,
      "step": 17
    },
    {
      "epoch": 0.045,
      "grad_norm": 5.055493354797363,
      "learning_rate": 2.7e-06,
      "loss": 0.6128,
      "step": 18
    },
    {
      "Batch Mean": -3.41650390625,
      "accuracy": 0.71875,
      "epoch": 0.045,
      "step": 18
    },
    {
      "Batch Mean": -3.077392578125,
      "accuracy": 0.75,
      "epoch": 0.045,
      "step": 18
    },
    {
      "Batch Mean": -3.1966552734375,
      "accuracy": 0.78125,
      "epoch": 0.045,
      "step": 18
    },
    {
      "Batch Mean": -3.2864990234375,
      "accuracy": 0.625,
      "epoch": 0.045,
      "step": 18
    },
    {
      "epoch": 0.0475,
      "grad_norm": 6.847475051879883,
      "learning_rate": 2.85e-06,
      "loss": 0.5578,
      "step": 19
    },
    {
      "Batch Mean": -3.0543212890625,
      "accuracy": 0.78125,
      "epoch": 0.0475,
      "step": 19
    },
    {
      "Batch Mean": -3.05322265625,
      "accuracy": 0.53125,
      "epoch": 0.0475,
      "step": 19
    },
    {
      "Batch Mean": -2.981201171875,
      "accuracy": 0.71875,
      "epoch": 0.0475,
      "step": 19
    },
    {
      "Batch Mean": -3.3494873046875,
      "accuracy": 0.625,
      "epoch": 0.0475,
      "step": 19
    },
    {
      "epoch": 0.05,
      "grad_norm": 5.472431659698486,
      "learning_rate": 3e-06,
      "loss": 0.5922,
      "step": 20
    },
    {
      "Batch Mean": -3.2989501953125,
      "accuracy": 0.8125,
      "epoch": 0.05,
      "step": 20
    },
    {
      "Batch Mean": -3.1324462890625,
      "accuracy": 0.8125,
      "epoch": 0.05,
      "step": 20
    },
    {
      "Batch Mean": -3.1845703125,
      "accuracy": 0.625,
      "epoch": 0.05,
      "step": 20
    },
    {
      "Batch Mean": -3.0384521484375,
      "accuracy": 0.65625,
      "epoch": 0.05,
      "step": 20
    },
    {
      "epoch": 0.0525,
      "grad_norm": 10.34755802154541,
      "learning_rate": 2.992105263157895e-06,
      "loss": 0.5812,
      "step": 21
    },
    {
      "Batch Mean": -3.0819091796875,
      "accuracy": 0.59375,
      "epoch": 0.0525,
      "step": 21
    },
    {
      "Batch Mean": -3.2308349609375,
      "accuracy": 0.71875,
      "epoch": 0.0525,
      "step": 21
    },
    {
      "Batch Mean": -3.0845947265625,
      "accuracy": 0.6875,
      "epoch": 0.0525,
      "step": 21
    },
    {
      "Batch Mean": -3.302001953125,
      "accuracy": 0.78125,
      "epoch": 0.0525,
      "step": 21
    },
    {
      "epoch": 0.055,
      "grad_norm": 6.563014030456543,
      "learning_rate": 2.9842105263157896e-06,
      "loss": 0.5442,
      "step": 22
    },
    {
      "Batch Mean": -3.26788330078125,
      "accuracy": 0.84375,
      "epoch": 0.055,
      "step": 22
    },
    {
      "Batch Mean": -3.18896484375,
      "accuracy": 0.6875,
      "epoch": 0.055,
      "step": 22
    },
    {
      "Batch Mean": -3.02227783203125,
      "accuracy": 0.8125,
      "epoch": 0.055,
      "step": 22
    },
    {
      "Batch Mean": -3.3311767578125,
      "accuracy": 0.75,
      "epoch": 0.055,
      "step": 22
    },
    {
      "epoch": 0.0575,
      "grad_norm": 6.604343891143799,
      "learning_rate": 2.9763157894736843e-06,
      "loss": 0.5241,
      "step": 23
    },
    {
      "Batch Mean": -3.5736083984375,
      "accuracy": 0.78125,
      "epoch": 0.0575,
      "step": 23
    },
    {
      "Batch Mean": -3.3741493225097656,
      "accuracy": 0.71875,
      "epoch": 0.0575,
      "step": 23
    },
    {
      "Batch Mean": -3.36822509765625,
      "accuracy": 0.71875,
      "epoch": 0.0575,
      "step": 23
    },
    {
      "Batch Mean": -3.2643585205078125,
      "accuracy": 0.8125,
      "epoch": 0.0575,
      "step": 23
    },
    {
      "epoch": 0.06,
      "grad_norm": 7.35377311706543,
      "learning_rate": 2.968421052631579e-06,
      "loss": 0.5049,
      "step": 24
    },
    {
      "Batch Mean": -3.401031494140625,
      "accuracy": 0.84375,
      "epoch": 0.06,
      "step": 24
    },
    {
      "Batch Mean": -3.914642333984375,
      "accuracy": 0.78125,
      "epoch": 0.06,
      "step": 24
    },
    {
      "Batch Mean": -3.43243408203125,
      "accuracy": 0.65625,
      "epoch": 0.06,
      "step": 24
    },
    {
      "Batch Mean": -3.382415771484375,
      "accuracy": 0.71875,
      "epoch": 0.06,
      "step": 24
    },
    {
      "epoch": 0.0625,
      "grad_norm": 11.941130638122559,
      "learning_rate": 2.960526315789474e-06,
      "loss": 0.5471,
      "step": 25
    },
    {
      "Batch Mean": -4.3116302490234375,
      "accuracy": 0.78125,
      "epoch": 0.0625,
      "step": 25
    },
    {
      "Batch Mean": -3.766510009765625,
      "accuracy": 0.8125,
      "epoch": 0.0625,
      "step": 25
    },
    {
      "Batch Mean": -3.777069091796875,
      "accuracy": 0.78125,
      "epoch": 0.0625,
      "step": 25
    },
    {
      "Batch Mean": -3.7879981994628906,
      "accuracy": 0.84375,
      "epoch": 0.0625,
      "step": 25
    },
    {
      "epoch": 0.065,
      "grad_norm": 9.934557914733887,
      "learning_rate": 2.9526315789473685e-06,
      "loss": 0.423,
      "step": 26
    },
    {
      "Batch Mean": -3.805145263671875,
      "accuracy": 0.8125,
      "epoch": 0.065,
      "step": 26
    },
    {
      "Batch Mean": -3.8750686645507812,
      "accuracy": 0.71875,
      "epoch": 0.065,
      "step": 26
    },
    {
      "Batch Mean": -3.9338016510009766,
      "accuracy": 0.53125,
      "epoch": 0.065,
      "step": 26
    },
    {
      "Batch Mean": -3.4400253295898438,
      "accuracy": 0.78125,
      "epoch": 0.065,
      "step": 26
    },
    {
      "epoch": 0.0675,
      "grad_norm": 11.906075477600098,
      "learning_rate": 2.9447368421052633e-06,
      "loss": 0.4943,
      "step": 27
    },
    {
      "Batch Mean": -4.218650817871094,
      "accuracy": 0.71875,
      "epoch": 0.0675,
      "step": 27
    },
    {
      "Batch Mean": -3.3725433349609375,
      "accuracy": 0.78125,
      "epoch": 0.0675,
      "step": 27
    },
    {
      "Batch Mean": -4.4810791015625,
      "accuracy": 0.6875,
      "epoch": 0.0675,
      "step": 27
    },
    {
      "Batch Mean": -4.03173828125,
      "accuracy": 0.6875,
      "epoch": 0.0675,
      "step": 27
    },
    {
      "epoch": 0.07,
      "grad_norm": 14.714795112609863,
      "learning_rate": 2.936842105263158e-06,
      "loss": 0.5273,
      "step": 28
    },
    {
      "Batch Mean": -4.141151428222656,
      "accuracy": 0.6875,
      "epoch": 0.07,
      "step": 28
    },
    {
      "Batch Mean": -3.8033447265625,
      "accuracy": 0.5625,
      "epoch": 0.07,
      "step": 28
    },
    {
      "Batch Mean": -3.9724884033203125,
      "accuracy": 0.59375,
      "epoch": 0.07,
      "step": 28
    },
    {
      "Batch Mean": -4.264892578125,
      "accuracy": 0.65625,
      "epoch": 0.07,
      "step": 28
    },
    {
      "epoch": 0.0725,
      "grad_norm": 24.04404640197754,
      "learning_rate": 2.9289473684210528e-06,
      "loss": 0.7219,
      "step": 29
    },
    {
      "Batch Mean": -4.1270751953125,
      "accuracy": 0.9375,
      "epoch": 0.0725,
      "step": 29
    },
    {
      "Batch Mean": -3.89520263671875,
      "accuracy": 0.6875,
      "epoch": 0.0725,
      "step": 29
    },
    {
      "Batch Mean": -3.8218345642089844,
      "accuracy": 0.59375,
      "epoch": 0.0725,
      "step": 29
    },
    {
      "Batch Mean": -4.262153625488281,
      "accuracy": 0.71875,
      "epoch": 0.0725,
      "step": 29
    },
    {
      "epoch": 0.075,
      "grad_norm": 12.918435096740723,
      "learning_rate": 2.9210526315789475e-06,
      "loss": 0.5446,
      "step": 30
    },
    {
      "Batch Mean": -4.5279541015625,
      "accuracy": 0.84375,
      "epoch": 0.075,
      "step": 30
    },
    {
      "Batch Mean": -4.4937744140625,
      "accuracy": 0.78125,
      "epoch": 0.075,
      "step": 30
    },
    {
      "Batch Mean": -4.4219970703125,
      "accuracy": 0.8125,
      "epoch": 0.075,
      "step": 30
    },
    {
      "Batch Mean": -4.4163818359375,
      "accuracy": 0.75,
      "epoch": 0.075,
      "step": 30
    },
    {
      "epoch": 0.0775,
      "grad_norm": 12.408222198486328,
      "learning_rate": 2.9131578947368423e-06,
      "loss": 0.4516,
      "step": 31
    },
    {
      "Batch Mean": -4.74365234375,
      "accuracy": 0.71875,
      "epoch": 0.0775,
      "step": 31
    },
    {
      "Batch Mean": -4.267730712890625,
      "accuracy": 0.75,
      "epoch": 0.0775,
      "step": 31
    },
    {
      "Batch Mean": -4.415737152099609,
      "accuracy": 0.78125,
      "epoch": 0.0775,
      "step": 31
    },
    {
      "Batch Mean": -4.21026611328125,
      "accuracy": 0.78125,
      "epoch": 0.0775,
      "step": 31
    },
    {
      "epoch": 0.08,
      "grad_norm": 11.342855453491211,
      "learning_rate": 2.905263157894737e-06,
      "loss": 0.4729,
      "step": 32
    },
    {
      "Batch Mean": -3.95538330078125,
      "accuracy": 0.71875,
      "epoch": 0.08,
      "step": 32
    },
    {
      "Batch Mean": -4.6029052734375,
      "accuracy": 0.75,
      "epoch": 0.08,
      "step": 32
    },
    {
      "Batch Mean": -4.25469970703125,
      "accuracy": 0.6875,
      "epoch": 0.08,
      "step": 32
    },
    {
      "Batch Mean": -4.155879974365234,
      "accuracy": 0.75,
      "epoch": 0.08,
      "step": 32
    },
    {
      "epoch": 0.0825,
      "grad_norm": 12.777791976928711,
      "learning_rate": 2.8973684210526318e-06,
      "loss": 0.5232,
      "step": 33
    },
    {
      "Batch Mean": -4.009521484375,
      "accuracy": 0.8125,
      "epoch": 0.0825,
      "step": 33
    },
    {
      "Batch Mean": -3.910858154296875,
      "accuracy": 0.8125,
      "epoch": 0.0825,
      "step": 33
    },
    {
      "Batch Mean": -4.120941162109375,
      "accuracy": 0.84375,
      "epoch": 0.0825,
      "step": 33
    },
    {
      "Batch Mean": -4.4345703125,
      "accuracy": 0.84375,
      "epoch": 0.0825,
      "step": 33
    },
    {
      "epoch": 0.085,
      "grad_norm": 11.19157886505127,
      "learning_rate": 2.8894736842105265e-06,
      "loss": 0.4562,
      "step": 34
    },
    {
      "Batch Mean": -3.889312744140625,
      "accuracy": 0.75,
      "epoch": 0.085,
      "step": 34
    },
    {
      "Batch Mean": -3.822784423828125,
      "accuracy": 0.6875,
      "epoch": 0.085,
      "step": 34
    },
    {
      "Batch Mean": -4.01025390625,
      "accuracy": 0.71875,
      "epoch": 0.085,
      "step": 34
    },
    {
      "Batch Mean": -4.082855224609375,
      "accuracy": 0.78125,
      "epoch": 0.085,
      "step": 34
    },
    {
      "epoch": 0.0875,
      "grad_norm": 12.107293128967285,
      "learning_rate": 2.8815789473684213e-06,
      "loss": 0.5443,
      "step": 35
    },
    {
      "Batch Mean": -3.8009033203125,
      "accuracy": 0.875,
      "epoch": 0.0875,
      "step": 35
    },
    {
      "Batch Mean": -3.89312744140625,
      "accuracy": 0.75,
      "epoch": 0.0875,
      "step": 35
    },
    {
      "Batch Mean": -4.010528564453125,
      "accuracy": 0.8125,
      "epoch": 0.0875,
      "step": 35
    },
    {
      "Batch Mean": -3.480602264404297,
      "accuracy": 0.8125,
      "epoch": 0.0875,
      "step": 35
    },
    {
      "epoch": 0.09,
      "grad_norm": 10.730036735534668,
      "learning_rate": 2.873684210526316e-06,
      "loss": 0.4968,
      "step": 36
    },
    {
      "Batch Mean": -3.891571044921875,
      "accuracy": 0.78125,
      "epoch": 0.09,
      "step": 36
    },
    {
      "Batch Mean": -3.62432861328125,
      "accuracy": 0.8125,
      "epoch": 0.09,
      "step": 36
    },
    {
      "Batch Mean": -3.650390625,
      "accuracy": 0.65625,
      "epoch": 0.09,
      "step": 36
    },
    {
      "Batch Mean": -3.6742515563964844,
      "accuracy": 0.75,
      "epoch": 0.09,
      "step": 36
    },
    {
      "epoch": 0.0925,
      "grad_norm": 8.428532600402832,
      "learning_rate": 2.8657894736842103e-06,
      "loss": 0.4515,
      "step": 37
    },
    {
      "Batch Mean": -3.5828857421875,
      "accuracy": 0.71875,
      "epoch": 0.0925,
      "step": 37
    },
    {
      "Batch Mean": -3.687744140625,
      "accuracy": 0.84375,
      "epoch": 0.0925,
      "step": 37
    },
    {
      "Batch Mean": -3.60595703125,
      "accuracy": 0.8125,
      "epoch": 0.0925,
      "step": 37
    },
    {
      "Batch Mean": -4.343017578125,
      "accuracy": 0.65625,
      "epoch": 0.0925,
      "step": 37
    },
    {
      "epoch": 0.095,
      "grad_norm": 8.982612609863281,
      "learning_rate": 2.857894736842105e-06,
      "loss": 0.4853,
      "step": 38
    },
    {
      "Batch Mean": -4.35125732421875,
      "accuracy": 0.875,
      "epoch": 0.095,
      "step": 38
    },
    {
      "Batch Mean": -4.2618408203125,
      "accuracy": 0.78125,
      "epoch": 0.095,
      "step": 38
    },
    {
      "Batch Mean": -3.67474365234375,
      "accuracy": 0.8125,
      "epoch": 0.095,
      "step": 38
    },
    {
      "Batch Mean": -4.13818359375,
      "accuracy": 0.6875,
      "epoch": 0.095,
      "step": 38
    },
    {
      "epoch": 0.0975,
      "grad_norm": 8.4310302734375,
      "learning_rate": 2.85e-06,
      "loss": 0.4476,
      "step": 39
    },
    {
      "Batch Mean": -4.23419189453125,
      "accuracy": 0.78125,
      "epoch": 0.0975,
      "step": 39
    },
    {
      "Batch Mean": -4.0997314453125,
      "accuracy": 0.59375,
      "epoch": 0.0975,
      "step": 39
    },
    {
      "Batch Mean": -4.07379150390625,
      "accuracy": 0.71875,
      "epoch": 0.0975,
      "step": 39
    },
    {
      "Batch Mean": -4.32989501953125,
      "accuracy": 1.0,
      "epoch": 0.0975,
      "step": 39
    },
    {
      "epoch": 0.1,
      "grad_norm": 7.986425399780273,
      "learning_rate": 2.8421052631578946e-06,
      "loss": 0.4334,
      "step": 40
    },
    {
      "Batch Mean": -4.50189208984375,
      "accuracy": 0.65625,
      "epoch": 0.1,
      "step": 40
    },
    {
      "Batch Mean": -3.809326171875,
      "accuracy": 0.75,
      "epoch": 0.1,
      "step": 40
    },
    {
      "Batch Mean": -4.21588134765625,
      "accuracy": 0.875,
      "epoch": 0.1,
      "step": 40
    },
    {
      "Batch Mean": -3.85302734375,
      "accuracy": 0.90625,
      "epoch": 0.1,
      "step": 40
    },
    {
      "epoch": 0.1025,
      "grad_norm": 8.613731384277344,
      "learning_rate": 2.8342105263157897e-06,
      "loss": 0.4477,
      "step": 41
    },
    {
      "Batch Mean": -3.824188232421875,
      "accuracy": 0.875,
      "epoch": 0.1025,
      "step": 41
    },
    {
      "Batch Mean": -3.658496856689453,
      "accuracy": 0.84375,
      "epoch": 0.1025,
      "step": 41
    },
    {
      "Batch Mean": -3.964691162109375,
      "accuracy": 0.84375,
      "epoch": 0.1025,
      "step": 41
    },
    {
      "Batch Mean": -3.8968505859375,
      "accuracy": 0.75,
      "epoch": 0.1025,
      "step": 41
    },
    {
      "epoch": 0.105,
      "grad_norm": 8.035860061645508,
      "learning_rate": 2.8263157894736845e-06,
      "loss": 0.3786,
      "step": 42
    },
    {
      "Batch Mean": -4.138427734375,
      "accuracy": 0.8125,
      "epoch": 0.105,
      "step": 42
    },
    {
      "Batch Mean": -3.946533203125,
      "accuracy": 0.78125,
      "epoch": 0.105,
      "step": 42
    },
    {
      "Batch Mean": -4.2305908203125,
      "accuracy": 0.8125,
      "epoch": 0.105,
      "step": 42
    },
    {
      "Batch Mean": -3.92535400390625,
      "accuracy": 0.875,
      "epoch": 0.105,
      "step": 42
    },
    {
      "epoch": 0.1075,
      "grad_norm": 8.79111385345459,
      "learning_rate": 2.8184210526315792e-06,
      "loss": 0.3971,
      "step": 43
    },
    {
      "Batch Mean": -3.7923431396484375,
      "accuracy": 0.65625,
      "epoch": 0.1075,
      "step": 43
    },
    {
      "Batch Mean": -3.5310258865356445,
      "accuracy": 0.84375,
      "epoch": 0.1075,
      "step": 43
    },
    {
      "Batch Mean": -3.6854248046875,
      "accuracy": 0.875,
      "epoch": 0.1075,
      "step": 43
    },
    {
      "Batch Mean": -3.468961715698242,
      "accuracy": 0.78125,
      "epoch": 0.1075,
      "step": 43
    },
    {
      "epoch": 0.11,
      "grad_norm": 10.596081733703613,
      "learning_rate": 2.810526315789474e-06,
      "loss": 0.4792,
      "step": 44
    },
    {
      "Batch Mean": -3.435342788696289,
      "accuracy": 0.875,
      "epoch": 0.11,
      "step": 44
    },
    {
      "Batch Mean": -3.907958984375,
      "accuracy": 0.71875,
      "epoch": 0.11,
      "step": 44
    },
    {
      "Batch Mean": -3.551055908203125,
      "accuracy": 0.71875,
      "epoch": 0.11,
      "step": 44
    },
    {
      "Batch Mean": -3.1349716186523438,
      "accuracy": 0.90625,
      "epoch": 0.11,
      "step": 44
    },
    {
      "epoch": 0.1125,
      "grad_norm": 8.473621368408203,
      "learning_rate": 2.8026315789473687e-06,
      "loss": 0.438,
      "step": 45
    },
    {
      "Batch Mean": -3.0492935180664062,
      "accuracy": 0.90625,
      "epoch": 0.1125,
      "step": 45
    },
    {
      "Batch Mean": -3.042724609375,
      "accuracy": 0.875,
      "epoch": 0.1125,
      "step": 45
    },
    {
      "Batch Mean": -3.8280296325683594,
      "accuracy": 0.90625,
      "epoch": 0.1125,
      "step": 45
    },
    {
      "Batch Mean": -3.4905853271484375,
      "accuracy": 0.9375,
      "epoch": 0.1125,
      "step": 45
    },
    {
      "epoch": 0.115,
      "grad_norm": 7.537892818450928,
      "learning_rate": 2.7947368421052635e-06,
      "loss": 0.2883,
      "step": 46
    },
    {
      "Batch Mean": -2.8027992248535156,
      "accuracy": 0.625,
      "epoch": 0.115,
      "step": 46
    },
    {
      "Batch Mean": -2.9038753509521484,
      "accuracy": 0.875,
      "epoch": 0.115,
      "step": 46
    },
    {
      "Batch Mean": -3.4294891357421875,
      "accuracy": 0.71875,
      "epoch": 0.115,
      "step": 46
    },
    {
      "Batch Mean": -2.933074951171875,
      "accuracy": 0.75,
      "epoch": 0.115,
      "step": 46
    },
    {
      "epoch": 0.1175,
      "grad_norm": 12.333584785461426,
      "learning_rate": 2.7868421052631578e-06,
      "loss": 0.5348,
      "step": 47
    },
    {
      "Batch Mean": -2.7773094177246094,
      "accuracy": 0.6875,
      "epoch": 0.1175,
      "step": 47
    },
    {
      "Batch Mean": -2.03399658203125,
      "accuracy": 0.84375,
      "epoch": 0.1175,
      "step": 47
    },
    {
      "Batch Mean": -2.9137496948242188,
      "accuracy": 0.84375,
      "epoch": 0.1175,
      "step": 47
    },
    {
      "Batch Mean": -3.201608657836914,
      "accuracy": 0.875,
      "epoch": 0.1175,
      "step": 47
    },
    {
      "epoch": 0.12,
      "grad_norm": 10.547114372253418,
      "learning_rate": 2.7789473684210525e-06,
      "loss": 0.4621,
      "step": 48
    },
    {
      "Batch Mean": -2.280449867248535,
      "accuracy": 0.78125,
      "epoch": 0.12,
      "step": 48
    },
    {
      "Batch Mean": -2.9716339111328125,
      "accuracy": 0.875,
      "epoch": 0.12,
      "step": 48
    },
    {
      "Batch Mean": -2.959075927734375,
      "accuracy": 0.71875,
      "epoch": 0.12,
      "step": 48
    },
    {
      "Batch Mean": -2.8322715759277344,
      "accuracy": 0.71875,
      "epoch": 0.12,
      "step": 48
    },
    {
      "epoch": 0.1225,
      "grad_norm": 12.012454986572266,
      "learning_rate": 2.7710526315789473e-06,
      "loss": 0.489,
      "step": 49
    },
    {
      "Batch Mean": -3.118091583251953,
      "accuracy": 0.8125,
      "epoch": 0.1225,
      "step": 49
    },
    {
      "Batch Mean": -2.2162322998046875,
      "accuracy": 0.75,
      "epoch": 0.1225,
      "step": 49
    },
    {
      "Batch Mean": -2.1783409118652344,
      "accuracy": 0.59375,
      "epoch": 0.1225,
      "step": 49
    },
    {
      "Batch Mean": -2.5770187377929688,
      "accuracy": 0.78125,
      "epoch": 0.1225,
      "step": 49
    },
    {
      "epoch": 0.125,
      "grad_norm": 13.930920600891113,
      "learning_rate": 2.763157894736842e-06,
      "loss": 0.5659,
      "step": 50
    },
    {
      "Batch Mean": -2.640003204345703,
      "accuracy": 0.875,
      "epoch": 0.125,
      "step": 50
    },
    {
      "Batch Mean": -3.2294464111328125,
      "accuracy": 0.84375,
      "epoch": 0.125,
      "step": 50
    },
    {
      "Batch Mean": -3.3978271484375,
      "accuracy": 0.65625,
      "epoch": 0.125,
      "step": 50
    },
    {
      "Batch Mean": -2.858466148376465,
      "accuracy": 0.8125,
      "epoch": 0.125,
      "step": 50
    },
    {
      "epoch": 0.1275,
      "grad_norm": 11.848394393920898,
      "learning_rate": 2.7552631578947368e-06,
      "loss": 0.5105,
      "step": 51
    },
    {
      "Batch Mean": -2.7342262268066406,
      "accuracy": 0.6875,
      "epoch": 0.1275,
      "step": 51
    },
    {
      "Batch Mean": -3.48663330078125,
      "accuracy": 0.65625,
      "epoch": 0.1275,
      "step": 51
    },
    {
      "Batch Mean": -3.0019655227661133,
      "accuracy": 0.71875,
      "epoch": 0.1275,
      "step": 51
    },
    {
      "Batch Mean": -3.550537109375,
      "accuracy": 0.6875,
      "epoch": 0.1275,
      "step": 51
    },
    {
      "epoch": 0.13,
      "grad_norm": 10.681953430175781,
      "learning_rate": 2.7473684210526315e-06,
      "loss": 0.5472,
      "step": 52
    },
    {
      "Batch Mean": -3.7133026123046875,
      "accuracy": 0.71875,
      "epoch": 0.13,
      "step": 52
    },
    {
      "Batch Mean": -3.6298141479492188,
      "accuracy": 0.71875,
      "epoch": 0.13,
      "step": 52
    },
    {
      "Batch Mean": -4.31036376953125,
      "accuracy": 0.8125,
      "epoch": 0.13,
      "step": 52
    },
    {
      "Batch Mean": -3.61016845703125,
      "accuracy": 0.71875,
      "epoch": 0.13,
      "step": 52
    },
    {
      "epoch": 0.1325,
      "grad_norm": 7.710223197937012,
      "learning_rate": 2.7394736842105263e-06,
      "loss": 0.5108,
      "step": 53
    },
    {
      "Batch Mean": -4.007965087890625,
      "accuracy": 0.75,
      "epoch": 0.1325,
      "step": 53
    },
    {
      "Batch Mean": -3.752777099609375,
      "accuracy": 0.84375,
      "epoch": 0.1325,
      "step": 53
    },
    {
      "Batch Mean": -3.687549591064453,
      "accuracy": 0.71875,
      "epoch": 0.1325,
      "step": 53
    },
    {
      "Batch Mean": -3.5942840576171875,
      "accuracy": 0.75,
      "epoch": 0.1325,
      "step": 53
    },
    {
      "epoch": 0.135,
      "grad_norm": 6.719354152679443,
      "learning_rate": 2.7315789473684214e-06,
      "loss": 0.4691,
      "step": 54
    },
    {
      "Batch Mean": -4.132904052734375,
      "accuracy": 0.90625,
      "epoch": 0.135,
      "step": 54
    },
    {
      "Batch Mean": -4.61541748046875,
      "accuracy": 0.78125,
      "epoch": 0.135,
      "step": 54
    },
    {
      "Batch Mean": -4.22991943359375,
      "accuracy": 0.8125,
      "epoch": 0.135,
      "step": 54
    },
    {
      "Batch Mean": -4.14007568359375,
      "accuracy": 0.6875,
      "epoch": 0.135,
      "step": 54
    },
    {
      "epoch": 0.1375,
      "grad_norm": 6.091842174530029,
      "learning_rate": 2.723684210526316e-06,
      "loss": 0.4276,
      "step": 55
    },
    {
      "Batch Mean": -4.385498046875,
      "accuracy": 0.9375,
      "epoch": 0.1375,
      "step": 55
    },
    {
      "Batch Mean": -4.61492919921875,
      "accuracy": 0.625,
      "epoch": 0.1375,
      "step": 55
    },
    {
      "Batch Mean": -4.2962646484375,
      "accuracy": 0.8125,
      "epoch": 0.1375,
      "step": 55
    },
    {
      "Batch Mean": -4.18157958984375,
      "accuracy": 0.6875,
      "epoch": 0.1375,
      "step": 55
    },
    {
      "epoch": 0.14,
      "grad_norm": 6.7918620109558105,
      "learning_rate": 2.715789473684211e-06,
      "loss": 0.4102,
      "step": 56
    },
    {
      "Batch Mean": -3.7819137573242188,
      "accuracy": 0.78125,
      "epoch": 0.14,
      "step": 56
    },
    {
      "Batch Mean": -3.8838958740234375,
      "accuracy": 0.90625,
      "epoch": 0.14,
      "step": 56
    },
    {
      "Batch Mean": -4.2425537109375,
      "accuracy": 0.625,
      "epoch": 0.14,
      "step": 56
    },
    {
      "Batch Mean": -4.3662109375,
      "accuracy": 0.90625,
      "epoch": 0.14,
      "step": 56
    },
    {
      "epoch": 0.1425,
      "grad_norm": 7.213972568511963,
      "learning_rate": 2.7078947368421052e-06,
      "loss": 0.457,
      "step": 57
    },
    {
      "Batch Mean": -3.624969482421875,
      "accuracy": 0.9375,
      "epoch": 0.1425,
      "step": 57
    },
    {
      "Batch Mean": -4.58807373046875,
      "accuracy": 0.71875,
      "epoch": 0.1425,
      "step": 57
    },
    {
      "Batch Mean": -4.09814453125,
      "accuracy": 0.75,
      "epoch": 0.1425,
      "step": 57
    },
    {
      "Batch Mean": -3.97235107421875,
      "accuracy": 0.8125,
      "epoch": 0.1425,
      "step": 57
    },
    {
      "epoch": 0.145,
      "grad_norm": 5.783757209777832,
      "learning_rate": 2.7e-06,
      "loss": 0.4118,
      "step": 58
    },
    {
      "Batch Mean": -3.2208633422851562,
      "accuracy": 0.90625,
      "epoch": 0.145,
      "step": 58
    },
    {
      "Batch Mean": -3.5696563720703125,
      "accuracy": 0.84375,
      "epoch": 0.145,
      "step": 58
    },
    {
      "Batch Mean": -3.245067596435547,
      "accuracy": 0.6875,
      "epoch": 0.145,
      "step": 58
    },
    {
      "Batch Mean": -3.427947998046875,
      "accuracy": 0.9375,
      "epoch": 0.145,
      "step": 58
    },
    {
      "epoch": 0.1475,
      "grad_norm": 6.49120569229126,
      "learning_rate": 2.6921052631578947e-06,
      "loss": 0.4278,
      "step": 59
    },
    {
      "Batch Mean": -3.1783447265625,
      "accuracy": 0.625,
      "epoch": 0.1475,
      "step": 59
    },
    {
      "Batch Mean": -3.128032684326172,
      "accuracy": 0.875,
      "epoch": 0.1475,
      "step": 59
    },
    {
      "Batch Mean": -3.291412353515625,
      "accuracy": 0.75,
      "epoch": 0.1475,
      "step": 59
    },
    {
      "Batch Mean": -3.10302734375,
      "accuracy": 0.71875,
      "epoch": 0.1475,
      "step": 59
    },
    {
      "epoch": 0.15,
      "grad_norm": 7.3600850105285645,
      "learning_rate": 2.6842105263157895e-06,
      "loss": 0.4719,
      "step": 60
    },
    {
      "Batch Mean": -2.8273773193359375,
      "accuracy": 0.75,
      "epoch": 0.15,
      "step": 60
    },
    {
      "Batch Mean": -3.152629852294922,
      "accuracy": 0.71875,
      "epoch": 0.15,
      "step": 60
    },
    {
      "Batch Mean": -2.8756637573242188,
      "accuracy": 0.6875,
      "epoch": 0.15,
      "step": 60
    },
    {
      "Batch Mean": -2.9803466796875,
      "accuracy": 0.84375,
      "epoch": 0.15,
      "step": 60
    },
    {
      "epoch": 0.1525,
      "grad_norm": 6.642537593841553,
      "learning_rate": 2.6763157894736842e-06,
      "loss": 0.4796,
      "step": 61
    },
    {
      "Batch Mean": -2.676004409790039,
      "accuracy": 0.84375,
      "epoch": 0.1525,
      "step": 61
    },
    {
      "Batch Mean": -2.830108642578125,
      "accuracy": 0.84375,
      "epoch": 0.1525,
      "step": 61
    },
    {
      "Batch Mean": -2.6599464416503906,
      "accuracy": 0.875,
      "epoch": 0.1525,
      "step": 61
    },
    {
      "Batch Mean": -2.9864349365234375,
      "accuracy": 0.8125,
      "epoch": 0.1525,
      "step": 61
    },
    {
      "epoch": 0.155,
      "grad_norm": 6.946270942687988,
      "learning_rate": 2.668421052631579e-06,
      "loss": 0.3668,
      "step": 62
    },
    {
      "Batch Mean": -2.9581680297851562,
      "accuracy": 0.875,
      "epoch": 0.155,
      "step": 62
    },
    {
      "Batch Mean": -2.5045700073242188,
      "accuracy": 0.84375,
      "epoch": 0.155,
      "step": 62
    },
    {
      "Batch Mean": -3.3170013427734375,
      "accuracy": 0.75,
      "epoch": 0.155,
      "step": 62
    },
    {
      "Batch Mean": -3.1421661376953125,
      "accuracy": 0.75,
      "epoch": 0.155,
      "step": 62
    },
    {
      "epoch": 0.1575,
      "grad_norm": 7.357174396514893,
      "learning_rate": 2.6605263157894737e-06,
      "loss": 0.4197,
      "step": 63
    },
    {
      "Batch Mean": -2.3607635498046875,
      "accuracy": 0.78125,
      "epoch": 0.1575,
      "step": 63
    },
    {
      "Batch Mean": -3.1617660522460938,
      "accuracy": 0.90625,
      "epoch": 0.1575,
      "step": 63
    },
    {
      "Batch Mean": -3.3164749145507812,
      "accuracy": 0.8125,
      "epoch": 0.1575,
      "step": 63
    },
    {
      "Batch Mean": -3.1099395751953125,
      "accuracy": 0.8125,
      "epoch": 0.1575,
      "step": 63
    },
    {
      "epoch": 0.16,
      "grad_norm": 7.538585662841797,
      "learning_rate": 2.6526315789473685e-06,
      "loss": 0.4143,
      "step": 64
    },
    {
      "Batch Mean": -2.6910171508789062,
      "accuracy": 0.84375,
      "epoch": 0.16,
      "step": 64
    },
    {
      "Batch Mean": -3.074167013168335,
      "accuracy": 0.8125,
      "epoch": 0.16,
      "step": 64
    },
    {
      "Batch Mean": -3.1822357177734375,
      "accuracy": 0.8125,
      "epoch": 0.16,
      "step": 64
    },
    {
      "Batch Mean": -3.3923683166503906,
      "accuracy": 0.78125,
      "epoch": 0.16,
      "step": 64
    },
    {
      "epoch": 0.1625,
      "grad_norm": 7.79094123840332,
      "learning_rate": 2.644736842105263e-06,
      "loss": 0.4276,
      "step": 65
    },
    {
      "Batch Mean": -2.9289722442626953,
      "accuracy": 0.75,
      "epoch": 0.1625,
      "step": 65
    },
    {
      "Batch Mean": -3.164053440093994,
      "accuracy": 0.75,
      "epoch": 0.1625,
      "step": 65
    },
    {
      "Batch Mean": -3.19818115234375,
      "accuracy": 0.71875,
      "epoch": 0.1625,
      "step": 65
    },
    {
      "Batch Mean": -3.493072509765625,
      "accuracy": 0.9375,
      "epoch": 0.1625,
      "step": 65
    },
    {
      "epoch": 0.165,
      "grad_norm": 8.591028213500977,
      "learning_rate": 2.636842105263158e-06,
      "loss": 0.5094,
      "step": 66
    },
    {
      "Batch Mean": -3.191915512084961,
      "accuracy": 0.75,
      "epoch": 0.165,
      "step": 66
    },
    {
      "Batch Mean": -3.100576400756836,
      "accuracy": 0.8125,
      "epoch": 0.165,
      "step": 66
    },
    {
      "Batch Mean": -3.4085922241210938,
      "accuracy": 0.96875,
      "epoch": 0.165,
      "step": 66
    },
    {
      "Batch Mean": -3.1348419189453125,
      "accuracy": 0.8125,
      "epoch": 0.165,
      "step": 66
    },
    {
      "epoch": 0.1675,
      "grad_norm": 8.065413475036621,
      "learning_rate": 2.6289473684210527e-06,
      "loss": 0.4552,
      "step": 67
    },
    {
      "Batch Mean": -4.05078125,
      "accuracy": 0.75,
      "epoch": 0.1675,
      "step": 67
    },
    {
      "Batch Mean": -3.196155548095703,
      "accuracy": 0.8125,
      "epoch": 0.1675,
      "step": 67
    },
    {
      "Batch Mean": -4.133514404296875,
      "accuracy": 0.71875,
      "epoch": 0.1675,
      "step": 67
    },
    {
      "Batch Mean": -3.7607345581054688,
      "accuracy": 0.84375,
      "epoch": 0.1675,
      "step": 67
    },
    {
      "epoch": 0.17,
      "grad_norm": 8.609820365905762,
      "learning_rate": 2.6210526315789474e-06,
      "loss": 0.4536,
      "step": 68
    },
    {
      "Batch Mean": -3.8547210693359375,
      "accuracy": 0.90625,
      "epoch": 0.17,
      "step": 68
    },
    {
      "Batch Mean": -3.7351226806640625,
      "accuracy": 0.71875,
      "epoch": 0.17,
      "step": 68
    },
    {
      "Batch Mean": -4.212074279785156,
      "accuracy": 0.78125,
      "epoch": 0.17,
      "step": 68
    },
    {
      "Batch Mean": -3.53668212890625,
      "accuracy": 0.84375,
      "epoch": 0.17,
      "step": 68
    },
    {
      "epoch": 0.1725,
      "grad_norm": 7.895672798156738,
      "learning_rate": 2.613157894736842e-06,
      "loss": 0.3879,
      "step": 69
    },
    {
      "Batch Mean": -4.088775634765625,
      "accuracy": 0.75,
      "epoch": 0.1725,
      "step": 69
    },
    {
      "Batch Mean": -3.75189208984375,
      "accuracy": 0.84375,
      "epoch": 0.1725,
      "step": 69
    },
    {
      "Batch Mean": -4.203272819519043,
      "accuracy": 0.8125,
      "epoch": 0.1725,
      "step": 69
    },
    {
      "Batch Mean": -3.900634765625,
      "accuracy": 0.75,
      "epoch": 0.1725,
      "step": 69
    },
    {
      "epoch": 0.175,
      "grad_norm": 7.224689483642578,
      "learning_rate": 2.605263157894737e-06,
      "loss": 0.422,
      "step": 70
    },
    {
      "Batch Mean": -4.03101921081543,
      "accuracy": 0.75,
      "epoch": 0.175,
      "step": 70
    },
    {
      "Batch Mean": -3.585662841796875,
      "accuracy": 0.78125,
      "epoch": 0.175,
      "step": 70
    },
    {
      "Batch Mean": -3.6973724365234375,
      "accuracy": 0.875,
      "epoch": 0.175,
      "step": 70
    },
    {
      "Batch Mean": -4.1641082763671875,
      "accuracy": 0.65625,
      "epoch": 0.175,
      "step": 70
    },
    {
      "epoch": 0.1775,
      "grad_norm": 6.848752498626709,
      "learning_rate": 2.5973684210526317e-06,
      "loss": 0.4386,
      "step": 71
    },
    {
      "Batch Mean": -3.96783447265625,
      "accuracy": 0.75,
      "epoch": 0.1775,
      "step": 71
    },
    {
      "Batch Mean": -4.127544403076172,
      "accuracy": 0.84375,
      "epoch": 0.1775,
      "step": 71
    },
    {
      "Batch Mean": -4.292510986328125,
      "accuracy": 0.53125,
      "epoch": 0.1775,
      "step": 71
    },
    {
      "Batch Mean": -3.774566650390625,
      "accuracy": 0.90625,
      "epoch": 0.1775,
      "step": 71
    },
    {
      "epoch": 0.18,
      "grad_norm": 6.717357635498047,
      "learning_rate": 2.5894736842105264e-06,
      "loss": 0.4317,
      "step": 72
    },
    {
      "Batch Mean": -4.0881805419921875,
      "accuracy": 0.9375,
      "epoch": 0.18,
      "step": 72
    },
    {
      "Batch Mean": -4.007743835449219,
      "accuracy": 0.84375,
      "epoch": 0.18,
      "step": 72
    },
    {
      "Batch Mean": -4.3395538330078125,
      "accuracy": 0.78125,
      "epoch": 0.18,
      "step": 72
    },
    {
      "Batch Mean": -3.7432098388671875,
      "accuracy": 0.90625,
      "epoch": 0.18,
      "step": 72
    },
    {
      "epoch": 0.1825,
      "grad_norm": 6.629477500915527,
      "learning_rate": 2.581578947368421e-06,
      "loss": 0.3347,
      "step": 73
    },
    {
      "Batch Mean": -4.6607513427734375,
      "accuracy": 0.6875,
      "epoch": 0.1825,
      "step": 73
    },
    {
      "Batch Mean": -4.265266418457031,
      "accuracy": 0.875,
      "epoch": 0.1825,
      "step": 73
    },
    {
      "Batch Mean": -3.5947265625,
      "accuracy": 0.8125,
      "epoch": 0.1825,
      "step": 73
    },
    {
      "Batch Mean": -4.227779388427734,
      "accuracy": 0.78125,
      "epoch": 0.1825,
      "step": 73
    },
    {
      "epoch": 0.185,
      "grad_norm": 6.081233501434326,
      "learning_rate": 2.573684210526316e-06,
      "loss": 0.3813,
      "step": 74
    },
    {
      "Batch Mean": -3.6422882080078125,
      "accuracy": 0.6875,
      "epoch": 0.185,
      "step": 74
    },
    {
      "Batch Mean": -4.190940856933594,
      "accuracy": 0.6875,
      "epoch": 0.185,
      "step": 74
    },
    {
      "Batch Mean": -3.8231201171875,
      "accuracy": 0.78125,
      "epoch": 0.185,
      "step": 74
    },
    {
      "Batch Mean": -4.39495849609375,
      "accuracy": 0.8125,
      "epoch": 0.185,
      "step": 74
    },
    {
      "epoch": 0.1875,
      "grad_norm": 7.641036033630371,
      "learning_rate": 2.5657894736842107e-06,
      "loss": 0.5041,
      "step": 75
    },
    {
      "Batch Mean": -3.682525634765625,
      "accuracy": 0.78125,
      "epoch": 0.1875,
      "step": 75
    },
    {
      "Batch Mean": -3.192474365234375,
      "accuracy": 0.6875,
      "epoch": 0.1875,
      "step": 75
    },
    {
      "Batch Mean": -2.80194091796875,
      "accuracy": 0.875,
      "epoch": 0.1875,
      "step": 75
    },
    {
      "Batch Mean": -3.7286148071289062,
      "accuracy": 0.78125,
      "epoch": 0.1875,
      "step": 75
    },
    {
      "epoch": 0.19,
      "grad_norm": 6.7662200927734375,
      "learning_rate": 2.5578947368421054e-06,
      "loss": 0.4441,
      "step": 76
    },
    {
      "Batch Mean": -3.60430908203125,
      "accuracy": 0.78125,
      "epoch": 0.19,
      "step": 76
    },
    {
      "Batch Mean": -3.5869674682617188,
      "accuracy": 0.875,
      "epoch": 0.19,
      "step": 76
    },
    {
      "Batch Mean": -3.536773681640625,
      "accuracy": 0.78125,
      "epoch": 0.19,
      "step": 76
    },
    {
      "Batch Mean": -3.7320709228515625,
      "accuracy": 0.8125,
      "epoch": 0.19,
      "step": 76
    },
    {
      "epoch": 0.1925,
      "grad_norm": 7.16535758972168,
      "learning_rate": 2.55e-06,
      "loss": 0.4266,
      "step": 77
    },
    {
      "Batch Mean": -2.750619411468506,
      "accuracy": 0.875,
      "epoch": 0.1925,
      "step": 77
    },
    {
      "Batch Mean": -3.563629150390625,
      "accuracy": 0.625,
      "epoch": 0.1925,
      "step": 77
    },
    {
      "Batch Mean": -3.890869140625,
      "accuracy": 0.8125,
      "epoch": 0.1925,
      "step": 77
    },
    {
      "Batch Mean": -3.540557861328125,
      "accuracy": 0.90625,
      "epoch": 0.1925,
      "step": 77
    },
    {
      "epoch": 0.195,
      "grad_norm": 8.778853416442871,
      "learning_rate": 2.542105263157895e-06,
      "loss": 0.4598,
      "step": 78
    },
    {
      "Batch Mean": -3.7507858276367188,
      "accuracy": 0.84375,
      "epoch": 0.195,
      "step": 78
    },
    {
      "Batch Mean": -4.087003707885742,
      "accuracy": 0.875,
      "epoch": 0.195,
      "step": 78
    },
    {
      "Batch Mean": -3.482025146484375,
      "accuracy": 0.75,
      "epoch": 0.195,
      "step": 78
    },
    {
      "Batch Mean": -3.1190052032470703,
      "accuracy": 0.9375,
      "epoch": 0.195,
      "step": 78
    },
    {
      "epoch": 0.1975,
      "grad_norm": 5.622673034667969,
      "learning_rate": 2.5342105263157892e-06,
      "loss": 0.3767,
      "step": 79
    },
    {
      "Batch Mean": -4.231414794921875,
      "accuracy": 0.65625,
      "epoch": 0.1975,
      "step": 79
    },
    {
      "Batch Mean": -3.4138107299804688,
      "accuracy": 0.78125,
      "epoch": 0.1975,
      "step": 79
    },
    {
      "Batch Mean": -3.5726470947265625,
      "accuracy": 0.78125,
      "epoch": 0.1975,
      "step": 79
    },
    {
      "Batch Mean": -3.738006591796875,
      "accuracy": 0.84375,
      "epoch": 0.1975,
      "step": 79
    },
    {
      "epoch": 0.2,
      "grad_norm": 6.790273189544678,
      "learning_rate": 2.526315789473684e-06,
      "loss": 0.437,
      "step": 80
    },
    {
      "Batch Mean": -3.6096343994140625,
      "accuracy": 0.71875,
      "epoch": 0.2,
      "step": 80
    },
    {
      "Batch Mean": -3.303373336791992,
      "accuracy": 0.9375,
      "epoch": 0.2,
      "step": 80
    },
    {
      "Batch Mean": -3.778522491455078,
      "accuracy": 0.90625,
      "epoch": 0.2,
      "step": 80
    },
    {
      "Batch Mean": -3.6217997074127197,
      "accuracy": 0.875,
      "epoch": 0.2,
      "step": 80
    },
    {
      "epoch": 0.2025,
      "grad_norm": 6.186145305633545,
      "learning_rate": 2.5184210526315787e-06,
      "loss": 0.3668,
      "step": 81
    },
    {
      "Batch Mean": -2.8534088134765625,
      "accuracy": 0.71875,
      "epoch": 0.2025,
      "step": 81
    },
    {
      "Batch Mean": -3.2958984375,
      "accuracy": 0.6875,
      "epoch": 0.2025,
      "step": 81
    },
    {
      "Batch Mean": -3.643817901611328,
      "accuracy": 0.75,
      "epoch": 0.2025,
      "step": 81
    },
    {
      "Batch Mean": -3.3983230590820312,
      "accuracy": 0.90625,
      "epoch": 0.2025,
      "step": 81
    },
    {
      "epoch": 0.205,
      "grad_norm": 6.6807379722595215,
      "learning_rate": 2.510526315789474e-06,
      "loss": 0.4352,
      "step": 82
    },
    {
      "Batch Mean": -3.2677268981933594,
      "accuracy": 0.75,
      "epoch": 0.205,
      "step": 82
    },
    {
      "Batch Mean": -3.808116912841797,
      "accuracy": 0.8125,
      "epoch": 0.205,
      "step": 82
    },
    {
      "Batch Mean": -3.621933937072754,
      "accuracy": 0.78125,
      "epoch": 0.205,
      "step": 82
    },
    {
      "Batch Mean": -3.544036865234375,
      "accuracy": 0.6875,
      "epoch": 0.205,
      "step": 82
    },
    {
      "epoch": 0.2075,
      "grad_norm": 6.687833309173584,
      "learning_rate": 2.5026315789473686e-06,
      "loss": 0.4091,
      "step": 83
    },
    {
      "Batch Mean": -3.397430419921875,
      "accuracy": 0.9375,
      "epoch": 0.2075,
      "step": 83
    },
    {
      "Batch Mean": -3.124340057373047,
      "accuracy": 0.875,
      "epoch": 0.2075,
      "step": 83
    },
    {
      "Batch Mean": -3.5037002563476562,
      "accuracy": 0.90625,
      "epoch": 0.2075,
      "step": 83
    },
    {
      "Batch Mean": -2.901216506958008,
      "accuracy": 0.75,
      "epoch": 0.2075,
      "step": 83
    },
    {
      "epoch": 0.21,
      "grad_norm": 6.103653907775879,
      "learning_rate": 2.4947368421052634e-06,
      "loss": 0.3172,
      "step": 84
    },
    {
      "Batch Mean": -3.0877227783203125,
      "accuracy": 0.90625,
      "epoch": 0.21,
      "step": 84
    },
    {
      "Batch Mean": -2.7988204956054688,
      "accuracy": 0.78125,
      "epoch": 0.21,
      "step": 84
    },
    {
      "Batch Mean": -3.3592453002929688,
      "accuracy": 0.8125,
      "epoch": 0.21,
      "step": 84
    },
    {
      "Batch Mean": -2.706587314605713,
      "accuracy": 0.6875,
      "epoch": 0.21,
      "step": 84
    },
    {
      "epoch": 0.2125,
      "grad_norm": 6.470891952514648,
      "learning_rate": 2.486842105263158e-06,
      "loss": 0.3878,
      "step": 85
    },
    {
      "Batch Mean": -2.30181884765625,
      "accuracy": 0.875,
      "epoch": 0.2125,
      "step": 85
    },
    {
      "Batch Mean": -2.7299463748931885,
      "accuracy": 0.71875,
      "epoch": 0.2125,
      "step": 85
    },
    {
      "Batch Mean": -2.6263694763183594,
      "accuracy": 0.6875,
      "epoch": 0.2125,
      "step": 85
    },
    {
      "Batch Mean": -2.6631546020507812,
      "accuracy": 0.78125,
      "epoch": 0.2125,
      "step": 85
    },
    {
      "epoch": 0.215,
      "grad_norm": 7.898273944854736,
      "learning_rate": 2.478947368421053e-06,
      "loss": 0.474,
      "step": 86
    },
    {
      "Batch Mean": -2.9964141845703125,
      "accuracy": 0.875,
      "epoch": 0.215,
      "step": 86
    },
    {
      "Batch Mean": -1.6867389678955078,
      "accuracy": 0.75,
      "epoch": 0.215,
      "step": 86
    },
    {
      "Batch Mean": -2.084423065185547,
      "accuracy": 0.875,
      "epoch": 0.215,
      "step": 86
    },
    {
      "Batch Mean": -2.3247833251953125,
      "accuracy": 0.90625,
      "epoch": 0.215,
      "step": 86
    },
    {
      "epoch": 0.2175,
      "grad_norm": 7.01542329788208,
      "learning_rate": 2.4710526315789476e-06,
      "loss": 0.354,
      "step": 87
    },
    {
      "Batch Mean": -1.62408447265625,
      "accuracy": 0.6875,
      "epoch": 0.2175,
      "step": 87
    },
    {
      "Batch Mean": -1.8180961608886719,
      "accuracy": 0.75,
      "epoch": 0.2175,
      "step": 87
    },
    {
      "Batch Mean": -2.39483642578125,
      "accuracy": 0.75,
      "epoch": 0.2175,
      "step": 87
    },
    {
      "Batch Mean": -2.1982975006103516,
      "accuracy": 0.9375,
      "epoch": 0.2175,
      "step": 87
    },
    {
      "epoch": 0.22,
      "grad_norm": 9.38858413696289,
      "learning_rate": 2.4631578947368424e-06,
      "loss": 0.463,
      "step": 88
    },
    {
      "Batch Mean": -1.4812946319580078,
      "accuracy": 0.8125,
      "epoch": 0.22,
      "step": 88
    },
    {
      "Batch Mean": -2.553070068359375,
      "accuracy": 0.8125,
      "epoch": 0.22,
      "step": 88
    },
    {
      "Batch Mean": -1.1305127143859863,
      "accuracy": 0.78125,
      "epoch": 0.22,
      "step": 88
    },
    {
      "Batch Mean": -1.9516029357910156,
      "accuracy": 0.875,
      "epoch": 0.22,
      "step": 88
    },
    {
      "epoch": 0.2225,
      "grad_norm": 8.485116004943848,
      "learning_rate": 2.4552631578947367e-06,
      "loss": 0.4295,
      "step": 89
    },
    {
      "Batch Mean": -1.5622539520263672,
      "accuracy": 0.8125,
      "epoch": 0.2225,
      "step": 89
    },
    {
      "Batch Mean": -2.8045387268066406,
      "accuracy": 0.875,
      "epoch": 0.2225,
      "step": 89
    },
    {
      "Batch Mean": -1.2656402587890625,
      "accuracy": 0.78125,
      "epoch": 0.2225,
      "step": 89
    },
    {
      "Batch Mean": -1.0702705383300781,
      "accuracy": 0.84375,
      "epoch": 0.2225,
      "step": 89
    },
    {
      "epoch": 0.225,
      "grad_norm": 6.422337532043457,
      "learning_rate": 2.4473684210526314e-06,
      "loss": 0.3633,
      "step": 90
    },
    {
      "Batch Mean": -1.141378402709961,
      "accuracy": 0.71875,
      "epoch": 0.225,
      "step": 90
    },
    {
      "Batch Mean": -1.7803688049316406,
      "accuracy": 0.875,
      "epoch": 0.225,
      "step": 90
    },
    {
      "Batch Mean": -1.991696834564209,
      "accuracy": 0.75,
      "epoch": 0.225,
      "step": 90
    },
    {
      "Batch Mean": -0.957183837890625,
      "accuracy": 0.875,
      "epoch": 0.225,
      "step": 90
    },
    {
      "epoch": 0.2275,
      "grad_norm": 8.08924388885498,
      "learning_rate": 2.439473684210526e-06,
      "loss": 0.4245,
      "step": 91
    },
    {
      "Batch Mean": -1.1860132217407227,
      "accuracy": 0.71875,
      "epoch": 0.2275,
      "step": 91
    },
    {
      "Batch Mean": -0.84332275390625,
      "accuracy": 0.90625,
      "epoch": 0.2275,
      "step": 91
    },
    {
      "Batch Mean": -2.0398807525634766,
      "accuracy": 0.75,
      "epoch": 0.2275,
      "step": 91
    },
    {
      "Batch Mean": -1.4838409423828125,
      "accuracy": 0.875,
      "epoch": 0.2275,
      "step": 91
    },
    {
      "epoch": 0.23,
      "grad_norm": 7.4404425621032715,
      "learning_rate": 2.431578947368421e-06,
      "loss": 0.3759,
      "step": 92
    },
    {
      "Batch Mean": -1.5761184692382812,
      "accuracy": 0.96875,
      "epoch": 0.23,
      "step": 92
    },
    {
      "Batch Mean": -2.564685821533203,
      "accuracy": 0.71875,
      "epoch": 0.23,
      "step": 92
    },
    {
      "Batch Mean": -1.5132369995117188,
      "accuracy": 0.6875,
      "epoch": 0.23,
      "step": 92
    },
    {
      "Batch Mean": -2.0555648803710938,
      "accuracy": 0.875,
      "epoch": 0.23,
      "step": 92
    },
    {
      "epoch": 0.2325,
      "grad_norm": 6.906028747558594,
      "learning_rate": 2.4236842105263157e-06,
      "loss": 0.4027,
      "step": 93
    },
    {
      "Batch Mean": -1.4121761322021484,
      "accuracy": 0.875,
      "epoch": 0.2325,
      "step": 93
    },
    {
      "Batch Mean": -1.95318603515625,
      "accuracy": 0.8125,
      "epoch": 0.2325,
      "step": 93
    },
    {
      "Batch Mean": -1.6953773498535156,
      "accuracy": 0.78125,
      "epoch": 0.2325,
      "step": 93
    },
    {
      "Batch Mean": -1.5923538208007812,
      "accuracy": 0.875,
      "epoch": 0.2325,
      "step": 93
    },
    {
      "epoch": 0.235,
      "grad_norm": 6.271661758422852,
      "learning_rate": 2.4157894736842104e-06,
      "loss": 0.3762,
      "step": 94
    },
    {
      "Batch Mean": -2.2878494262695312,
      "accuracy": 0.71875,
      "epoch": 0.235,
      "step": 94
    },
    {
      "Batch Mean": -2.286632537841797,
      "accuracy": 0.78125,
      "epoch": 0.235,
      "step": 94
    },
    {
      "Batch Mean": -2.0052947998046875,
      "accuracy": 0.78125,
      "epoch": 0.235,
      "step": 94
    },
    {
      "Batch Mean": -2.1512603759765625,
      "accuracy": 0.8125,
      "epoch": 0.235,
      "step": 94
    },
    {
      "epoch": 0.2375,
      "grad_norm": 7.963528156280518,
      "learning_rate": 2.4078947368421056e-06,
      "loss": 0.4721,
      "step": 95
    },
    {
      "Batch Mean": -1.9903564453125,
      "accuracy": 0.78125,
      "epoch": 0.2375,
      "step": 95
    },
    {
      "Batch Mean": -2.5484466552734375,
      "accuracy": 0.78125,
      "epoch": 0.2375,
      "step": 95
    },
    {
      "Batch Mean": -2.6534271240234375,
      "accuracy": 0.75,
      "epoch": 0.2375,
      "step": 95
    },
    {
      "Batch Mean": -1.515106201171875,
      "accuracy": 0.78125,
      "epoch": 0.2375,
      "step": 95
    },
    {
      "epoch": 0.24,
      "grad_norm": 6.653286933898926,
      "learning_rate": 2.4000000000000003e-06,
      "loss": 0.4651,
      "step": 96
    },
    {
      "Batch Mean": -2.765472412109375,
      "accuracy": 0.8125,
      "epoch": 0.24,
      "step": 96
    },
    {
      "Batch Mean": -2.008441925048828,
      "accuracy": 0.875,
      "epoch": 0.24,
      "step": 96
    },
    {
      "Batch Mean": -2.3530349731445312,
      "accuracy": 0.8125,
      "epoch": 0.24,
      "step": 96
    },
    {
      "Batch Mean": -2.2541656494140625,
      "accuracy": 0.75,
      "epoch": 0.24,
      "step": 96
    },
    {
      "epoch": 0.2425,
      "grad_norm": 6.258959770202637,
      "learning_rate": 2.392105263157895e-06,
      "loss": 0.3895,
      "step": 97
    },
    {
      "Batch Mean": -2.4180755615234375,
      "accuracy": 0.84375,
      "epoch": 0.2425,
      "step": 97
    },
    {
      "Batch Mean": -2.4765663146972656,
      "accuracy": 0.78125,
      "epoch": 0.2425,
      "step": 97
    },
    {
      "Batch Mean": -2.5517921447753906,
      "accuracy": 0.8125,
      "epoch": 0.2425,
      "step": 97
    },
    {
      "Batch Mean": -2.794464111328125,
      "accuracy": 0.75,
      "epoch": 0.2425,
      "step": 97
    },
    {
      "epoch": 0.245,
      "grad_norm": 6.597422122955322,
      "learning_rate": 2.38421052631579e-06,
      "loss": 0.3898,
      "step": 98
    },
    {
      "Batch Mean": -2.1800403594970703,
      "accuracy": 0.6875,
      "epoch": 0.245,
      "step": 98
    },
    {
      "Batch Mean": -2.8349609375,
      "accuracy": 0.8125,
      "epoch": 0.245,
      "step": 98
    },
    {
      "Batch Mean": -2.4977874755859375,
      "accuracy": 0.8125,
      "epoch": 0.245,
      "step": 98
    },
    {
      "Batch Mean": -2.3620100021362305,
      "accuracy": 0.78125,
      "epoch": 0.245,
      "step": 98
    },
    {
      "epoch": 0.2475,
      "grad_norm": 7.636416912078857,
      "learning_rate": 2.376315789473684e-06,
      "loss": 0.5111,
      "step": 99
    },
    {
      "Batch Mean": -2.890850067138672,
      "accuracy": 0.84375,
      "epoch": 0.2475,
      "step": 99
    },
    {
      "Batch Mean": -2.570037841796875,
      "accuracy": 0.8125,
      "epoch": 0.2475,
      "step": 99
    },
    {
      "Batch Mean": -2.489429473876953,
      "accuracy": 0.90625,
      "epoch": 0.2475,
      "step": 99
    },
    {
      "Batch Mean": -2.4925973415374756,
      "accuracy": 0.84375,
      "epoch": 0.2475,
      "step": 99
    },
    {
      "epoch": 0.25,
      "grad_norm": 5.276621341705322,
      "learning_rate": 2.368421052631579e-06,
      "loss": 0.3493,
      "step": 100
    },
    {
      "Batch Mean": -2.685060501098633,
      "accuracy": 0.8125,
      "epoch": 0.25,
      "step": 100
    },
    {
      "Batch Mean": -3.0683860778808594,
      "accuracy": 0.75,
      "epoch": 0.25,
      "step": 100
    },
    {
      "Batch Mean": -2.406707763671875,
      "accuracy": 0.84375,
      "epoch": 0.25,
      "step": 100
    },
    {
      "Batch Mean": -3.204519271850586,
      "accuracy": 0.75,
      "epoch": 0.25,
      "step": 100
    },
    {
      "epoch": 0.2525,
      "grad_norm": 7.729552745819092,
      "learning_rate": 2.3605263157894736e-06,
      "loss": 0.4712,
      "step": 101
    },
    {
      "Batch Mean": -2.516643524169922,
      "accuracy": 0.875,
      "epoch": 0.2525,
      "step": 101
    },
    {
      "Batch Mean": -2.89825439453125,
      "accuracy": 0.90625,
      "epoch": 0.2525,
      "step": 101
    },
    {
      "Batch Mean": -2.921875,
      "accuracy": 0.8125,
      "epoch": 0.2525,
      "step": 101
    },
    {
      "Batch Mean": -2.51778507232666,
      "accuracy": 0.78125,
      "epoch": 0.2525,
      "step": 101
    },
    {
      "epoch": 0.255,
      "grad_norm": 6.247341156005859,
      "learning_rate": 2.3526315789473684e-06,
      "loss": 0.3784,
      "step": 102
    },
    {
      "Batch Mean": -2.3507378101348877,
      "accuracy": 0.8125,
      "epoch": 0.255,
      "step": 102
    },
    {
      "Batch Mean": -2.923431396484375,
      "accuracy": 0.875,
      "epoch": 0.255,
      "step": 102
    },
    {
      "Batch Mean": -2.9839935302734375,
      "accuracy": 0.84375,
      "epoch": 0.255,
      "step": 102
    },
    {
      "Batch Mean": -2.5507564544677734,
      "accuracy": 0.90625,
      "epoch": 0.255,
      "step": 102
    },
    {
      "epoch": 0.2575,
      "grad_norm": 5.437046051025391,
      "learning_rate": 2.344736842105263e-06,
      "loss": 0.3234,
      "step": 103
    },
    {
      "Batch Mean": -2.7343482971191406,
      "accuracy": 0.75,
      "epoch": 0.2575,
      "step": 103
    },
    {
      "Batch Mean": -3.000023365020752,
      "accuracy": 0.75,
      "epoch": 0.2575,
      "step": 103
    },
    {
      "Batch Mean": -3.117097854614258,
      "accuracy": 0.8125,
      "epoch": 0.2575,
      "step": 103
    },
    {
      "Batch Mean": -3.2165985107421875,
      "accuracy": 0.8125,
      "epoch": 0.2575,
      "step": 103
    },
    {
      "epoch": 0.26,
      "grad_norm": 7.148366451263428,
      "learning_rate": 2.336842105263158e-06,
      "loss": 0.4215,
      "step": 104
    },
    {
      "Batch Mean": -3.3616867065429688,
      "accuracy": 0.84375,
      "epoch": 0.26,
      "step": 104
    },
    {
      "Batch Mean": -3.173351287841797,
      "accuracy": 0.875,
      "epoch": 0.26,
      "step": 104
    },
    {
      "Batch Mean": -2.648517608642578,
      "accuracy": 0.71875,
      "epoch": 0.26,
      "step": 104
    },
    {
      "Batch Mean": -3.0380096435546875,
      "accuracy": 0.78125,
      "epoch": 0.26,
      "step": 104
    },
    {
      "epoch": 0.2625,
      "grad_norm": 7.0915021896362305,
      "learning_rate": 2.3289473684210526e-06,
      "loss": 0.3746,
      "step": 105
    },
    {
      "Batch Mean": -2.9630088806152344,
      "accuracy": 0.90625,
      "epoch": 0.2625,
      "step": 105
    },
    {
      "Batch Mean": -2.8781280517578125,
      "accuracy": 0.75,
      "epoch": 0.2625,
      "step": 105
    },
    {
      "Batch Mean": -3.432220458984375,
      "accuracy": 0.65625,
      "epoch": 0.2625,
      "step": 105
    },
    {
      "Batch Mean": -2.7377090454101562,
      "accuracy": 0.875,
      "epoch": 0.2625,
      "step": 105
    },
    {
      "epoch": 0.265,
      "grad_norm": 6.841515064239502,
      "learning_rate": 2.3210526315789473e-06,
      "loss": 0.3739,
      "step": 106
    },
    {
      "Batch Mean": -2.955535888671875,
      "accuracy": 0.875,
      "epoch": 0.265,
      "step": 106
    },
    {
      "Batch Mean": -2.5553359985351562,
      "accuracy": 0.71875,
      "epoch": 0.265,
      "step": 106
    },
    {
      "Batch Mean": -3.257892608642578,
      "accuracy": 0.75,
      "epoch": 0.265,
      "step": 106
    },
    {
      "Batch Mean": -2.7374954223632812,
      "accuracy": 0.8125,
      "epoch": 0.265,
      "step": 106
    },
    {
      "epoch": 0.2675,
      "grad_norm": 6.91084623336792,
      "learning_rate": 2.313157894736842e-06,
      "loss": 0.3924,
      "step": 107
    },
    {
      "Batch Mean": -3.3003501892089844,
      "accuracy": 0.84375,
      "epoch": 0.2675,
      "step": 107
    },
    {
      "Batch Mean": -2.7673654556274414,
      "accuracy": 0.65625,
      "epoch": 0.2675,
      "step": 107
    },
    {
      "Batch Mean": -2.5366382598876953,
      "accuracy": 0.75,
      "epoch": 0.2675,
      "step": 107
    },
    {
      "Batch Mean": -2.509608507156372,
      "accuracy": 0.90625,
      "epoch": 0.2675,
      "step": 107
    },
    {
      "epoch": 0.27,
      "grad_norm": 7.229374885559082,
      "learning_rate": 2.305263157894737e-06,
      "loss": 0.4284,
      "step": 108
    },
    {
      "Batch Mean": -3.148345947265625,
      "accuracy": 0.8125,
      "epoch": 0.27,
      "step": 108
    },
    {
      "Batch Mean": -3.0621566772460938,
      "accuracy": 0.90625,
      "epoch": 0.27,
      "step": 108
    },
    {
      "Batch Mean": -2.859434127807617,
      "accuracy": 0.71875,
      "epoch": 0.27,
      "step": 108
    },
    {
      "Batch Mean": -3.1439666748046875,
      "accuracy": 0.84375,
      "epoch": 0.27,
      "step": 108
    },
    {
      "epoch": 0.2725,
      "grad_norm": 8.356084823608398,
      "learning_rate": 2.2973684210526316e-06,
      "loss": 0.4372,
      "step": 109
    },
    {
      "Batch Mean": -2.7968883514404297,
      "accuracy": 0.875,
      "epoch": 0.2725,
      "step": 109
    },
    {
      "Batch Mean": -2.700042724609375,
      "accuracy": 0.8125,
      "epoch": 0.2725,
      "step": 109
    },
    {
      "Batch Mean": -2.6135177612304688,
      "accuracy": 0.90625,
      "epoch": 0.2725,
      "step": 109
    },
    {
      "Batch Mean": -3.498870849609375,
      "accuracy": 0.78125,
      "epoch": 0.2725,
      "step": 109
    },
    {
      "epoch": 0.275,
      "grad_norm": 6.937132358551025,
      "learning_rate": 2.2894736842105263e-06,
      "loss": 0.338,
      "step": 110
    },
    {
      "Batch Mean": -2.8419265747070312,
      "accuracy": 0.90625,
      "epoch": 0.275,
      "step": 110
    },
    {
      "Batch Mean": -1.7743732929229736,
      "accuracy": 0.90625,
      "epoch": 0.275,
      "step": 110
    },
    {
      "Batch Mean": -2.9847335815429688,
      "accuracy": 0.90625,
      "epoch": 0.275,
      "step": 110
    },
    {
      "Batch Mean": -2.8534889221191406,
      "accuracy": 0.84375,
      "epoch": 0.275,
      "step": 110
    },
    {
      "epoch": 0.2775,
      "grad_norm": 6.012710094451904,
      "learning_rate": 2.281578947368421e-06,
      "loss": 0.2648,
      "step": 111
    },
    {
      "Batch Mean": -2.719818115234375,
      "accuracy": 0.96875,
      "epoch": 0.2775,
      "step": 111
    },
    {
      "Batch Mean": -2.7094573974609375,
      "accuracy": 0.8125,
      "epoch": 0.2775,
      "step": 111
    },
    {
      "Batch Mean": -3.1852569580078125,
      "accuracy": 0.9375,
      "epoch": 0.2775,
      "step": 111
    },
    {
      "Batch Mean": -2.9012527465820312,
      "accuracy": 0.9375,
      "epoch": 0.2775,
      "step": 111
    },
    {
      "epoch": 0.28,
      "grad_norm": 5.613126754760742,
      "learning_rate": 2.273684210526316e-06,
      "loss": 0.2261,
      "step": 112
    },
    {
      "Batch Mean": -2.317493438720703,
      "accuracy": 0.84375,
      "epoch": 0.28,
      "step": 112
    },
    {
      "Batch Mean": -2.7786102294921875,
      "accuracy": 0.875,
      "epoch": 0.28,
      "step": 112
    },
    {
      "Batch Mean": -2.4584197998046875,
      "accuracy": 0.75,
      "epoch": 0.28,
      "step": 112
    },
    {
      "Batch Mean": -3.82269287109375,
      "accuracy": 0.84375,
      "epoch": 0.28,
      "step": 112
    },
    {
      "epoch": 0.2825,
      "grad_norm": 8.895098686218262,
      "learning_rate": 2.2657894736842106e-06,
      "loss": 0.4254,
      "step": 113
    },
    {
      "Batch Mean": -2.82708740234375,
      "accuracy": 0.84375,
      "epoch": 0.2825,
      "step": 113
    },
    {
      "Batch Mean": -2.834275245666504,
      "accuracy": 0.96875,
      "epoch": 0.2825,
      "step": 113
    },
    {
      "Batch Mean": -2.4896316528320312,
      "accuracy": 0.90625,
      "epoch": 0.2825,
      "step": 113
    },
    {
      "Batch Mean": -2.9547224044799805,
      "accuracy": 0.875,
      "epoch": 0.2825,
      "step": 113
    },
    {
      "epoch": 0.285,
      "grad_norm": 6.938596725463867,
      "learning_rate": 2.2578947368421053e-06,
      "loss": 0.2756,
      "step": 114
    },
    {
      "Batch Mean": -3.3860321044921875,
      "accuracy": 0.84375,
      "epoch": 0.285,
      "step": 114
    },
    {
      "Batch Mean": -1.289031982421875,
      "accuracy": 0.75,
      "epoch": 0.285,
      "step": 114
    },
    {
      "Batch Mean": -1.973724365234375,
      "accuracy": 0.6875,
      "epoch": 0.285,
      "step": 114
    },
    {
      "Batch Mean": -1.4242782592773438,
      "accuracy": 0.78125,
      "epoch": 0.285,
      "step": 114
    },
    {
      "epoch": 0.2875,
      "grad_norm": 10.572776794433594,
      "learning_rate": 2.25e-06,
      "loss": 0.4922,
      "step": 115
    },
    {
      "Batch Mean": -2.7473220825195312,
      "accuracy": 0.6875,
      "epoch": 0.2875,
      "step": 115
    },
    {
      "Batch Mean": -2.382610321044922,
      "accuracy": 0.90625,
      "epoch": 0.2875,
      "step": 115
    },
    {
      "Batch Mean": -2.4156417846679688,
      "accuracy": 0.75,
      "epoch": 0.2875,
      "step": 115
    },
    {
      "Batch Mean": -2.8010101318359375,
      "accuracy": 0.9375,
      "epoch": 0.2875,
      "step": 115
    },
    {
      "epoch": 0.29,
      "grad_norm": 8.355461120605469,
      "learning_rate": 2.242105263157895e-06,
      "loss": 0.3628,
      "step": 116
    },
    {
      "Batch Mean": -3.490283966064453,
      "accuracy": 0.96875,
      "epoch": 0.29,
      "step": 116
    },
    {
      "Batch Mean": -3.179001808166504,
      "accuracy": 0.84375,
      "epoch": 0.29,
      "step": 116
    },
    {
      "Batch Mean": -2.5853271484375,
      "accuracy": 0.84375,
      "epoch": 0.29,
      "step": 116
    },
    {
      "Batch Mean": -2.0840072631835938,
      "accuracy": 0.75,
      "epoch": 0.29,
      "step": 116
    },
    {
      "epoch": 0.2925,
      "grad_norm": 7.448988437652588,
      "learning_rate": 2.2342105263157895e-06,
      "loss": 0.3418,
      "step": 117
    },
    {
      "Batch Mean": -3.2340087890625,
      "accuracy": 0.84375,
      "epoch": 0.2925,
      "step": 117
    },
    {
      "Batch Mean": -2.6928272247314453,
      "accuracy": 0.78125,
      "epoch": 0.2925,
      "step": 117
    },
    {
      "Batch Mean": -2.367229461669922,
      "accuracy": 0.9375,
      "epoch": 0.2925,
      "step": 117
    },
    {
      "Batch Mean": -2.940413475036621,
      "accuracy": 0.84375,
      "epoch": 0.2925,
      "step": 117
    },
    {
      "epoch": 0.295,
      "grad_norm": 7.369281768798828,
      "learning_rate": 2.2263157894736843e-06,
      "loss": 0.3758,
      "step": 118
    },
    {
      "Batch Mean": -3.154327392578125,
      "accuracy": 0.75,
      "epoch": 0.295,
      "step": 118
    },
    {
      "Batch Mean": -3.162120819091797,
      "accuracy": 0.78125,
      "epoch": 0.295,
      "step": 118
    },
    {
      "Batch Mean": -2.571208953857422,
      "accuracy": 0.75,
      "epoch": 0.295,
      "step": 118
    },
    {
      "Batch Mean": -2.0688514709472656,
      "accuracy": 0.78125,
      "epoch": 0.295,
      "step": 118
    },
    {
      "epoch": 0.2975,
      "grad_norm": 9.859661102294922,
      "learning_rate": 2.218421052631579e-06,
      "loss": 0.4781,
      "step": 119
    },
    {
      "Batch Mean": -2.288646697998047,
      "accuracy": 0.8125,
      "epoch": 0.2975,
      "step": 119
    },
    {
      "Batch Mean": -2.9682788848876953,
      "accuracy": 0.75,
      "epoch": 0.2975,
      "step": 119
    },
    {
      "Batch Mean": -3.6280517578125,
      "accuracy": 0.78125,
      "epoch": 0.2975,
      "step": 119
    },
    {
      "Batch Mean": -2.7621307373046875,
      "accuracy": 0.875,
      "epoch": 0.2975,
      "step": 119
    },
    {
      "epoch": 0.3,
      "grad_norm": 7.310153961181641,
      "learning_rate": 2.2105263157894738e-06,
      "loss": 0.3414,
      "step": 120
    },
    {
      "Batch Mean": -3.35638427734375,
      "accuracy": 0.8125,
      "epoch": 0.3,
      "step": 120
    },
    {
      "Batch Mean": -3.4458541870117188,
      "accuracy": 0.75,
      "epoch": 0.3,
      "step": 120
    },
    {
      "Batch Mean": -3.32684326171875,
      "accuracy": 0.8125,
      "epoch": 0.3,
      "step": 120
    },
    {
      "Batch Mean": -3.52008056640625,
      "accuracy": 0.875,
      "epoch": 0.3,
      "step": 120
    },
    {
      "epoch": 0.3025,
      "grad_norm": 6.901857852935791,
      "learning_rate": 2.2026315789473685e-06,
      "loss": 0.3663,
      "step": 121
    },
    {
      "Batch Mean": -3.2964706420898438,
      "accuracy": 0.90625,
      "epoch": 0.3025,
      "step": 121
    },
    {
      "Batch Mean": -4.0569610595703125,
      "accuracy": 0.75,
      "epoch": 0.3025,
      "step": 121
    },
    {
      "Batch Mean": -3.8861026763916016,
      "accuracy": 0.84375,
      "epoch": 0.3025,
      "step": 121
    },
    {
      "Batch Mean": -4.03857421875,
      "accuracy": 0.8125,
      "epoch": 0.3025,
      "step": 121
    },
    {
      "epoch": 0.305,
      "grad_norm": 6.965332508087158,
      "learning_rate": 2.1947368421052633e-06,
      "loss": 0.3583,
      "step": 122
    },
    {
      "Batch Mean": -4.39910888671875,
      "accuracy": 0.90625,
      "epoch": 0.305,
      "step": 122
    },
    {
      "Batch Mean": -3.1727142333984375,
      "accuracy": 0.78125,
      "epoch": 0.305,
      "step": 122
    },
    {
      "Batch Mean": -3.89581298828125,
      "accuracy": 0.8125,
      "epoch": 0.305,
      "step": 122
    },
    {
      "Batch Mean": -3.6239490509033203,
      "accuracy": 0.84375,
      "epoch": 0.305,
      "step": 122
    },
    {
      "epoch": 0.3075,
      "grad_norm": 6.396092414855957,
      "learning_rate": 2.186842105263158e-06,
      "loss": 0.321,
      "step": 123
    },
    {
      "Batch Mean": -3.5579376220703125,
      "accuracy": 0.90625,
      "epoch": 0.3075,
      "step": 123
    },
    {
      "Batch Mean": -3.6140670776367188,
      "accuracy": 0.78125,
      "epoch": 0.3075,
      "step": 123
    },
    {
      "Batch Mean": -3.25006103515625,
      "accuracy": 0.875,
      "epoch": 0.3075,
      "step": 123
    },
    {
      "Batch Mean": -3.859375,
      "accuracy": 0.84375,
      "epoch": 0.3075,
      "step": 123
    },
    {
      "epoch": 0.31,
      "grad_norm": 6.925971984863281,
      "learning_rate": 2.1789473684210528e-06,
      "loss": 0.3773,
      "step": 124
    },
    {
      "Batch Mean": -3.74835205078125,
      "accuracy": 0.9375,
      "epoch": 0.31,
      "step": 124
    },
    {
      "Batch Mean": -2.98968505859375,
      "accuracy": 0.71875,
      "epoch": 0.31,
      "step": 124
    },
    {
      "Batch Mean": -3.8388538360595703,
      "accuracy": 0.78125,
      "epoch": 0.31,
      "step": 124
    },
    {
      "Batch Mean": -3.46600341796875,
      "accuracy": 0.84375,
      "epoch": 0.31,
      "step": 124
    },
    {
      "epoch": 0.3125,
      "grad_norm": 7.271778106689453,
      "learning_rate": 2.1710526315789475e-06,
      "loss": 0.4283,
      "step": 125
    },
    {
      "Batch Mean": -3.1362533569335938,
      "accuracy": 0.8125,
      "epoch": 0.3125,
      "step": 125
    },
    {
      "Batch Mean": -3.978717803955078,
      "accuracy": 0.90625,
      "epoch": 0.3125,
      "step": 125
    },
    {
      "Batch Mean": -2.7917633056640625,
      "accuracy": 0.75,
      "epoch": 0.3125,
      "step": 125
    },
    {
      "Batch Mean": -3.4448699951171875,
      "accuracy": 0.75,
      "epoch": 0.3125,
      "step": 125
    },
    {
      "epoch": 0.315,
      "grad_norm": 6.391500473022461,
      "learning_rate": 2.1631578947368423e-06,
      "loss": 0.377,
      "step": 126
    },
    {
      "Batch Mean": -3.4984054565429688,
      "accuracy": 0.90625,
      "epoch": 0.315,
      "step": 126
    },
    {
      "Batch Mean": -3.4216766357421875,
      "accuracy": 0.9375,
      "epoch": 0.315,
      "step": 126
    },
    {
      "Batch Mean": -3.199005126953125,
      "accuracy": 0.8125,
      "epoch": 0.315,
      "step": 126
    },
    {
      "Batch Mean": -3.3887786865234375,
      "accuracy": 0.6875,
      "epoch": 0.315,
      "step": 126
    },
    {
      "epoch": 0.3175,
      "grad_norm": 7.132022380828857,
      "learning_rate": 2.155263157894737e-06,
      "loss": 0.348,
      "step": 127
    },
    {
      "Batch Mean": -3.533205032348633,
      "accuracy": 0.875,
      "epoch": 0.3175,
      "step": 127
    },
    {
      "Batch Mean": -3.0708465576171875,
      "accuracy": 0.84375,
      "epoch": 0.3175,
      "step": 127
    },
    {
      "Batch Mean": -2.5532379150390625,
      "accuracy": 0.875,
      "epoch": 0.3175,
      "step": 127
    },
    {
      "Batch Mean": -3.242755889892578,
      "accuracy": 0.8125,
      "epoch": 0.3175,
      "step": 127
    },
    {
      "epoch": 0.32,
      "grad_norm": 6.543447971343994,
      "learning_rate": 2.1473684210526317e-06,
      "loss": 0.3184,
      "step": 128
    },
    {
      "Batch Mean": -3.64508056640625,
      "accuracy": 0.90625,
      "epoch": 0.32,
      "step": 128
    },
    {
      "Batch Mean": -4.324256896972656,
      "accuracy": 0.9375,
      "epoch": 0.32,
      "step": 128
    },
    {
      "Batch Mean": -3.880157470703125,
      "accuracy": 0.90625,
      "epoch": 0.32,
      "step": 128
    },
    {
      "Batch Mean": -3.4844741821289062,
      "accuracy": 0.78125,
      "epoch": 0.32,
      "step": 128
    },
    {
      "epoch": 0.3225,
      "grad_norm": 7.542953014373779,
      "learning_rate": 2.1394736842105265e-06,
      "loss": 0.3573,
      "step": 129
    },
    {
      "Batch Mean": -3.59429931640625,
      "accuracy": 0.78125,
      "epoch": 0.3225,
      "step": 129
    },
    {
      "Batch Mean": -3.3390655517578125,
      "accuracy": 0.9375,
      "epoch": 0.3225,
      "step": 129
    },
    {
      "Batch Mean": -3.0296974182128906,
      "accuracy": 0.8125,
      "epoch": 0.3225,
      "step": 129
    },
    {
      "Batch Mean": -3.78515625,
      "accuracy": 0.8125,
      "epoch": 0.3225,
      "step": 129
    },
    {
      "epoch": 0.325,
      "grad_norm": 6.553402900695801,
      "learning_rate": 2.1315789473684212e-06,
      "loss": 0.3219,
      "step": 130
    },
    {
      "Batch Mean": -2.7899856567382812,
      "accuracy": 0.8125,
      "epoch": 0.325,
      "step": 130
    },
    {
      "Batch Mean": -3.6958160400390625,
      "accuracy": 0.84375,
      "epoch": 0.325,
      "step": 130
    },
    {
      "Batch Mean": -2.806610107421875,
      "accuracy": 0.75,
      "epoch": 0.325,
      "step": 130
    },
    {
      "Batch Mean": -3.182114601135254,
      "accuracy": 0.78125,
      "epoch": 0.325,
      "step": 130
    },
    {
      "epoch": 0.3275,
      "grad_norm": 7.584489822387695,
      "learning_rate": 2.123684210526316e-06,
      "loss": 0.4417,
      "step": 131
    },
    {
      "Batch Mean": -3.492685317993164,
      "accuracy": 0.875,
      "epoch": 0.3275,
      "step": 131
    },
    {
      "Batch Mean": -3.4088897705078125,
      "accuracy": 0.90625,
      "epoch": 0.3275,
      "step": 131
    },
    {
      "Batch Mean": -3.0204315185546875,
      "accuracy": 0.84375,
      "epoch": 0.3275,
      "step": 131
    },
    {
      "Batch Mean": -3.449188232421875,
      "accuracy": 0.9375,
      "epoch": 0.3275,
      "step": 131
    },
    {
      "epoch": 0.33,
      "grad_norm": 5.928311347961426,
      "learning_rate": 2.1157894736842103e-06,
      "loss": 0.3043,
      "step": 132
    },
    {
      "Batch Mean": -2.6694793701171875,
      "accuracy": 0.8125,
      "epoch": 0.33,
      "step": 132
    },
    {
      "Batch Mean": -2.967529296875,
      "accuracy": 0.9375,
      "epoch": 0.33,
      "step": 132
    },
    {
      "Batch Mean": -4.03662109375,
      "accuracy": 0.71875,
      "epoch": 0.33,
      "step": 132
    },
    {
      "Batch Mean": -3.179962158203125,
      "accuracy": 0.65625,
      "epoch": 0.33,
      "step": 132
    },
    {
      "epoch": 0.3325,
      "grad_norm": 8.017102241516113,
      "learning_rate": 2.107894736842105e-06,
      "loss": 0.4024,
      "step": 133
    },
    {
      "Batch Mean": -3.1867218017578125,
      "accuracy": 0.90625,
      "epoch": 0.3325,
      "step": 133
    },
    {
      "Batch Mean": -3.424471855163574,
      "accuracy": 0.78125,
      "epoch": 0.3325,
      "step": 133
    },
    {
      "Batch Mean": -3.25,
      "accuracy": 0.8125,
      "epoch": 0.3325,
      "step": 133
    },
    {
      "Batch Mean": -3.68438720703125,
      "accuracy": 0.6875,
      "epoch": 0.3325,
      "step": 133
    },
    {
      "epoch": 0.335,
      "grad_norm": 7.597751140594482,
      "learning_rate": 2.1e-06,
      "loss": 0.4027,
      "step": 134
    },
    {
      "Batch Mean": -2.4028396606445312,
      "accuracy": 0.75,
      "epoch": 0.335,
      "step": 134
    },
    {
      "Batch Mean": -3.6143722534179688,
      "accuracy": 0.8125,
      "epoch": 0.335,
      "step": 134
    },
    {
      "Batch Mean": -3.1368408203125,
      "accuracy": 0.875,
      "epoch": 0.335,
      "step": 134
    },
    {
      "Batch Mean": -2.922149658203125,
      "accuracy": 0.8125,
      "epoch": 0.335,
      "step": 134
    },
    {
      "epoch": 0.3375,
      "grad_norm": 7.709555149078369,
      "learning_rate": 2.0921052631578945e-06,
      "loss": 0.4432,
      "step": 135
    },
    {
      "Batch Mean": -4.095916748046875,
      "accuracy": 0.8125,
      "epoch": 0.3375,
      "step": 135
    },
    {
      "Batch Mean": -3.7461395263671875,
      "accuracy": 0.9375,
      "epoch": 0.3375,
      "step": 135
    },
    {
      "Batch Mean": -3.748504161834717,
      "accuracy": 0.8125,
      "epoch": 0.3375,
      "step": 135
    },
    {
      "Batch Mean": -2.7383270263671875,
      "accuracy": 0.8125,
      "epoch": 0.3375,
      "step": 135
    },
    {
      "epoch": 0.34,
      "grad_norm": 6.083080291748047,
      "learning_rate": 2.0842105263157897e-06,
      "loss": 0.3501,
      "step": 136
    },
    {
      "Batch Mean": -3.571258544921875,
      "accuracy": 0.84375,
      "epoch": 0.34,
      "step": 136
    },
    {
      "Batch Mean": -3.6791038513183594,
      "accuracy": 0.875,
      "epoch": 0.34,
      "step": 136
    },
    {
      "Batch Mean": -3.7348480224609375,
      "accuracy": 0.875,
      "epoch": 0.34,
      "step": 136
    },
    {
      "Batch Mean": -3.2222461700439453,
      "accuracy": 0.71875,
      "epoch": 0.34,
      "step": 136
    },
    {
      "epoch": 0.3425,
      "grad_norm": 6.974395275115967,
      "learning_rate": 2.0763157894736845e-06,
      "loss": 0.358,
      "step": 137
    },
    {
      "Batch Mean": -4.6036376953125,
      "accuracy": 0.8125,
      "epoch": 0.3425,
      "step": 137
    },
    {
      "Batch Mean": -3.4961585998535156,
      "accuracy": 0.84375,
      "epoch": 0.3425,
      "step": 137
    },
    {
      "Batch Mean": -3.3908309936523438,
      "accuracy": 0.78125,
      "epoch": 0.3425,
      "step": 137
    },
    {
      "Batch Mean": -4.088287353515625,
      "accuracy": 0.78125,
      "epoch": 0.3425,
      "step": 137
    },
    {
      "epoch": 0.345,
      "grad_norm": 7.275962829589844,
      "learning_rate": 2.068421052631579e-06,
      "loss": 0.3928,
      "step": 138
    },
    {
      "Batch Mean": -4.08306884765625,
      "accuracy": 0.8125,
      "epoch": 0.345,
      "step": 138
    },
    {
      "Batch Mean": -3.62261962890625,
      "accuracy": 0.78125,
      "epoch": 0.345,
      "step": 138
    },
    {
      "Batch Mean": -4.1116790771484375,
      "accuracy": 0.8125,
      "epoch": 0.345,
      "step": 138
    },
    {
      "Batch Mean": -4.20538330078125,
      "accuracy": 0.875,
      "epoch": 0.345,
      "step": 138
    },
    {
      "epoch": 0.3475,
      "grad_norm": 6.272771835327148,
      "learning_rate": 2.060526315789474e-06,
      "loss": 0.3163,
      "step": 139
    },
    {
      "Batch Mean": -4.465492248535156,
      "accuracy": 0.8125,
      "epoch": 0.3475,
      "step": 139
    },
    {
      "Batch Mean": -4.651458740234375,
      "accuracy": 0.875,
      "epoch": 0.3475,
      "step": 139
    },
    {
      "Batch Mean": -4.1647186279296875,
      "accuracy": 0.90625,
      "epoch": 0.3475,
      "step": 139
    },
    {
      "Batch Mean": -2.7436389923095703,
      "accuracy": 0.875,
      "epoch": 0.3475,
      "step": 139
    },
    {
      "epoch": 0.35,
      "grad_norm": 6.057549953460693,
      "learning_rate": 2.0526315789473687e-06,
      "loss": 0.3358,
      "step": 140
    },
    {
      "Batch Mean": -3.641510009765625,
      "accuracy": 0.84375,
      "epoch": 0.35,
      "step": 140
    },
    {
      "Batch Mean": -3.993194580078125,
      "accuracy": 0.875,
      "epoch": 0.35,
      "step": 140
    },
    {
      "Batch Mean": -4.205657958984375,
      "accuracy": 0.8125,
      "epoch": 0.35,
      "step": 140
    },
    {
      "Batch Mean": -3.8518409729003906,
      "accuracy": 0.90625,
      "epoch": 0.35,
      "step": 140
    },
    {
      "epoch": 0.3525,
      "grad_norm": 6.492579936981201,
      "learning_rate": 2.0447368421052634e-06,
      "loss": 0.36,
      "step": 141
    },
    {
      "Batch Mean": -3.0039520263671875,
      "accuracy": 0.625,
      "epoch": 0.3525,
      "step": 141
    },
    {
      "Batch Mean": -3.5678176879882812,
      "accuracy": 0.78125,
      "epoch": 0.3525,
      "step": 141
    },
    {
      "Batch Mean": -3.959442138671875,
      "accuracy": 0.78125,
      "epoch": 0.3525,
      "step": 141
    },
    {
      "Batch Mean": -3.90399169921875,
      "accuracy": 0.84375,
      "epoch": 0.3525,
      "step": 141
    },
    {
      "epoch": 0.355,
      "grad_norm": 8.822011947631836,
      "learning_rate": 2.0368421052631578e-06,
      "loss": 0.4625,
      "step": 142
    },
    {
      "Batch Mean": -4.228359222412109,
      "accuracy": 0.8125,
      "epoch": 0.355,
      "step": 142
    },
    {
      "Batch Mean": -4.1596527099609375,
      "accuracy": 0.65625,
      "epoch": 0.355,
      "step": 142
    },
    {
      "Batch Mean": -3.49688720703125,
      "accuracy": 0.90625,
      "epoch": 0.355,
      "step": 142
    },
    {
      "Batch Mean": -3.519866943359375,
      "accuracy": 0.875,
      "epoch": 0.355,
      "step": 142
    },
    {
      "epoch": 0.3575,
      "grad_norm": 7.293342590332031,
      "learning_rate": 2.0289473684210525e-06,
      "loss": 0.4173,
      "step": 143
    },
    {
      "Batch Mean": -4.350372314453125,
      "accuracy": 0.875,
      "epoch": 0.3575,
      "step": 143
    },
    {
      "Batch Mean": -4.087388038635254,
      "accuracy": 0.8125,
      "epoch": 0.3575,
      "step": 143
    },
    {
      "Batch Mean": -3.1829357147216797,
      "accuracy": 0.90625,
      "epoch": 0.3575,
      "step": 143
    },
    {
      "Batch Mean": -4.2586517333984375,
      "accuracy": 0.90625,
      "epoch": 0.3575,
      "step": 143
    },
    {
      "epoch": 0.36,
      "grad_norm": 5.839393138885498,
      "learning_rate": 2.0210526315789473e-06,
      "loss": 0.3149,
      "step": 144
    },
    {
      "Batch Mean": -3.851898193359375,
      "accuracy": 0.8125,
      "epoch": 0.36,
      "step": 144
    },
    {
      "Batch Mean": -3.5305938720703125,
      "accuracy": 0.90625,
      "epoch": 0.36,
      "step": 144
    },
    {
      "Batch Mean": -4.296562194824219,
      "accuracy": 0.71875,
      "epoch": 0.36,
      "step": 144
    },
    {
      "Batch Mean": -4.097373962402344,
      "accuracy": 0.84375,
      "epoch": 0.36,
      "step": 144
    },
    {
      "epoch": 0.3625,
      "grad_norm": 6.38557767868042,
      "learning_rate": 2.013157894736842e-06,
      "loss": 0.3809,
      "step": 145
    },
    {
      "Batch Mean": -3.644855499267578,
      "accuracy": 0.8125,
      "epoch": 0.3625,
      "step": 145
    },
    {
      "Batch Mean": -4.092620849609375,
      "accuracy": 0.875,
      "epoch": 0.3625,
      "step": 145
    },
    {
      "Batch Mean": -3.9543728828430176,
      "accuracy": 0.875,
      "epoch": 0.3625,
      "step": 145
    },
    {
      "Batch Mean": -3.901203155517578,
      "accuracy": 0.9375,
      "epoch": 0.3625,
      "step": 145
    },
    {
      "epoch": 0.365,
      "grad_norm": 6.386345386505127,
      "learning_rate": 2.0052631578947367e-06,
      "loss": 0.3469,
      "step": 146
    },
    {
      "Batch Mean": -3.872467041015625,
      "accuracy": 0.875,
      "epoch": 0.365,
      "step": 146
    },
    {
      "Batch Mean": -3.0312156677246094,
      "accuracy": 0.8125,
      "epoch": 0.365,
      "step": 146
    },
    {
      "Batch Mean": -3.2101287841796875,
      "accuracy": 0.84375,
      "epoch": 0.365,
      "step": 146
    },
    {
      "Batch Mean": -3.727203369140625,
      "accuracy": 0.6875,
      "epoch": 0.365,
      "step": 146
    },
    {
      "epoch": 0.3675,
      "grad_norm": 7.31825590133667,
      "learning_rate": 1.9973684210526315e-06,
      "loss": 0.3775,
      "step": 147
    },
    {
      "Batch Mean": -4.53240966796875,
      "accuracy": 0.8125,
      "epoch": 0.3675,
      "step": 147
    },
    {
      "Batch Mean": -4.39666748046875,
      "accuracy": 0.8125,
      "epoch": 0.3675,
      "step": 147
    },
    {
      "Batch Mean": -4.245336532592773,
      "accuracy": 0.84375,
      "epoch": 0.3675,
      "step": 147
    },
    {
      "Batch Mean": -3.6735334396362305,
      "accuracy": 0.8125,
      "epoch": 0.3675,
      "step": 147
    },
    {
      "epoch": 0.37,
      "grad_norm": 8.167656898498535,
      "learning_rate": 1.9894736842105262e-06,
      "loss": 0.4151,
      "step": 148
    },
    {
      "Batch Mean": -4.223419189453125,
      "accuracy": 0.90625,
      "epoch": 0.37,
      "step": 148
    },
    {
      "Batch Mean": -2.857851028442383,
      "accuracy": 0.78125,
      "epoch": 0.37,
      "step": 148
    },
    {
      "Batch Mean": -3.902324676513672,
      "accuracy": 0.71875,
      "epoch": 0.37,
      "step": 148
    },
    {
      "Batch Mean": -4.022743225097656,
      "accuracy": 0.75,
      "epoch": 0.37,
      "step": 148
    },
    {
      "epoch": 0.3725,
      "grad_norm": 6.420680522918701,
      "learning_rate": 1.9815789473684214e-06,
      "loss": 0.4168,
      "step": 149
    },
    {
      "Batch Mean": -4.02081298828125,
      "accuracy": 0.8125,
      "epoch": 0.3725,
      "step": 149
    },
    {
      "Batch Mean": -3.3986377716064453,
      "accuracy": 0.9375,
      "epoch": 0.3725,
      "step": 149
    },
    {
      "Batch Mean": -4.357936859130859,
      "accuracy": 0.875,
      "epoch": 0.3725,
      "step": 149
    },
    {
      "Batch Mean": -3.590606689453125,
      "accuracy": 0.9375,
      "epoch": 0.3725,
      "step": 149
    },
    {
      "epoch": 0.375,
      "grad_norm": 5.989567279815674,
      "learning_rate": 1.973684210526316e-06,
      "loss": 0.2872,
      "step": 150
    },
    {
      "Batch Mean": -3.69549560546875,
      "accuracy": 0.9375,
      "epoch": 0.375,
      "step": 150
    },
    {
      "Batch Mean": -3.0214691162109375,
      "accuracy": 0.71875,
      "epoch": 0.375,
      "step": 150
    },
    {
      "Batch Mean": -3.49407958984375,
      "accuracy": 0.8125,
      "epoch": 0.375,
      "step": 150
    },
    {
      "Batch Mean": -3.97283935546875,
      "accuracy": 0.8125,
      "epoch": 0.375,
      "step": 150
    },
    {
      "epoch": 0.3775,
      "grad_norm": 6.451083660125732,
      "learning_rate": 1.965789473684211e-06,
      "loss": 0.3594,
      "step": 151
    },
    {
      "Batch Mean": -3.495819091796875,
      "accuracy": 0.875,
      "epoch": 0.3775,
      "step": 151
    },
    {
      "Batch Mean": -4.1375274658203125,
      "accuracy": 0.8125,
      "epoch": 0.3775,
      "step": 151
    },
    {
      "Batch Mean": -3.330834150314331,
      "accuracy": 0.78125,
      "epoch": 0.3775,
      "step": 151
    },
    {
      "Batch Mean": -3.2013702392578125,
      "accuracy": 0.875,
      "epoch": 0.3775,
      "step": 151
    },
    {
      "epoch": 0.38,
      "grad_norm": 7.380356788635254,
      "learning_rate": 1.9578947368421052e-06,
      "loss": 0.4359,
      "step": 152
    },
    {
      "Batch Mean": -3.5101318359375,
      "accuracy": 0.65625,
      "epoch": 0.38,
      "step": 152
    },
    {
      "Batch Mean": -3.5372161865234375,
      "accuracy": 0.8125,
      "epoch": 0.38,
      "step": 152
    },
    {
      "Batch Mean": -2.7825088500976562,
      "accuracy": 0.8125,
      "epoch": 0.38,
      "step": 152
    },
    {
      "Batch Mean": -3.661346435546875,
      "accuracy": 0.96875,
      "epoch": 0.38,
      "step": 152
    },
    {
      "epoch": 0.3825,
      "grad_norm": 5.823904514312744,
      "learning_rate": 1.95e-06,
      "loss": 0.3266,
      "step": 153
    },
    {
      "Batch Mean": -2.8103408813476562,
      "accuracy": 0.84375,
      "epoch": 0.3825,
      "step": 153
    },
    {
      "Batch Mean": -3.345386505126953,
      "accuracy": 0.84375,
      "epoch": 0.3825,
      "step": 153
    },
    {
      "Batch Mean": -3.250551223754883,
      "accuracy": 0.8125,
      "epoch": 0.3825,
      "step": 153
    },
    {
      "Batch Mean": -3.2271499633789062,
      "accuracy": 0.8125,
      "epoch": 0.3825,
      "step": 153
    },
    {
      "epoch": 0.385,
      "grad_norm": 8.050492286682129,
      "learning_rate": 1.9421052631578947e-06,
      "loss": 0.3497,
      "step": 154
    },
    {
      "Batch Mean": -2.0474777221679688,
      "accuracy": 0.90625,
      "epoch": 0.385,
      "step": 154
    },
    {
      "Batch Mean": -3.3988208770751953,
      "accuracy": 0.90625,
      "epoch": 0.385,
      "step": 154
    },
    {
      "Batch Mean": -3.7056427001953125,
      "accuracy": 0.78125,
      "epoch": 0.385,
      "step": 154
    },
    {
      "Batch Mean": -2.6432647705078125,
      "accuracy": 0.84375,
      "epoch": 0.385,
      "step": 154
    },
    {
      "epoch": 0.3875,
      "grad_norm": 5.581839561462402,
      "learning_rate": 1.9342105263157895e-06,
      "loss": 0.3112,
      "step": 155
    },
    {
      "Batch Mean": -2.814176559448242,
      "accuracy": 0.84375,
      "epoch": 0.3875,
      "step": 155
    },
    {
      "Batch Mean": -2.0308380126953125,
      "accuracy": 0.90625,
      "epoch": 0.3875,
      "step": 155
    },
    {
      "Batch Mean": -3.164947509765625,
      "accuracy": 0.90625,
      "epoch": 0.3875,
      "step": 155
    },
    {
      "Batch Mean": -2.7253189086914062,
      "accuracy": 0.84375,
      "epoch": 0.3875,
      "step": 155
    },
    {
      "epoch": 0.39,
      "grad_norm": 6.545515537261963,
      "learning_rate": 1.926315789473684e-06,
      "loss": 0.3156,
      "step": 156
    },
    {
      "Batch Mean": -2.7420883178710938,
      "accuracy": 0.8125,
      "epoch": 0.39,
      "step": 156
    },
    {
      "Batch Mean": -1.8706493377685547,
      "accuracy": 0.875,
      "epoch": 0.39,
      "step": 156
    },
    {
      "Batch Mean": -1.612701416015625,
      "accuracy": 0.9375,
      "epoch": 0.39,
      "step": 156
    },
    {
      "Batch Mean": -2.74774169921875,
      "accuracy": 0.75,
      "epoch": 0.39,
      "step": 156
    },
    {
      "epoch": 0.3925,
      "grad_norm": 6.551059246063232,
      "learning_rate": 1.918421052631579e-06,
      "loss": 0.3152,
      "step": 157
    },
    {
      "Batch Mean": -3.313770294189453,
      "accuracy": 0.875,
      "epoch": 0.3925,
      "step": 157
    },
    {
      "Batch Mean": -2.3776092529296875,
      "accuracy": 0.78125,
      "epoch": 0.3925,
      "step": 157
    },
    {
      "Batch Mean": -2.6562118530273438,
      "accuracy": 0.84375,
      "epoch": 0.3925,
      "step": 157
    },
    {
      "Batch Mean": -2.5965747833251953,
      "accuracy": 0.9375,
      "epoch": 0.3925,
      "step": 157
    },
    {
      "epoch": 0.395,
      "grad_norm": 7.611530303955078,
      "learning_rate": 1.9105263157894737e-06,
      "loss": 0.2848,
      "step": 158
    },
    {
      "Batch Mean": -2.9923248291015625,
      "accuracy": 0.84375,
      "epoch": 0.395,
      "step": 158
    },
    {
      "Batch Mean": -3.1744384765625,
      "accuracy": 0.875,
      "epoch": 0.395,
      "step": 158
    },
    {
      "Batch Mean": -1.654348373413086,
      "accuracy": 0.84375,
      "epoch": 0.395,
      "step": 158
    },
    {
      "Batch Mean": -2.4622344970703125,
      "accuracy": 0.875,
      "epoch": 0.395,
      "step": 158
    },
    {
      "epoch": 0.3975,
      "grad_norm": 6.978616237640381,
      "learning_rate": 1.9026315789473684e-06,
      "loss": 0.3135,
      "step": 159
    },
    {
      "Batch Mean": -2.66357421875,
      "accuracy": 0.9375,
      "epoch": 0.3975,
      "step": 159
    },
    {
      "Batch Mean": -0.5262050628662109,
      "accuracy": 0.875,
      "epoch": 0.3975,
      "step": 159
    },
    {
      "Batch Mean": -2.160125732421875,
      "accuracy": 0.84375,
      "epoch": 0.3975,
      "step": 159
    },
    {
      "Batch Mean": -2.8070220947265625,
      "accuracy": 0.8125,
      "epoch": 0.3975,
      "step": 159
    },
    {
      "epoch": 0.4,
      "grad_norm": 6.540273666381836,
      "learning_rate": 1.8947368421052632e-06,
      "loss": 0.2625,
      "step": 160
    },
    {
      "Batch Mean": -1.2659626007080078,
      "accuracy": 0.90625,
      "epoch": 0.4,
      "step": 160
    },
    {
      "Batch Mean": -1.5436820983886719,
      "accuracy": 0.84375,
      "epoch": 0.4,
      "step": 160
    },
    {
      "Batch Mean": -1.8903522491455078,
      "accuracy": 0.78125,
      "epoch": 0.4,
      "step": 160
    },
    {
      "Batch Mean": -2.8146209716796875,
      "accuracy": 0.84375,
      "epoch": 0.4,
      "step": 160
    },
    {
      "epoch": 0.4025,
      "grad_norm": 8.280950546264648,
      "learning_rate": 1.8868421052631577e-06,
      "loss": 0.3544,
      "step": 161
    },
    {
      "Batch Mean": -0.720062255859375,
      "accuracy": 0.90625,
      "epoch": 0.4025,
      "step": 161
    },
    {
      "Batch Mean": -1.4061965942382812,
      "accuracy": 0.78125,
      "epoch": 0.4025,
      "step": 161
    },
    {
      "Batch Mean": -1.1033363342285156,
      "accuracy": 0.84375,
      "epoch": 0.4025,
      "step": 161
    },
    {
      "Batch Mean": -1.5503730773925781,
      "accuracy": 0.96875,
      "epoch": 0.4025,
      "step": 161
    },
    {
      "epoch": 0.405,
      "grad_norm": 7.286800384521484,
      "learning_rate": 1.8789473684210525e-06,
      "loss": 0.3147,
      "step": 162
    },
    {
      "Batch Mean": -1.6750932931900024,
      "accuracy": 0.6875,
      "epoch": 0.405,
      "step": 162
    },
    {
      "Batch Mean": -2.2614898681640625,
      "accuracy": 0.875,
      "epoch": 0.405,
      "step": 162
    },
    {
      "Batch Mean": -2.33111572265625,
      "accuracy": 0.9375,
      "epoch": 0.405,
      "step": 162
    },
    {
      "Batch Mean": -1.9058303833007812,
      "accuracy": 0.8125,
      "epoch": 0.405,
      "step": 162
    },
    {
      "epoch": 0.4075,
      "grad_norm": 8.709750175476074,
      "learning_rate": 1.8710526315789476e-06,
      "loss": 0.3208,
      "step": 163
    },
    {
      "Batch Mean": -1.1716461181640625,
      "accuracy": 0.84375,
      "epoch": 0.4075,
      "step": 163
    },
    {
      "Batch Mean": -2.4022750854492188,
      "accuracy": 0.8125,
      "epoch": 0.4075,
      "step": 163
    },
    {
      "Batch Mean": -2.6826629638671875,
      "accuracy": 0.8125,
      "epoch": 0.4075,
      "step": 163
    },
    {
      "Batch Mean": -2.4820892810821533,
      "accuracy": 0.875,
      "epoch": 0.4075,
      "step": 163
    },
    {
      "epoch": 0.41,
      "grad_norm": 8.75256633758545,
      "learning_rate": 1.8631578947368424e-06,
      "loss": 0.3381,
      "step": 164
    },
    {
      "Batch Mean": -2.472442626953125,
      "accuracy": 0.875,
      "epoch": 0.41,
      "step": 164
    },
    {
      "Batch Mean": -1.6739931106567383,
      "accuracy": 0.875,
      "epoch": 0.41,
      "step": 164
    },
    {
      "Batch Mean": -0.72857666015625,
      "accuracy": 0.90625,
      "epoch": 0.41,
      "step": 164
    },
    {
      "Batch Mean": -3.306723117828369,
      "accuracy": 0.84375,
      "epoch": 0.41,
      "step": 164
    },
    {
      "epoch": 0.4125,
      "grad_norm": 10.206024169921875,
      "learning_rate": 1.855263157894737e-06,
      "loss": 0.3731,
      "step": 165
    },
    {
      "Batch Mean": -3.639404296875,
      "accuracy": 0.875,
      "epoch": 0.4125,
      "step": 165
    },
    {
      "Batch Mean": -2.616802215576172,
      "accuracy": 0.8125,
      "epoch": 0.4125,
      "step": 165
    },
    {
      "Batch Mean": -2.4031362533569336,
      "accuracy": 0.875,
      "epoch": 0.4125,
      "step": 165
    },
    {
      "Batch Mean": -2.225605010986328,
      "accuracy": 0.90625,
      "epoch": 0.4125,
      "step": 165
    },
    {
      "epoch": 0.415,
      "grad_norm": 9.089224815368652,
      "learning_rate": 1.8473684210526317e-06,
      "loss": 0.2955,
      "step": 166
    },
    {
      "Batch Mean": -2.55029296875,
      "accuracy": 0.875,
      "epoch": 0.415,
      "step": 166
    },
    {
      "Batch Mean": -2.450855255126953,
      "accuracy": 0.875,
      "epoch": 0.415,
      "step": 166
    },
    {
      "Batch Mean": -2.3015074729919434,
      "accuracy": 0.8125,
      "epoch": 0.415,
      "step": 166
    },
    {
      "Batch Mean": -1.580474853515625,
      "accuracy": 0.84375,
      "epoch": 0.415,
      "step": 166
    },
    {
      "epoch": 0.4175,
      "grad_norm": 8.66561222076416,
      "learning_rate": 1.8394736842105264e-06,
      "loss": 0.3365,
      "step": 167
    },
    {
      "Batch Mean": -2.9533843994140625,
      "accuracy": 0.96875,
      "epoch": 0.4175,
      "step": 167
    },
    {
      "Batch Mean": -2.373016357421875,
      "accuracy": 0.875,
      "epoch": 0.4175,
      "step": 167
    },
    {
      "Batch Mean": -0.9237747192382812,
      "accuracy": 0.875,
      "epoch": 0.4175,
      "step": 167
    },
    {
      "Batch Mean": -2.4250221252441406,
      "accuracy": 0.8125,
      "epoch": 0.4175,
      "step": 167
    },
    {
      "epoch": 0.42,
      "grad_norm": 7.428339004516602,
      "learning_rate": 1.8315789473684211e-06,
      "loss": 0.2774,
      "step": 168
    },
    {
      "Batch Mean": -3.4917097091674805,
      "accuracy": 0.84375,
      "epoch": 0.42,
      "step": 168
    },
    {
      "Batch Mean": -3.013397216796875,
      "accuracy": 0.71875,
      "epoch": 0.42,
      "step": 168
    },
    {
      "Batch Mean": -2.8196258544921875,
      "accuracy": 0.875,
      "epoch": 0.42,
      "step": 168
    },
    {
      "Batch Mean": -3.11566162109375,
      "accuracy": 0.84375,
      "epoch": 0.42,
      "step": 168
    },
    {
      "epoch": 0.4225,
      "grad_norm": 10.0361328125,
      "learning_rate": 1.8236842105263159e-06,
      "loss": 0.3954,
      "step": 169
    },
    {
      "Batch Mean": -2.8026123046875,
      "accuracy": 0.90625,
      "epoch": 0.4225,
      "step": 169
    },
    {
      "Batch Mean": -2.0240249633789062,
      "accuracy": 0.78125,
      "epoch": 0.4225,
      "step": 169
    },
    {
      "Batch Mean": -3.3115146160125732,
      "accuracy": 0.84375,
      "epoch": 0.4225,
      "step": 169
    },
    {
      "Batch Mean": -3.3820533752441406,
      "accuracy": 0.84375,
      "epoch": 0.4225,
      "step": 169
    },
    {
      "epoch": 0.425,
      "grad_norm": 8.027361869812012,
      "learning_rate": 1.8157894736842106e-06,
      "loss": 0.3511,
      "step": 170
    },
    {
      "Batch Mean": -3.3656005859375,
      "accuracy": 0.875,
      "epoch": 0.425,
      "step": 170
    },
    {
      "Batch Mean": -2.1733436584472656,
      "accuracy": 0.875,
      "epoch": 0.425,
      "step": 170
    },
    {
      "Batch Mean": -4.290916442871094,
      "accuracy": 0.75,
      "epoch": 0.425,
      "step": 170
    },
    {
      "Batch Mean": -2.6019363403320312,
      "accuracy": 0.9375,
      "epoch": 0.425,
      "step": 170
    },
    {
      "epoch": 0.4275,
      "grad_norm": 7.092214584350586,
      "learning_rate": 1.8078947368421052e-06,
      "loss": 0.2611,
      "step": 171
    },
    {
      "Batch Mean": -3.9479827880859375,
      "accuracy": 0.75,
      "epoch": 0.4275,
      "step": 171
    },
    {
      "Batch Mean": -4.346954345703125,
      "accuracy": 0.875,
      "epoch": 0.4275,
      "step": 171
    },
    {
      "Batch Mean": -4.29962158203125,
      "accuracy": 0.9375,
      "epoch": 0.4275,
      "step": 171
    },
    {
      "Batch Mean": -2.8163528442382812,
      "accuracy": 0.71875,
      "epoch": 0.4275,
      "step": 171
    },
    {
      "epoch": 0.43,
      "grad_norm": 8.232298851013184,
      "learning_rate": 1.8e-06,
      "loss": 0.3665,
      "step": 172
    },
    {
      "Batch Mean": -4.021289825439453,
      "accuracy": 0.875,
      "epoch": 0.43,
      "step": 172
    },
    {
      "Batch Mean": -4.265247344970703,
      "accuracy": 0.75,
      "epoch": 0.43,
      "step": 172
    },
    {
      "Batch Mean": -3.1769113540649414,
      "accuracy": 0.8125,
      "epoch": 0.43,
      "step": 172
    },
    {
      "Batch Mean": -3.112884521484375,
      "accuracy": 0.90625,
      "epoch": 0.43,
      "step": 172
    },
    {
      "epoch": 0.4325,
      "grad_norm": 6.9392218589782715,
      "learning_rate": 1.7921052631578947e-06,
      "loss": 0.3095,
      "step": 173
    },
    {
      "Batch Mean": -5.00799560546875,
      "accuracy": 0.78125,
      "epoch": 0.4325,
      "step": 173
    },
    {
      "Batch Mean": -4.173049449920654,
      "accuracy": 0.90625,
      "epoch": 0.4325,
      "step": 173
    },
    {
      "Batch Mean": -4.1201171875,
      "accuracy": 0.78125,
      "epoch": 0.4325,
      "step": 173
    },
    {
      "Batch Mean": -3.61419677734375,
      "accuracy": 0.84375,
      "epoch": 0.4325,
      "step": 173
    },
    {
      "epoch": 0.435,
      "grad_norm": 7.176342964172363,
      "learning_rate": 1.7842105263157894e-06,
      "loss": 0.3698,
      "step": 174
    },
    {
      "Batch Mean": -3.4647297859191895,
      "accuracy": 0.78125,
      "epoch": 0.435,
      "step": 174
    },
    {
      "Batch Mean": -3.5292892456054688,
      "accuracy": 0.875,
      "epoch": 0.435,
      "step": 174
    },
    {
      "Batch Mean": -3.6618194580078125,
      "accuracy": 0.84375,
      "epoch": 0.435,
      "step": 174
    },
    {
      "Batch Mean": -3.8795318603515625,
      "accuracy": 0.84375,
      "epoch": 0.435,
      "step": 174
    },
    {
      "epoch": 0.4375,
      "grad_norm": 7.459259510040283,
      "learning_rate": 1.7763157894736842e-06,
      "loss": 0.3615,
      "step": 175
    },
    {
      "Batch Mean": -5.0499267578125,
      "accuracy": 0.875,
      "epoch": 0.4375,
      "step": 175
    },
    {
      "Batch Mean": -3.0543928146362305,
      "accuracy": 0.84375,
      "epoch": 0.4375,
      "step": 175
    },
    {
      "Batch Mean": -4.691841125488281,
      "accuracy": 0.84375,
      "epoch": 0.4375,
      "step": 175
    },
    {
      "Batch Mean": -4.546966552734375,
      "accuracy": 0.75,
      "epoch": 0.4375,
      "step": 175
    },
    {
      "epoch": 0.44,
      "grad_norm": 6.923594951629639,
      "learning_rate": 1.768421052631579e-06,
      "loss": 0.3677,
      "step": 176
    },
    {
      "Batch Mean": -3.6771240234375,
      "accuracy": 0.875,
      "epoch": 0.44,
      "step": 176
    },
    {
      "Batch Mean": -4.0308990478515625,
      "accuracy": 0.96875,
      "epoch": 0.44,
      "step": 176
    },
    {
      "Batch Mean": -3.9250564575195312,
      "accuracy": 0.78125,
      "epoch": 0.44,
      "step": 176
    },
    {
      "Batch Mean": -4.5969390869140625,
      "accuracy": 0.90625,
      "epoch": 0.44,
      "step": 176
    },
    {
      "epoch": 0.4425,
      "grad_norm": 6.972391128540039,
      "learning_rate": 1.7605263157894739e-06,
      "loss": 0.3915,
      "step": 177
    },
    {
      "Batch Mean": -4.8792572021484375,
      "accuracy": 0.84375,
      "epoch": 0.4425,
      "step": 177
    },
    {
      "Batch Mean": -4.31494140625,
      "accuracy": 0.9375,
      "epoch": 0.4425,
      "step": 177
    },
    {
      "Batch Mean": -4.7841796875,
      "accuracy": 0.84375,
      "epoch": 0.4425,
      "step": 177
    },
    {
      "Batch Mean": -3.2111968994140625,
      "accuracy": 0.9375,
      "epoch": 0.4425,
      "step": 177
    },
    {
      "epoch": 0.445,
      "grad_norm": 6.1882805824279785,
      "learning_rate": 1.7526315789473686e-06,
      "loss": 0.2838,
      "step": 178
    },
    {
      "Batch Mean": -4.4821577072143555,
      "accuracy": 0.90625,
      "epoch": 0.445,
      "step": 178
    },
    {
      "Batch Mean": -3.9442481994628906,
      "accuracy": 0.78125,
      "epoch": 0.445,
      "step": 178
    },
    {
      "Batch Mean": -3.8130874633789062,
      "accuracy": 0.84375,
      "epoch": 0.445,
      "step": 178
    },
    {
      "Batch Mean": -4.3104095458984375,
      "accuracy": 0.875,
      "epoch": 0.445,
      "step": 178
    },
    {
      "epoch": 0.4475,
      "grad_norm": 8.147394180297852,
      "learning_rate": 1.7447368421052633e-06,
      "loss": 0.33,
      "step": 179
    },
    {
      "Batch Mean": -5.016021728515625,
      "accuracy": 0.78125,
      "epoch": 0.4475,
      "step": 179
    },
    {
      "Batch Mean": -5.16357421875,
      "accuracy": 0.75,
      "epoch": 0.4475,
      "step": 179
    },
    {
      "Batch Mean": -5.1396484375,
      "accuracy": 0.78125,
      "epoch": 0.4475,
      "step": 179
    },
    {
      "Batch Mean": -4.346986770629883,
      "accuracy": 0.875,
      "epoch": 0.4475,
      "step": 179
    },
    {
      "epoch": 0.45,
      "grad_norm": 6.089241027832031,
      "learning_rate": 1.736842105263158e-06,
      "loss": 0.3783,
      "step": 180
    },
    {
      "Batch Mean": -4.29742431640625,
      "accuracy": 0.84375,
      "epoch": 0.45,
      "step": 180
    },
    {
      "Batch Mean": -4.288475036621094,
      "accuracy": 0.875,
      "epoch": 0.45,
      "step": 180
    },
    {
      "Batch Mean": -3.9849853515625,
      "accuracy": 0.71875,
      "epoch": 0.45,
      "step": 180
    },
    {
      "Batch Mean": -3.968719482421875,
      "accuracy": 0.875,
      "epoch": 0.45,
      "step": 180
    },
    {
      "epoch": 0.4525,
      "grad_norm": 6.3386101722717285,
      "learning_rate": 1.7289473684210526e-06,
      "loss": 0.3564,
      "step": 181
    },
    {
      "Batch Mean": -4.3591766357421875,
      "accuracy": 0.90625,
      "epoch": 0.4525,
      "step": 181
    },
    {
      "Batch Mean": -3.95111083984375,
      "accuracy": 0.875,
      "epoch": 0.4525,
      "step": 181
    },
    {
      "Batch Mean": -3.8838043212890625,
      "accuracy": 0.84375,
      "epoch": 0.4525,
      "step": 181
    },
    {
      "Batch Mean": -4.525177001953125,
      "accuracy": 0.875,
      "epoch": 0.4525,
      "step": 181
    },
    {
      "epoch": 0.455,
      "grad_norm": 5.666683197021484,
      "learning_rate": 1.7210526315789474e-06,
      "loss": 0.2878,
      "step": 182
    },
    {
      "Batch Mean": -3.9508824348449707,
      "accuracy": 0.71875,
      "epoch": 0.455,
      "step": 182
    },
    {
      "Batch Mean": -4.3285064697265625,
      "accuracy": 0.8125,
      "epoch": 0.455,
      "step": 182
    },
    {
      "Batch Mean": -4.2314043045043945,
      "accuracy": 0.75,
      "epoch": 0.455,
      "step": 182
    },
    {
      "Batch Mean": -3.8064422607421875,
      "accuracy": 0.8125,
      "epoch": 0.455,
      "step": 182
    },
    {
      "epoch": 0.4575,
      "grad_norm": 8.353093147277832,
      "learning_rate": 1.7131578947368421e-06,
      "loss": 0.469,
      "step": 183
    },
    {
      "Batch Mean": -4.291778564453125,
      "accuracy": 0.78125,
      "epoch": 0.4575,
      "step": 183
    },
    {
      "Batch Mean": -4.387228488922119,
      "accuracy": 0.8125,
      "epoch": 0.4575,
      "step": 183
    },
    {
      "Batch Mean": -4.6067657470703125,
      "accuracy": 0.8125,
      "epoch": 0.4575,
      "step": 183
    },
    {
      "Batch Mean": -4.279502868652344,
      "accuracy": 0.75,
      "epoch": 0.4575,
      "step": 183
    },
    {
      "epoch": 0.46,
      "grad_norm": 6.954326629638672,
      "learning_rate": 1.7052631578947369e-06,
      "loss": 0.4025,
      "step": 184
    },
    {
      "Batch Mean": -4.3614501953125,
      "accuracy": 0.71875,
      "epoch": 0.46,
      "step": 184
    },
    {
      "Batch Mean": -5.14569091796875,
      "accuracy": 0.9375,
      "epoch": 0.46,
      "step": 184
    },
    {
      "Batch Mean": -4.415924072265625,
      "accuracy": 0.78125,
      "epoch": 0.46,
      "step": 184
    },
    {
      "Batch Mean": -4.2044525146484375,
      "accuracy": 0.78125,
      "epoch": 0.46,
      "step": 184
    },
    {
      "epoch": 0.4625,
      "grad_norm": 8.812761306762695,
      "learning_rate": 1.6973684210526316e-06,
      "loss": 0.4675,
      "step": 185
    },
    {
      "Batch Mean": -4.315742492675781,
      "accuracy": 0.78125,
      "epoch": 0.4625,
      "step": 185
    },
    {
      "Batch Mean": -2.924093246459961,
      "accuracy": 0.84375,
      "epoch": 0.4625,
      "step": 185
    },
    {
      "Batch Mean": -4.596168518066406,
      "accuracy": 0.71875,
      "epoch": 0.4625,
      "step": 185
    },
    {
      "Batch Mean": -4.245391845703125,
      "accuracy": 0.96875,
      "epoch": 0.4625,
      "step": 185
    },
    {
      "epoch": 0.465,
      "grad_norm": 6.642449378967285,
      "learning_rate": 1.6894736842105264e-06,
      "loss": 0.343,
      "step": 186
    },
    {
      "Batch Mean": -3.6098060607910156,
      "accuracy": 0.78125,
      "epoch": 0.465,
      "step": 186
    },
    {
      "Batch Mean": -3.5883445739746094,
      "accuracy": 0.96875,
      "epoch": 0.465,
      "step": 186
    },
    {
      "Batch Mean": -3.1055908203125,
      "accuracy": 0.84375,
      "epoch": 0.465,
      "step": 186
    },
    {
      "Batch Mean": -4.519744873046875,
      "accuracy": 0.90625,
      "epoch": 0.465,
      "step": 186
    },
    {
      "epoch": 0.4675,
      "grad_norm": 5.832669258117676,
      "learning_rate": 1.6815789473684209e-06,
      "loss": 0.3375,
      "step": 187
    },
    {
      "Batch Mean": -4.752597808837891,
      "accuracy": 0.75,
      "epoch": 0.4675,
      "step": 187
    },
    {
      "Batch Mean": -4.298095703125,
      "accuracy": 0.8125,
      "epoch": 0.4675,
      "step": 187
    },
    {
      "Batch Mean": -3.3523178100585938,
      "accuracy": 0.96875,
      "epoch": 0.4675,
      "step": 187
    },
    {
      "Batch Mean": -4.147176742553711,
      "accuracy": 0.9375,
      "epoch": 0.4675,
      "step": 187
    },
    {
      "epoch": 0.47,
      "grad_norm": 5.989990234375,
      "learning_rate": 1.6736842105263156e-06,
      "loss": 0.3268,
      "step": 188
    },
    {
      "Batch Mean": -3.7307815551757812,
      "accuracy": 0.90625,
      "epoch": 0.47,
      "step": 188
    },
    {
      "Batch Mean": -3.8831558227539062,
      "accuracy": 0.84375,
      "epoch": 0.47,
      "step": 188
    },
    {
      "Batch Mean": -4.1096649169921875,
      "accuracy": 0.71875,
      "epoch": 0.47,
      "step": 188
    },
    {
      "Batch Mean": -5.082061767578125,
      "accuracy": 0.75,
      "epoch": 0.47,
      "step": 188
    },
    {
      "epoch": 0.4725,
      "grad_norm": 7.336109161376953,
      "learning_rate": 1.6657894736842104e-06,
      "loss": 0.3981,
      "step": 189
    },
    {
      "Batch Mean": -4.101593017578125,
      "accuracy": 0.9375,
      "epoch": 0.4725,
      "step": 189
    },
    {
      "Batch Mean": -4.2955322265625,
      "accuracy": 0.78125,
      "epoch": 0.4725,
      "step": 189
    },
    {
      "Batch Mean": -4.952667236328125,
      "accuracy": 0.84375,
      "epoch": 0.4725,
      "step": 189
    },
    {
      "Batch Mean": -3.91229248046875,
      "accuracy": 0.90625,
      "epoch": 0.4725,
      "step": 189
    },
    {
      "epoch": 0.475,
      "grad_norm": 6.2155442237854,
      "learning_rate": 1.6578947368421056e-06,
      "loss": 0.2867,
      "step": 190
    },
    {
      "Batch Mean": -3.2933349609375,
      "accuracy": 0.84375,
      "epoch": 0.475,
      "step": 190
    },
    {
      "Batch Mean": -4.35699462890625,
      "accuracy": 0.875,
      "epoch": 0.475,
      "step": 190
    },
    {
      "Batch Mean": -3.69720458984375,
      "accuracy": 0.71875,
      "epoch": 0.475,
      "step": 190
    },
    {
      "Batch Mean": -3.815704345703125,
      "accuracy": 0.9375,
      "epoch": 0.475,
      "step": 190
    },
    {
      "epoch": 0.4775,
      "grad_norm": 8.016613960266113,
      "learning_rate": 1.65e-06,
      "loss": 0.3731,
      "step": 191
    },
    {
      "Batch Mean": -4.8086395263671875,
      "accuracy": 0.9375,
      "epoch": 0.4775,
      "step": 191
    },
    {
      "Batch Mean": -4.474151611328125,
      "accuracy": 0.875,
      "epoch": 0.4775,
      "step": 191
    },
    {
      "Batch Mean": -4.273475646972656,
      "accuracy": 0.96875,
      "epoch": 0.4775,
      "step": 191
    },
    {
      "Batch Mean": -3.0001220703125,
      "accuracy": 0.875,
      "epoch": 0.4775,
      "step": 191
    },
    {
      "epoch": 0.48,
      "grad_norm": 5.50937032699585,
      "learning_rate": 1.6421052631578948e-06,
      "loss": 0.2694,
      "step": 192
    },
    {
      "Batch Mean": -3.7930126190185547,
      "accuracy": 0.75,
      "epoch": 0.48,
      "step": 192
    },
    {
      "Batch Mean": -4.763885498046875,
      "accuracy": 0.96875,
      "epoch": 0.48,
      "step": 192
    },
    {
      "Batch Mean": -4.5084228515625,
      "accuracy": 0.875,
      "epoch": 0.48,
      "step": 192
    },
    {
      "Batch Mean": -3.997528076171875,
      "accuracy": 0.875,
      "epoch": 0.48,
      "step": 192
    },
    {
      "epoch": 0.4825,
      "grad_norm": 5.860997676849365,
      "learning_rate": 1.6342105263157896e-06,
      "loss": 0.3189,
      "step": 193
    },
    {
      "Batch Mean": -4.019233703613281,
      "accuracy": 0.875,
      "epoch": 0.4825,
      "step": 193
    },
    {
      "Batch Mean": -2.669097900390625,
      "accuracy": 0.8125,
      "epoch": 0.4825,
      "step": 193
    },
    {
      "Batch Mean": -3.9401063919067383,
      "accuracy": 0.875,
      "epoch": 0.4825,
      "step": 193
    },
    {
      "Batch Mean": -3.343963623046875,
      "accuracy": 0.78125,
      "epoch": 0.4825,
      "step": 193
    },
    {
      "epoch": 0.485,
      "grad_norm": 7.13978910446167,
      "learning_rate": 1.6263157894736843e-06,
      "loss": 0.3526,
      "step": 194
    },
    {
      "Batch Mean": -3.36846923828125,
      "accuracy": 0.6875,
      "epoch": 0.485,
      "step": 194
    },
    {
      "Batch Mean": -2.7085227966308594,
      "accuracy": 0.8125,
      "epoch": 0.485,
      "step": 194
    },
    {
      "Batch Mean": -3.2451019287109375,
      "accuracy": 0.875,
      "epoch": 0.485,
      "step": 194
    },
    {
      "Batch Mean": -3.0539398193359375,
      "accuracy": 0.78125,
      "epoch": 0.485,
      "step": 194
    },
    {
      "epoch": 0.4875,
      "grad_norm": 7.440718173980713,
      "learning_rate": 1.618421052631579e-06,
      "loss": 0.3791,
      "step": 195
    },
    {
      "Batch Mean": -3.2764711380004883,
      "accuracy": 0.78125,
      "epoch": 0.4875,
      "step": 195
    },
    {
      "Batch Mean": -3.397979736328125,
      "accuracy": 0.90625,
      "epoch": 0.4875,
      "step": 195
    },
    {
      "Batch Mean": -4.21051025390625,
      "accuracy": 0.75,
      "epoch": 0.4875,
      "step": 195
    },
    {
      "Batch Mean": -3.2050552368164062,
      "accuracy": 0.8125,
      "epoch": 0.4875,
      "step": 195
    },
    {
      "epoch": 0.49,
      "grad_norm": 7.265270233154297,
      "learning_rate": 1.6105263157894738e-06,
      "loss": 0.3847,
      "step": 196
    },
    {
      "Batch Mean": -3.199951171875,
      "accuracy": 0.875,
      "epoch": 0.49,
      "step": 196
    },
    {
      "Batch Mean": -4.533942699432373,
      "accuracy": 0.8125,
      "epoch": 0.49,
      "step": 196
    },
    {
      "Batch Mean": -3.7108230590820312,
      "accuracy": 0.84375,
      "epoch": 0.49,
      "step": 196
    },
    {
      "Batch Mean": -3.08642578125,
      "accuracy": 0.9375,
      "epoch": 0.49,
      "step": 196
    },
    {
      "epoch": 0.4925,
      "grad_norm": 7.191283226013184,
      "learning_rate": 1.6026315789473683e-06,
      "loss": 0.2875,
      "step": 197
    },
    {
      "Batch Mean": -3.5214691162109375,
      "accuracy": 0.84375,
      "epoch": 0.4925,
      "step": 197
    },
    {
      "Batch Mean": -4.107030868530273,
      "accuracy": 0.8125,
      "epoch": 0.4925,
      "step": 197
    },
    {
      "Batch Mean": -3.35797119140625,
      "accuracy": 0.9375,
      "epoch": 0.4925,
      "step": 197
    },
    {
      "Batch Mean": -3.8758201599121094,
      "accuracy": 0.875,
      "epoch": 0.4925,
      "step": 197
    },
    {
      "epoch": 0.495,
      "grad_norm": 6.3946709632873535,
      "learning_rate": 1.594736842105263e-06,
      "loss": 0.2981,
      "step": 198
    },
    {
      "Batch Mean": -2.5393829345703125,
      "accuracy": 0.96875,
      "epoch": 0.495,
      "step": 198
    },
    {
      "Batch Mean": -3.3964996337890625,
      "accuracy": 0.84375,
      "epoch": 0.495,
      "step": 198
    },
    {
      "Batch Mean": -4.194858551025391,
      "accuracy": 0.84375,
      "epoch": 0.495,
      "step": 198
    },
    {
      "Batch Mean": -4.430961608886719,
      "accuracy": 0.875,
      "epoch": 0.495,
      "step": 198
    },
    {
      "epoch": 0.4975,
      "grad_norm": 6.2794599533081055,
      "learning_rate": 1.5868421052631578e-06,
      "loss": 0.2904,
      "step": 199
    },
    {
      "Batch Mean": -3.3924102783203125,
      "accuracy": 0.96875,
      "epoch": 0.4975,
      "step": 199
    },
    {
      "Batch Mean": -3.5726470947265625,
      "accuracy": 0.875,
      "epoch": 0.4975,
      "step": 199
    },
    {
      "Batch Mean": -2.8596115112304688,
      "accuracy": 0.9375,
      "epoch": 0.4975,
      "step": 199
    },
    {
      "Batch Mean": -4.352705001831055,
      "accuracy": 0.75,
      "epoch": 0.4975,
      "step": 199
    },
    {
      "epoch": 0.5,
      "grad_norm": 6.9940595626831055,
      "learning_rate": 1.5789473684210526e-06,
      "loss": 0.2861,
      "step": 200
    },
    {
      "Batch Mean": -4.12518310546875,
      "accuracy": 0.84375,
      "epoch": 0.5,
      "step": 200
    },
    {
      "Batch Mean": -1.9889678955078125,
      "accuracy": 0.71875,
      "epoch": 0.5,
      "step": 200
    },
    {
      "Batch Mean": -4.243873596191406,
      "accuracy": 0.875,
      "epoch": 0.5,
      "step": 200
    },
    {
      "Batch Mean": -3.7152099609375,
      "accuracy": 0.875,
      "epoch": 0.5,
      "step": 200
    },
    {
      "epoch": 0.5025,
      "grad_norm": 7.452747344970703,
      "learning_rate": 1.5710526315789473e-06,
      "loss": 0.3522,
      "step": 201
    },
    {
      "Batch Mean": -3.8014869689941406,
      "accuracy": 0.9375,
      "epoch": 0.5025,
      "step": 201
    },
    {
      "Batch Mean": -3.8201751708984375,
      "accuracy": 0.90625,
      "epoch": 0.5025,
      "step": 201
    },
    {
      "Batch Mean": -4.052543640136719,
      "accuracy": 0.90625,
      "epoch": 0.5025,
      "step": 201
    },
    {
      "Batch Mean": -4.5098114013671875,
      "accuracy": 0.875,
      "epoch": 0.5025,
      "step": 201
    },
    {
      "epoch": 0.505,
      "grad_norm": 6.652421951293945,
      "learning_rate": 1.563157894736842e-06,
      "loss": 0.2527,
      "step": 202
    },
    {
      "Batch Mean": -4.42303466796875,
      "accuracy": 0.9375,
      "epoch": 0.505,
      "step": 202
    },
    {
      "Batch Mean": -3.4556884765625,
      "accuracy": 0.875,
      "epoch": 0.505,
      "step": 202
    },
    {
      "Batch Mean": -3.7808380126953125,
      "accuracy": 0.8125,
      "epoch": 0.505,
      "step": 202
    },
    {
      "Batch Mean": -3.71234130859375,
      "accuracy": 0.90625,
      "epoch": 0.505,
      "step": 202
    },
    {
      "epoch": 0.5075,
      "grad_norm": 6.491587162017822,
      "learning_rate": 1.5552631578947368e-06,
      "loss": 0.2616,
      "step": 203
    },
    {
      "Batch Mean": -3.1049137115478516,
      "accuracy": 0.84375,
      "epoch": 0.5075,
      "step": 203
    },
    {
      "Batch Mean": -4.4124908447265625,
      "accuracy": 0.8125,
      "epoch": 0.5075,
      "step": 203
    },
    {
      "Batch Mean": -3.9703140258789062,
      "accuracy": 0.875,
      "epoch": 0.5075,
      "step": 203
    },
    {
      "Batch Mean": -3.132171630859375,
      "accuracy": 0.84375,
      "epoch": 0.5075,
      "step": 203
    },
    {
      "epoch": 0.51,
      "grad_norm": 8.836193084716797,
      "learning_rate": 1.5473684210526318e-06,
      "loss": 0.3677,
      "step": 204
    },
    {
      "Batch Mean": -3.517852783203125,
      "accuracy": 0.875,
      "epoch": 0.51,
      "step": 204
    },
    {
      "Batch Mean": -3.1400108337402344,
      "accuracy": 0.75,
      "epoch": 0.51,
      "step": 204
    },
    {
      "Batch Mean": -3.8603515625,
      "accuracy": 0.84375,
      "epoch": 0.51,
      "step": 204
    },
    {
      "Batch Mean": -3.963531494140625,
      "accuracy": 0.90625,
      "epoch": 0.51,
      "step": 204
    },
    {
      "epoch": 0.5125,
      "grad_norm": 9.294251441955566,
      "learning_rate": 1.5394736842105265e-06,
      "loss": 0.3596,
      "step": 205
    },
    {
      "Batch Mean": -2.659269332885742,
      "accuracy": 0.71875,
      "epoch": 0.5125,
      "step": 205
    },
    {
      "Batch Mean": -3.768524169921875,
      "accuracy": 0.8125,
      "epoch": 0.5125,
      "step": 205
    },
    {
      "Batch Mean": -2.410369873046875,
      "accuracy": 0.78125,
      "epoch": 0.5125,
      "step": 205
    },
    {
      "Batch Mean": -4.1054840087890625,
      "accuracy": 0.84375,
      "epoch": 0.5125,
      "step": 205
    },
    {
      "epoch": 0.515,
      "grad_norm": 8.42018985748291,
      "learning_rate": 1.5315789473684213e-06,
      "loss": 0.373,
      "step": 206
    },
    {
      "Batch Mean": -3.9416961669921875,
      "accuracy": 0.96875,
      "epoch": 0.515,
      "step": 206
    },
    {
      "Batch Mean": -4.67938232421875,
      "accuracy": 0.75,
      "epoch": 0.515,
      "step": 206
    },
    {
      "Batch Mean": -3.5211868286132812,
      "accuracy": 0.75,
      "epoch": 0.515,
      "step": 206
    },
    {
      "Batch Mean": -3.649818181991577,
      "accuracy": 0.90625,
      "epoch": 0.515,
      "step": 206
    },
    {
      "epoch": 0.5175,
      "grad_norm": 8.131293296813965,
      "learning_rate": 1.5236842105263158e-06,
      "loss": 0.316,
      "step": 207
    },
    {
      "Batch Mean": -3.995166778564453,
      "accuracy": 0.84375,
      "epoch": 0.5175,
      "step": 207
    },
    {
      "Batch Mean": -3.2599563598632812,
      "accuracy": 0.90625,
      "epoch": 0.5175,
      "step": 207
    },
    {
      "Batch Mean": -4.12115478515625,
      "accuracy": 0.875,
      "epoch": 0.5175,
      "step": 207
    },
    {
      "Batch Mean": -3.947113037109375,
      "accuracy": 0.875,
      "epoch": 0.5175,
      "step": 207
    },
    {
      "epoch": 0.52,
      "grad_norm": 8.336821556091309,
      "learning_rate": 1.5157894736842105e-06,
      "loss": 0.2943,
      "step": 208
    },
    {
      "Batch Mean": -3.06756591796875,
      "accuracy": 0.875,
      "epoch": 0.52,
      "step": 208
    },
    {
      "Batch Mean": -2.88372802734375,
      "accuracy": 0.90625,
      "epoch": 0.52,
      "step": 208
    },
    {
      "Batch Mean": -2.9774584770202637,
      "accuracy": 0.84375,
      "epoch": 0.52,
      "step": 208
    },
    {
      "Batch Mean": -3.4167098999023438,
      "accuracy": 0.9375,
      "epoch": 0.52,
      "step": 208
    },
    {
      "epoch": 0.5225,
      "grad_norm": 7.195529937744141,
      "learning_rate": 1.5078947368421053e-06,
      "loss": 0.3139,
      "step": 209
    },
    {
      "Batch Mean": -3.744873046875,
      "accuracy": 0.9375,
      "epoch": 0.5225,
      "step": 209
    },
    {
      "Batch Mean": -3.461726188659668,
      "accuracy": 0.90625,
      "epoch": 0.5225,
      "step": 209
    },
    {
      "Batch Mean": -3.9631195068359375,
      "accuracy": 0.8125,
      "epoch": 0.5225,
      "step": 209
    },
    {
      "Batch Mean": -3.3913650512695312,
      "accuracy": 0.78125,
      "epoch": 0.5225,
      "step": 209
    },
    {
      "epoch": 0.525,
      "grad_norm": 7.90736722946167,
      "learning_rate": 1.5e-06,
      "loss": 0.332,
      "step": 210
    },
    {
      "Batch Mean": -3.917583465576172,
      "accuracy": 0.75,
      "epoch": 0.525,
      "step": 210
    },
    {
      "Batch Mean": -3.137542724609375,
      "accuracy": 0.9375,
      "epoch": 0.525,
      "step": 210
    },
    {
      "Batch Mean": -3.7052459716796875,
      "accuracy": 0.75,
      "epoch": 0.525,
      "step": 210
    },
    {
      "Batch Mean": -4.4464111328125,
      "accuracy": 0.84375,
      "epoch": 0.525,
      "step": 210
    },
    {
      "epoch": 0.5275,
      "grad_norm": 11.173701286315918,
      "learning_rate": 1.4921052631578948e-06,
      "loss": 0.3781,
      "step": 211
    },
    {
      "Batch Mean": -2.9488754272460938,
      "accuracy": 0.71875,
      "epoch": 0.5275,
      "step": 211
    },
    {
      "Batch Mean": -3.9233360290527344,
      "accuracy": 0.9375,
      "epoch": 0.5275,
      "step": 211
    },
    {
      "Batch Mean": -3.617340087890625,
      "accuracy": 0.875,
      "epoch": 0.5275,
      "step": 211
    },
    {
      "Batch Mean": -3.932525634765625,
      "accuracy": 0.84375,
      "epoch": 0.5275,
      "step": 211
    },
    {
      "epoch": 0.53,
      "grad_norm": 7.895691394805908,
      "learning_rate": 1.4842105263157895e-06,
      "loss": 0.3269,
      "step": 212
    },
    {
      "Batch Mean": -3.774271011352539,
      "accuracy": 0.78125,
      "epoch": 0.53,
      "step": 212
    },
    {
      "Batch Mean": -3.649505615234375,
      "accuracy": 0.84375,
      "epoch": 0.53,
      "step": 212
    },
    {
      "Batch Mean": -3.3257598876953125,
      "accuracy": 0.78125,
      "epoch": 0.53,
      "step": 212
    },
    {
      "Batch Mean": -2.99652099609375,
      "accuracy": 0.875,
      "epoch": 0.53,
      "step": 212
    },
    {
      "epoch": 0.5325,
      "grad_norm": 7.724608898162842,
      "learning_rate": 1.4763157894736843e-06,
      "loss": 0.3798,
      "step": 213
    },
    {
      "Batch Mean": -3.405029296875,
      "accuracy": 0.8125,
      "epoch": 0.5325,
      "step": 213
    },
    {
      "Batch Mean": -3.267181396484375,
      "accuracy": 0.84375,
      "epoch": 0.5325,
      "step": 213
    },
    {
      "Batch Mean": -3.329376220703125,
      "accuracy": 0.84375,
      "epoch": 0.5325,
      "step": 213
    },
    {
      "Batch Mean": -3.267730712890625,
      "accuracy": 0.84375,
      "epoch": 0.5325,
      "step": 213
    },
    {
      "epoch": 0.535,
      "grad_norm": 7.690370082855225,
      "learning_rate": 1.468421052631579e-06,
      "loss": 0.3397,
      "step": 214
    },
    {
      "Batch Mean": -3.646892547607422,
      "accuracy": 0.875,
      "epoch": 0.535,
      "step": 214
    },
    {
      "Batch Mean": -2.780792236328125,
      "accuracy": 0.8125,
      "epoch": 0.535,
      "step": 214
    },
    {
      "Batch Mean": -3.8013041019439697,
      "accuracy": 0.875,
      "epoch": 0.535,
      "step": 214
    },
    {
      "Batch Mean": -3.426971435546875,
      "accuracy": 0.875,
      "epoch": 0.535,
      "step": 214
    },
    {
      "epoch": 0.5375,
      "grad_norm": 8.358918190002441,
      "learning_rate": 1.4605263157894738e-06,
      "loss": 0.3611,
      "step": 215
    },
    {
      "Batch Mean": -3.7685012817382812,
      "accuracy": 0.90625,
      "epoch": 0.5375,
      "step": 215
    },
    {
      "Batch Mean": -3.950042724609375,
      "accuracy": 0.875,
      "epoch": 0.5375,
      "step": 215
    },
    {
      "Batch Mean": -3.021089553833008,
      "accuracy": 0.8125,
      "epoch": 0.5375,
      "step": 215
    },
    {
      "Batch Mean": -3.802825927734375,
      "accuracy": 0.75,
      "epoch": 0.5375,
      "step": 215
    },
    {
      "epoch": 0.54,
      "grad_norm": 8.37688159942627,
      "learning_rate": 1.4526315789473685e-06,
      "loss": 0.3714,
      "step": 216
    },
    {
      "Batch Mean": -2.695831298828125,
      "accuracy": 0.78125,
      "epoch": 0.54,
      "step": 216
    },
    {
      "Batch Mean": -3.305999755859375,
      "accuracy": 0.9375,
      "epoch": 0.54,
      "step": 216
    },
    {
      "Batch Mean": -4.085424423217773,
      "accuracy": 0.84375,
      "epoch": 0.54,
      "step": 216
    },
    {
      "Batch Mean": -3.650482177734375,
      "accuracy": 0.84375,
      "epoch": 0.54,
      "step": 216
    },
    {
      "epoch": 0.5425,
      "grad_norm": 7.070443153381348,
      "learning_rate": 1.4447368421052633e-06,
      "loss": 0.3216,
      "step": 217
    },
    {
      "Batch Mean": -3.5268707275390625,
      "accuracy": 0.875,
      "epoch": 0.5425,
      "step": 217
    },
    {
      "Batch Mean": -2.8703842163085938,
      "accuracy": 0.8125,
      "epoch": 0.5425,
      "step": 217
    },
    {
      "Batch Mean": -3.6777610778808594,
      "accuracy": 0.875,
      "epoch": 0.5425,
      "step": 217
    },
    {
      "Batch Mean": -3.273221015930176,
      "accuracy": 0.90625,
      "epoch": 0.5425,
      "step": 217
    },
    {
      "epoch": 0.545,
      "grad_norm": 7.268734455108643,
      "learning_rate": 1.436842105263158e-06,
      "loss": 0.3018,
      "step": 218
    },
    {
      "Batch Mean": -4.14947509765625,
      "accuracy": 0.84375,
      "epoch": 0.545,
      "step": 218
    },
    {
      "Batch Mean": -3.9094696044921875,
      "accuracy": 0.875,
      "epoch": 0.545,
      "step": 218
    },
    {
      "Batch Mean": -4.183448791503906,
      "accuracy": 0.90625,
      "epoch": 0.545,
      "step": 218
    },
    {
      "Batch Mean": -3.0260162353515625,
      "accuracy": 0.875,
      "epoch": 0.545,
      "step": 218
    },
    {
      "epoch": 0.5475,
      "grad_norm": 6.658324718475342,
      "learning_rate": 1.4289473684210525e-06,
      "loss": 0.3162,
      "step": 219
    },
    {
      "Batch Mean": -3.3106613159179688,
      "accuracy": 0.96875,
      "epoch": 0.5475,
      "step": 219
    },
    {
      "Batch Mean": -3.244659423828125,
      "accuracy": 0.90625,
      "epoch": 0.5475,
      "step": 219
    },
    {
      "Batch Mean": -3.70648193359375,
      "accuracy": 0.84375,
      "epoch": 0.5475,
      "step": 219
    },
    {
      "Batch Mean": -3.5211639404296875,
      "accuracy": 0.875,
      "epoch": 0.5475,
      "step": 219
    },
    {
      "epoch": 0.55,
      "grad_norm": 6.178283214569092,
      "learning_rate": 1.4210526315789473e-06,
      "loss": 0.3051,
      "step": 220
    },
    {
      "Batch Mean": -4.166107177734375,
      "accuracy": 0.875,
      "epoch": 0.55,
      "step": 220
    },
    {
      "Batch Mean": -3.3558349609375,
      "accuracy": 0.84375,
      "epoch": 0.55,
      "step": 220
    },
    {
      "Batch Mean": -4.33941650390625,
      "accuracy": 0.8125,
      "epoch": 0.55,
      "step": 220
    },
    {
      "Batch Mean": -4.407003402709961,
      "accuracy": 0.78125,
      "epoch": 0.55,
      "step": 220
    },
    {
      "epoch": 0.5525,
      "grad_norm": 7.264347076416016,
      "learning_rate": 1.4131578947368422e-06,
      "loss": 0.3635,
      "step": 221
    },
    {
      "Batch Mean": -3.1469879150390625,
      "accuracy": 0.78125,
      "epoch": 0.5525,
      "step": 221
    },
    {
      "Batch Mean": -3.2413787841796875,
      "accuracy": 0.75,
      "epoch": 0.5525,
      "step": 221
    },
    {
      "Batch Mean": -3.6842041015625,
      "accuracy": 0.84375,
      "epoch": 0.5525,
      "step": 221
    },
    {
      "Batch Mean": -3.679645538330078,
      "accuracy": 0.90625,
      "epoch": 0.5525,
      "step": 221
    },
    {
      "epoch": 0.555,
      "grad_norm": 7.31732177734375,
      "learning_rate": 1.405263157894737e-06,
      "loss": 0.3973,
      "step": 222
    },
    {
      "Batch Mean": -3.34716796875,
      "accuracy": 0.8125,
      "epoch": 0.555,
      "step": 222
    },
    {
      "Batch Mean": -3.9062652587890625,
      "accuracy": 0.78125,
      "epoch": 0.555,
      "step": 222
    },
    {
      "Batch Mean": -3.495391845703125,
      "accuracy": 0.84375,
      "epoch": 0.555,
      "step": 222
    },
    {
      "Batch Mean": -2.7776336669921875,
      "accuracy": 0.78125,
      "epoch": 0.555,
      "step": 222
    },
    {
      "epoch": 0.5575,
      "grad_norm": 8.380857467651367,
      "learning_rate": 1.3973684210526317e-06,
      "loss": 0.3659,
      "step": 223
    },
    {
      "Batch Mean": -3.2745742797851562,
      "accuracy": 0.90625,
      "epoch": 0.5575,
      "step": 223
    },
    {
      "Batch Mean": -3.9029083251953125,
      "accuracy": 0.875,
      "epoch": 0.5575,
      "step": 223
    },
    {
      "Batch Mean": -4.170623779296875,
      "accuracy": 0.75,
      "epoch": 0.5575,
      "step": 223
    },
    {
      "Batch Mean": -2.839531898498535,
      "accuracy": 0.9375,
      "epoch": 0.5575,
      "step": 223
    },
    {
      "epoch": 0.56,
      "grad_norm": 6.545968055725098,
      "learning_rate": 1.3894736842105263e-06,
      "loss": 0.2995,
      "step": 224
    },
    {
      "Batch Mean": -3.87921142578125,
      "accuracy": 0.84375,
      "epoch": 0.56,
      "step": 224
    },
    {
      "Batch Mean": -3.514899730682373,
      "accuracy": 0.84375,
      "epoch": 0.56,
      "step": 224
    },
    {
      "Batch Mean": -4.507991790771484,
      "accuracy": 0.84375,
      "epoch": 0.56,
      "step": 224
    },
    {
      "Batch Mean": -4.523223876953125,
      "accuracy": 0.875,
      "epoch": 0.56,
      "step": 224
    },
    {
      "epoch": 0.5625,
      "grad_norm": 8.860806465148926,
      "learning_rate": 1.381578947368421e-06,
      "loss": 0.3704,
      "step": 225
    },
    {
      "Batch Mean": -3.1665496826171875,
      "accuracy": 0.84375,
      "epoch": 0.5625,
      "step": 225
    },
    {
      "Batch Mean": -3.83599853515625,
      "accuracy": 0.875,
      "epoch": 0.5625,
      "step": 225
    },
    {
      "Batch Mean": -3.524555206298828,
      "accuracy": 0.75,
      "epoch": 0.5625,
      "step": 225
    },
    {
      "Batch Mean": -3.394775390625,
      "accuracy": 0.9375,
      "epoch": 0.5625,
      "step": 225
    },
    {
      "epoch": 0.565,
      "grad_norm": 6.077930450439453,
      "learning_rate": 1.3736842105263158e-06,
      "loss": 0.318,
      "step": 226
    },
    {
      "Batch Mean": -3.356473922729492,
      "accuracy": 0.96875,
      "epoch": 0.565,
      "step": 226
    },
    {
      "Batch Mean": -4.112098693847656,
      "accuracy": 0.8125,
      "epoch": 0.565,
      "step": 226
    },
    {
      "Batch Mean": -4.091426849365234,
      "accuracy": 0.90625,
      "epoch": 0.565,
      "step": 226
    },
    {
      "Batch Mean": -3.0035018920898438,
      "accuracy": 0.90625,
      "epoch": 0.565,
      "step": 226
    },
    {
      "epoch": 0.5675,
      "grad_norm": 6.0616631507873535,
      "learning_rate": 1.3657894736842107e-06,
      "loss": 0.2958,
      "step": 227
    },
    {
      "Batch Mean": -3.436004638671875,
      "accuracy": 0.9375,
      "epoch": 0.5675,
      "step": 227
    },
    {
      "Batch Mean": -3.202392578125,
      "accuracy": 0.84375,
      "epoch": 0.5675,
      "step": 227
    },
    {
      "Batch Mean": -3.9043502807617188,
      "accuracy": 0.65625,
      "epoch": 0.5675,
      "step": 227
    },
    {
      "Batch Mean": -3.1567840576171875,
      "accuracy": 0.75,
      "epoch": 0.5675,
      "step": 227
    },
    {
      "epoch": 0.57,
      "grad_norm": 7.531952381134033,
      "learning_rate": 1.3578947368421055e-06,
      "loss": 0.4195,
      "step": 228
    },
    {
      "Batch Mean": -3.8647918701171875,
      "accuracy": 0.8125,
      "epoch": 0.57,
      "step": 228
    },
    {
      "Batch Mean": -3.0856170654296875,
      "accuracy": 0.875,
      "epoch": 0.57,
      "step": 228
    },
    {
      "Batch Mean": -4.292762756347656,
      "accuracy": 0.90625,
      "epoch": 0.57,
      "step": 228
    },
    {
      "Batch Mean": -3.574859619140625,
      "accuracy": 0.71875,
      "epoch": 0.57,
      "step": 228
    },
    {
      "epoch": 0.5725,
      "grad_norm": 7.590925216674805,
      "learning_rate": 1.35e-06,
      "loss": 0.3986,
      "step": 229
    },
    {
      "Batch Mean": -4.2539825439453125,
      "accuracy": 0.90625,
      "epoch": 0.5725,
      "step": 229
    },
    {
      "Batch Mean": -4.0859375,
      "accuracy": 0.9375,
      "epoch": 0.5725,
      "step": 229
    },
    {
      "Batch Mean": -4.0404815673828125,
      "accuracy": 0.9375,
      "epoch": 0.5725,
      "step": 229
    },
    {
      "Batch Mean": -4.4549407958984375,
      "accuracy": 0.8125,
      "epoch": 0.5725,
      "step": 229
    },
    {
      "epoch": 0.575,
      "grad_norm": 6.640595436096191,
      "learning_rate": 1.3421052631578947e-06,
      "loss": 0.283,
      "step": 230
    },
    {
      "Batch Mean": -3.2370572090148926,
      "accuracy": 0.8125,
      "epoch": 0.575,
      "step": 230
    },
    {
      "Batch Mean": -4.13724422454834,
      "accuracy": 0.75,
      "epoch": 0.575,
      "step": 230
    },
    {
      "Batch Mean": -3.491596221923828,
      "accuracy": 0.78125,
      "epoch": 0.575,
      "step": 230
    },
    {
      "Batch Mean": -3.4756851196289062,
      "accuracy": 0.8125,
      "epoch": 0.575,
      "step": 230
    },
    {
      "epoch": 0.5775,
      "grad_norm": 8.793740272521973,
      "learning_rate": 1.3342105263157895e-06,
      "loss": 0.4038,
      "step": 231
    },
    {
      "Batch Mean": -4.296232223510742,
      "accuracy": 0.84375,
      "epoch": 0.5775,
      "step": 231
    },
    {
      "Batch Mean": -3.5285520553588867,
      "accuracy": 0.84375,
      "epoch": 0.5775,
      "step": 231
    },
    {
      "Batch Mean": -4.08984375,
      "accuracy": 0.65625,
      "epoch": 0.5775,
      "step": 231
    },
    {
      "Batch Mean": -3.6899566650390625,
      "accuracy": 0.875,
      "epoch": 0.5775,
      "step": 231
    },
    {
      "epoch": 0.58,
      "grad_norm": 7.992417335510254,
      "learning_rate": 1.3263157894736842e-06,
      "loss": 0.3747,
      "step": 232
    },
    {
      "Batch Mean": -3.7143096923828125,
      "accuracy": 0.78125,
      "epoch": 0.58,
      "step": 232
    },
    {
      "Batch Mean": -3.961048126220703,
      "accuracy": 0.84375,
      "epoch": 0.58,
      "step": 232
    },
    {
      "Batch Mean": -3.7069091796875,
      "accuracy": 0.75,
      "epoch": 0.58,
      "step": 232
    },
    {
      "Batch Mean": -3.9899253845214844,
      "accuracy": 0.84375,
      "epoch": 0.58,
      "step": 232
    },
    {
      "epoch": 0.5825,
      "grad_norm": 7.035994052886963,
      "learning_rate": 1.318421052631579e-06,
      "loss": 0.3526,
      "step": 233
    },
    {
      "Batch Mean": -3.759960174560547,
      "accuracy": 0.96875,
      "epoch": 0.5825,
      "step": 233
    },
    {
      "Batch Mean": -3.8244781494140625,
      "accuracy": 0.78125,
      "epoch": 0.5825,
      "step": 233
    },
    {
      "Batch Mean": -3.9124221801757812,
      "accuracy": 0.84375,
      "epoch": 0.5825,
      "step": 233
    },
    {
      "Batch Mean": -3.5677642822265625,
      "accuracy": 0.875,
      "epoch": 0.5825,
      "step": 233
    },
    {
      "epoch": 0.585,
      "grad_norm": 6.324237823486328,
      "learning_rate": 1.3105263157894737e-06,
      "loss": 0.2934,
      "step": 234
    },
    {
      "Batch Mean": -3.8089447021484375,
      "accuracy": 0.84375,
      "epoch": 0.585,
      "step": 234
    },
    {
      "Batch Mean": -3.088848114013672,
      "accuracy": 0.84375,
      "epoch": 0.585,
      "step": 234
    },
    {
      "Batch Mean": -3.75396728515625,
      "accuracy": 0.9375,
      "epoch": 0.585,
      "step": 234
    },
    {
      "Batch Mean": -3.730072021484375,
      "accuracy": 0.84375,
      "epoch": 0.585,
      "step": 234
    },
    {
      "epoch": 0.5875,
      "grad_norm": 6.127070426940918,
      "learning_rate": 1.3026315789473685e-06,
      "loss": 0.2932,
      "step": 235
    },
    {
      "Batch Mean": -3.4142894744873047,
      "accuracy": 0.90625,
      "epoch": 0.5875,
      "step": 235
    },
    {
      "Batch Mean": -2.889068603515625,
      "accuracy": 0.78125,
      "epoch": 0.5875,
      "step": 235
    },
    {
      "Batch Mean": -3.9111099243164062,
      "accuracy": 0.84375,
      "epoch": 0.5875,
      "step": 235
    },
    {
      "Batch Mean": -2.477039337158203,
      "accuracy": 0.875,
      "epoch": 0.5875,
      "step": 235
    },
    {
      "epoch": 0.59,
      "grad_norm": 6.313676357269287,
      "learning_rate": 1.2947368421052632e-06,
      "loss": 0.3009,
      "step": 236
    },
    {
      "Batch Mean": -3.6894073486328125,
      "accuracy": 0.96875,
      "epoch": 0.59,
      "step": 236
    },
    {
      "Batch Mean": -3.5066070556640625,
      "accuracy": 0.84375,
      "epoch": 0.59,
      "step": 236
    },
    {
      "Batch Mean": -3.67718505859375,
      "accuracy": 0.875,
      "epoch": 0.59,
      "step": 236
    },
    {
      "Batch Mean": -2.9031982421875,
      "accuracy": 0.875,
      "epoch": 0.59,
      "step": 236
    },
    {
      "epoch": 0.5925,
      "grad_norm": 5.945924758911133,
      "learning_rate": 1.286842105263158e-06,
      "loss": 0.2873,
      "step": 237
    },
    {
      "Batch Mean": -4.34759521484375,
      "accuracy": 0.875,
      "epoch": 0.5925,
      "step": 237
    },
    {
      "Batch Mean": -3.4844703674316406,
      "accuracy": 0.875,
      "epoch": 0.5925,
      "step": 237
    },
    {
      "Batch Mean": -3.9878082275390625,
      "accuracy": 0.75,
      "epoch": 0.5925,
      "step": 237
    },
    {
      "Batch Mean": -3.9766464233398438,
      "accuracy": 0.96875,
      "epoch": 0.5925,
      "step": 237
    },
    {
      "epoch": 0.595,
      "grad_norm": 6.510786533355713,
      "learning_rate": 1.2789473684210527e-06,
      "loss": 0.2744,
      "step": 238
    },
    {
      "Batch Mean": -3.688953399658203,
      "accuracy": 0.84375,
      "epoch": 0.595,
      "step": 238
    },
    {
      "Batch Mean": -3.54766845703125,
      "accuracy": 0.84375,
      "epoch": 0.595,
      "step": 238
    },
    {
      "Batch Mean": -2.0823516845703125,
      "accuracy": 0.84375,
      "epoch": 0.595,
      "step": 238
    },
    {
      "Batch Mean": -3.4054336547851562,
      "accuracy": 0.8125,
      "epoch": 0.595,
      "step": 238
    },
    {
      "epoch": 0.5975,
      "grad_norm": 6.6380228996276855,
      "learning_rate": 1.2710526315789474e-06,
      "loss": 0.2938,
      "step": 239
    },
    {
      "Batch Mean": -2.8431396484375,
      "accuracy": 0.90625,
      "epoch": 0.5975,
      "step": 239
    },
    {
      "Batch Mean": -2.9769973754882812,
      "accuracy": 0.75,
      "epoch": 0.5975,
      "step": 239
    },
    {
      "Batch Mean": -2.3684401512145996,
      "accuracy": 0.875,
      "epoch": 0.5975,
      "step": 239
    },
    {
      "Batch Mean": -1.8329415321350098,
      "accuracy": 0.8125,
      "epoch": 0.5975,
      "step": 239
    },
    {
      "epoch": 0.6,
      "grad_norm": 8.833738327026367,
      "learning_rate": 1.263157894736842e-06,
      "loss": 0.3339,
      "step": 240
    },
    {
      "Batch Mean": -2.77374267578125,
      "accuracy": 0.78125,
      "epoch": 0.6,
      "step": 240
    },
    {
      "Batch Mean": -2.96490478515625,
      "accuracy": 0.84375,
      "epoch": 0.6,
      "step": 240
    },
    {
      "Batch Mean": -3.2462158203125,
      "accuracy": 0.90625,
      "epoch": 0.6,
      "step": 240
    },
    {
      "Batch Mean": -3.204761505126953,
      "accuracy": 0.84375,
      "epoch": 0.6,
      "step": 240
    },
    {
      "epoch": 0.6025,
      "grad_norm": 7.8431501388549805,
      "learning_rate": 1.255263157894737e-06,
      "loss": 0.377,
      "step": 241
    },
    {
      "Batch Mean": -2.714019775390625,
      "accuracy": 0.8125,
      "epoch": 0.6025,
      "step": 241
    },
    {
      "Batch Mean": -2.3566207885742188,
      "accuracy": 0.9375,
      "epoch": 0.6025,
      "step": 241
    },
    {
      "Batch Mean": -2.1364517211914062,
      "accuracy": 0.8125,
      "epoch": 0.6025,
      "step": 241
    },
    {
      "Batch Mean": -2.121917724609375,
      "accuracy": 0.9375,
      "epoch": 0.6025,
      "step": 241
    },
    {
      "epoch": 0.605,
      "grad_norm": 6.37907600402832,
      "learning_rate": 1.2473684210526317e-06,
      "loss": 0.2626,
      "step": 242
    },
    {
      "Batch Mean": -2.95263671875,
      "accuracy": 0.875,
      "epoch": 0.605,
      "step": 242
    },
    {
      "Batch Mean": -3.3397293090820312,
      "accuracy": 0.84375,
      "epoch": 0.605,
      "step": 242
    },
    {
      "Batch Mean": -3.6249160766601562,
      "accuracy": 0.90625,
      "epoch": 0.605,
      "step": 242
    },
    {
      "Batch Mean": -3.3070068359375,
      "accuracy": 0.78125,
      "epoch": 0.605,
      "step": 242
    },
    {
      "epoch": 0.6075,
      "grad_norm": 7.880499362945557,
      "learning_rate": 1.2394736842105264e-06,
      "loss": 0.3124,
      "step": 243
    },
    {
      "Batch Mean": -2.3552703857421875,
      "accuracy": 0.78125,
      "epoch": 0.6075,
      "step": 243
    },
    {
      "Batch Mean": -2.283773422241211,
      "accuracy": 0.75,
      "epoch": 0.6075,
      "step": 243
    },
    {
      "Batch Mean": -3.1748132705688477,
      "accuracy": 0.78125,
      "epoch": 0.6075,
      "step": 243
    },
    {
      "Batch Mean": -1.2377548217773438,
      "accuracy": 0.875,
      "epoch": 0.6075,
      "step": 243
    },
    {
      "epoch": 0.61,
      "grad_norm": 9.455083847045898,
      "learning_rate": 1.2315789473684212e-06,
      "loss": 0.4212,
      "step": 244
    },
    {
      "Batch Mean": -2.047351837158203,
      "accuracy": 1.0,
      "epoch": 0.61,
      "step": 244
    },
    {
      "Batch Mean": -3.2393999099731445,
      "accuracy": 0.71875,
      "epoch": 0.61,
      "step": 244
    },
    {
      "Batch Mean": -2.2280426025390625,
      "accuracy": 0.84375,
      "epoch": 0.61,
      "step": 244
    },
    {
      "Batch Mean": -3.0465545654296875,
      "accuracy": 0.71875,
      "epoch": 0.61,
      "step": 244
    },
    {
      "epoch": 0.6125,
      "grad_norm": 8.660555839538574,
      "learning_rate": 1.2236842105263157e-06,
      "loss": 0.3798,
      "step": 245
    },
    {
      "Batch Mean": -2.966043472290039,
      "accuracy": 0.71875,
      "epoch": 0.6125,
      "step": 245
    },
    {
      "Batch Mean": -2.2515945434570312,
      "accuracy": 0.875,
      "epoch": 0.6125,
      "step": 245
    },
    {
      "Batch Mean": -4.3316497802734375,
      "accuracy": 0.8125,
      "epoch": 0.6125,
      "step": 245
    },
    {
      "Batch Mean": -2.0645751953125,
      "accuracy": 0.78125,
      "epoch": 0.6125,
      "step": 245
    },
    {
      "epoch": 0.615,
      "grad_norm": 11.141060829162598,
      "learning_rate": 1.2157894736842105e-06,
      "loss": 0.4981,
      "step": 246
    },
    {
      "Batch Mean": -2.4443702697753906,
      "accuracy": 0.8125,
      "epoch": 0.615,
      "step": 246
    },
    {
      "Batch Mean": -1.937652587890625,
      "accuracy": 0.875,
      "epoch": 0.615,
      "step": 246
    },
    {
      "Batch Mean": -4.549560546875,
      "accuracy": 0.8125,
      "epoch": 0.615,
      "step": 246
    },
    {
      "Batch Mean": -2.3571548461914062,
      "accuracy": 0.84375,
      "epoch": 0.615,
      "step": 246
    },
    {
      "epoch": 0.6175,
      "grad_norm": 7.784694194793701,
      "learning_rate": 1.2078947368421052e-06,
      "loss": 0.3372,
      "step": 247
    },
    {
      "Batch Mean": -3.070587158203125,
      "accuracy": 0.84375,
      "epoch": 0.6175,
      "step": 247
    },
    {
      "Batch Mean": -3.364349365234375,
      "accuracy": 0.84375,
      "epoch": 0.6175,
      "step": 247
    },
    {
      "Batch Mean": -3.2482681274414062,
      "accuracy": 0.90625,
      "epoch": 0.6175,
      "step": 247
    },
    {
      "Batch Mean": -2.4936294555664062,
      "accuracy": 0.90625,
      "epoch": 0.6175,
      "step": 247
    },
    {
      "epoch": 0.62,
      "grad_norm": 6.7486982345581055,
      "learning_rate": 1.2000000000000002e-06,
      "loss": 0.3296,
      "step": 248
    },
    {
      "Batch Mean": -2.79473876953125,
      "accuracy": 0.78125,
      "epoch": 0.62,
      "step": 248
    },
    {
      "Batch Mean": -2.90679931640625,
      "accuracy": 0.84375,
      "epoch": 0.62,
      "step": 248
    },
    {
      "Batch Mean": -3.117279052734375,
      "accuracy": 0.8125,
      "epoch": 0.62,
      "step": 248
    },
    {
      "Batch Mean": -3.5787353515625,
      "accuracy": 0.78125,
      "epoch": 0.62,
      "step": 248
    },
    {
      "epoch": 0.6225,
      "grad_norm": 8.39495849609375,
      "learning_rate": 1.192105263157895e-06,
      "loss": 0.3904,
      "step": 249
    },
    {
      "Batch Mean": -2.7425079345703125,
      "accuracy": 0.75,
      "epoch": 0.6225,
      "step": 249
    },
    {
      "Batch Mean": -4.147431373596191,
      "accuracy": 0.75,
      "epoch": 0.6225,
      "step": 249
    },
    {
      "Batch Mean": -3.2316246032714844,
      "accuracy": 0.8125,
      "epoch": 0.6225,
      "step": 249
    },
    {
      "Batch Mean": -2.8546524047851562,
      "accuracy": 0.875,
      "epoch": 0.6225,
      "step": 249
    },
    {
      "epoch": 0.625,
      "grad_norm": 7.723196029663086,
      "learning_rate": 1.1842105263157894e-06,
      "loss": 0.3809,
      "step": 250
    },
    {
      "Batch Mean": -2.8010005950927734,
      "accuracy": 0.90625,
      "epoch": 0.625,
      "step": 250
    },
    {
      "Batch Mean": -3.46478271484375,
      "accuracy": 0.875,
      "epoch": 0.625,
      "step": 250
    },
    {
      "Batch Mean": -4.16009521484375,
      "accuracy": 0.9375,
      "epoch": 0.625,
      "step": 250
    },
    {
      "Batch Mean": -3.533355712890625,
      "accuracy": 0.84375,
      "epoch": 0.625,
      "step": 250
    },
    {
      "epoch": 0.6275,
      "grad_norm": 5.696111679077148,
      "learning_rate": 1.1763157894736842e-06,
      "loss": 0.2774,
      "step": 251
    },
    {
      "Batch Mean": -3.386138916015625,
      "accuracy": 0.875,
      "epoch": 0.6275,
      "step": 251
    },
    {
      "Batch Mean": -3.6047067642211914,
      "accuracy": 0.8125,
      "epoch": 0.6275,
      "step": 251
    },
    {
      "Batch Mean": -3.3834381103515625,
      "accuracy": 0.78125,
      "epoch": 0.6275,
      "step": 251
    },
    {
      "Batch Mean": -3.3063201904296875,
      "accuracy": 0.84375,
      "epoch": 0.6275,
      "step": 251
    },
    {
      "epoch": 0.63,
      "grad_norm": 6.556052207946777,
      "learning_rate": 1.168421052631579e-06,
      "loss": 0.3641,
      "step": 252
    },
    {
      "Batch Mean": -3.0179519653320312,
      "accuracy": 0.65625,
      "epoch": 0.63,
      "step": 252
    },
    {
      "Batch Mean": -3.4812660217285156,
      "accuracy": 0.9375,
      "epoch": 0.63,
      "step": 252
    },
    {
      "Batch Mean": -3.9097061157226562,
      "accuracy": 0.90625,
      "epoch": 0.63,
      "step": 252
    },
    {
      "Batch Mean": -2.435272216796875,
      "accuracy": 0.78125,
      "epoch": 0.63,
      "step": 252
    },
    {
      "epoch": 0.6325,
      "grad_norm": 6.269940376281738,
      "learning_rate": 1.1605263157894737e-06,
      "loss": 0.3371,
      "step": 253
    },
    {
      "Batch Mean": -4.071680068969727,
      "accuracy": 0.96875,
      "epoch": 0.6325,
      "step": 253
    },
    {
      "Batch Mean": -2.9765701293945312,
      "accuracy": 0.90625,
      "epoch": 0.6325,
      "step": 253
    },
    {
      "Batch Mean": -3.8900146484375,
      "accuracy": 0.84375,
      "epoch": 0.6325,
      "step": 253
    },
    {
      "Batch Mean": -2.765000343322754,
      "accuracy": 0.90625,
      "epoch": 0.6325,
      "step": 253
    },
    {
      "epoch": 0.635,
      "grad_norm": 6.220406532287598,
      "learning_rate": 1.1526315789473684e-06,
      "loss": 0.2673,
      "step": 254
    },
    {
      "Batch Mean": -3.6123905181884766,
      "accuracy": 0.71875,
      "epoch": 0.635,
      "step": 254
    },
    {
      "Batch Mean": -4.041259765625,
      "accuracy": 0.8125,
      "epoch": 0.635,
      "step": 254
    },
    {
      "Batch Mean": -3.0314865112304688,
      "accuracy": 0.875,
      "epoch": 0.635,
      "step": 254
    },
    {
      "Batch Mean": -2.9288330078125,
      "accuracy": 0.75,
      "epoch": 0.635,
      "step": 254
    },
    {
      "epoch": 0.6375,
      "grad_norm": 8.09074592590332,
      "learning_rate": 1.1447368421052632e-06,
      "loss": 0.4475,
      "step": 255
    },
    {
      "Batch Mean": -3.0372314453125,
      "accuracy": 0.75,
      "epoch": 0.6375,
      "step": 255
    },
    {
      "Batch Mean": -2.77508544921875,
      "accuracy": 0.90625,
      "epoch": 0.6375,
      "step": 255
    },
    {
      "Batch Mean": -3.7425918579101562,
      "accuracy": 0.84375,
      "epoch": 0.6375,
      "step": 255
    },
    {
      "Batch Mean": -3.914762496948242,
      "accuracy": 0.71875,
      "epoch": 0.6375,
      "step": 255
    },
    {
      "epoch": 0.64,
      "grad_norm": 6.071317672729492,
      "learning_rate": 1.136842105263158e-06,
      "loss": 0.378,
      "step": 256
    },
    {
      "Batch Mean": -3.5071868896484375,
      "accuracy": 0.84375,
      "epoch": 0.64,
      "step": 256
    },
    {
      "Batch Mean": -3.4954605102539062,
      "accuracy": 0.90625,
      "epoch": 0.64,
      "step": 256
    },
    {
      "Batch Mean": -2.872222900390625,
      "accuracy": 0.875,
      "epoch": 0.64,
      "step": 256
    },
    {
      "Batch Mean": -3.0280609130859375,
      "accuracy": 0.78125,
      "epoch": 0.64,
      "step": 256
    },
    {
      "epoch": 0.6425,
      "grad_norm": 6.662141799926758,
      "learning_rate": 1.1289473684210527e-06,
      "loss": 0.3832,
      "step": 257
    },
    {
      "Batch Mean": -3.240509033203125,
      "accuracy": 0.875,
      "epoch": 0.6425,
      "step": 257
    },
    {
      "Batch Mean": -3.8335418701171875,
      "accuracy": 0.78125,
      "epoch": 0.6425,
      "step": 257
    },
    {
      "Batch Mean": -3.164426803588867,
      "accuracy": 0.90625,
      "epoch": 0.6425,
      "step": 257
    },
    {
      "Batch Mean": -3.7109527587890625,
      "accuracy": 0.78125,
      "epoch": 0.6425,
      "step": 257
    },
    {
      "epoch": 0.645,
      "grad_norm": 5.634103298187256,
      "learning_rate": 1.1210526315789474e-06,
      "loss": 0.3235,
      "step": 258
    },
    {
      "Batch Mean": -3.70538330078125,
      "accuracy": 0.78125,
      "epoch": 0.645,
      "step": 258
    },
    {
      "Batch Mean": -4.567169189453125,
      "accuracy": 0.8125,
      "epoch": 0.645,
      "step": 258
    },
    {
      "Batch Mean": -3.396778106689453,
      "accuracy": 0.96875,
      "epoch": 0.645,
      "step": 258
    },
    {
      "Batch Mean": -4.2032470703125,
      "accuracy": 0.84375,
      "epoch": 0.645,
      "step": 258
    },
    {
      "epoch": 0.6475,
      "grad_norm": 5.471774101257324,
      "learning_rate": 1.1131578947368421e-06,
      "loss": 0.3283,
      "step": 259
    },
    {
      "Batch Mean": -4.028106689453125,
      "accuracy": 0.875,
      "epoch": 0.6475,
      "step": 259
    },
    {
      "Batch Mean": -4.26116943359375,
      "accuracy": 0.90625,
      "epoch": 0.6475,
      "step": 259
    },
    {
      "Batch Mean": -2.7089462280273438,
      "accuracy": 0.8125,
      "epoch": 0.6475,
      "step": 259
    },
    {
      "Batch Mean": -2.54803466796875,
      "accuracy": 0.84375,
      "epoch": 0.6475,
      "step": 259
    },
    {
      "epoch": 0.65,
      "grad_norm": 5.883737564086914,
      "learning_rate": 1.1052631578947369e-06,
      "loss": 0.3364,
      "step": 260
    },
    {
      "Batch Mean": -3.8737640380859375,
      "accuracy": 0.8125,
      "epoch": 0.65,
      "step": 260
    },
    {
      "Batch Mean": -2.4150657653808594,
      "accuracy": 0.90625,
      "epoch": 0.65,
      "step": 260
    },
    {
      "Batch Mean": -3.8108320236206055,
      "accuracy": 0.8125,
      "epoch": 0.65,
      "step": 260
    },
    {
      "Batch Mean": -3.62054443359375,
      "accuracy": 0.90625,
      "epoch": 0.65,
      "step": 260
    },
    {
      "epoch": 0.6525,
      "grad_norm": 5.747334957122803,
      "learning_rate": 1.0973684210526316e-06,
      "loss": 0.3397,
      "step": 261
    },
    {
      "Batch Mean": -3.2949142456054688,
      "accuracy": 0.90625,
      "epoch": 0.6525,
      "step": 261
    },
    {
      "Batch Mean": -3.8467559814453125,
      "accuracy": 0.84375,
      "epoch": 0.6525,
      "step": 261
    },
    {
      "Batch Mean": -3.0838470458984375,
      "accuracy": 0.84375,
      "epoch": 0.6525,
      "step": 261
    },
    {
      "Batch Mean": -3.0746688842773438,
      "accuracy": 0.875,
      "epoch": 0.6525,
      "step": 261
    },
    {
      "epoch": 0.655,
      "grad_norm": 5.860691547393799,
      "learning_rate": 1.0894736842105264e-06,
      "loss": 0.3437,
      "step": 262
    },
    {
      "Batch Mean": -3.8531265258789062,
      "accuracy": 0.78125,
      "epoch": 0.655,
      "step": 262
    },
    {
      "Batch Mean": -2.9960174560546875,
      "accuracy": 0.8125,
      "epoch": 0.655,
      "step": 262
    },
    {
      "Batch Mean": -3.86285400390625,
      "accuracy": 0.9375,
      "epoch": 0.655,
      "step": 262
    },
    {
      "Batch Mean": -4.966447830200195,
      "accuracy": 0.8125,
      "epoch": 0.655,
      "step": 262
    },
    {
      "epoch": 0.6575,
      "grad_norm": 7.351145267486572,
      "learning_rate": 1.0815789473684211e-06,
      "loss": 0.3677,
      "step": 263
    },
    {
      "Batch Mean": -3.3981552124023438,
      "accuracy": 0.96875,
      "epoch": 0.6575,
      "step": 263
    },
    {
      "Batch Mean": -3.6040496826171875,
      "accuracy": 0.84375,
      "epoch": 0.6575,
      "step": 263
    },
    {
      "Batch Mean": -4.476898193359375,
      "accuracy": 0.71875,
      "epoch": 0.6575,
      "step": 263
    },
    {
      "Batch Mean": -4.4954986572265625,
      "accuracy": 0.875,
      "epoch": 0.6575,
      "step": 263
    },
    {
      "epoch": 0.66,
      "grad_norm": 5.997040748596191,
      "learning_rate": 1.0736842105263159e-06,
      "loss": 0.3175,
      "step": 264
    },
    {
      "Batch Mean": -4.134864807128906,
      "accuracy": 0.90625,
      "epoch": 0.66,
      "step": 264
    },
    {
      "Batch Mean": -3.3722915649414062,
      "accuracy": 0.8125,
      "epoch": 0.66,
      "step": 264
    },
    {
      "Batch Mean": -3.886749267578125,
      "accuracy": 0.875,
      "epoch": 0.66,
      "step": 264
    },
    {
      "Batch Mean": -3.9037017822265625,
      "accuracy": 0.78125,
      "epoch": 0.66,
      "step": 264
    },
    {
      "epoch": 0.6625,
      "grad_norm": 6.369904041290283,
      "learning_rate": 1.0657894736842106e-06,
      "loss": 0.3005,
      "step": 265
    },
    {
      "Batch Mean": -2.43988037109375,
      "accuracy": 0.84375,
      "epoch": 0.6625,
      "step": 265
    },
    {
      "Batch Mean": -3.7969512939453125,
      "accuracy": 0.78125,
      "epoch": 0.6625,
      "step": 265
    },
    {
      "Batch Mean": -2.8551597595214844,
      "accuracy": 0.78125,
      "epoch": 0.6625,
      "step": 265
    },
    {
      "Batch Mean": -4.376708984375,
      "accuracy": 0.875,
      "epoch": 0.6625,
      "step": 265
    },
    {
      "epoch": 0.665,
      "grad_norm": 7.589053630828857,
      "learning_rate": 1.0578947368421052e-06,
      "loss": 0.378,
      "step": 266
    },
    {
      "Batch Mean": -2.819580078125,
      "accuracy": 0.8125,
      "epoch": 0.665,
      "step": 266
    },
    {
      "Batch Mean": -3.7125625610351562,
      "accuracy": 0.875,
      "epoch": 0.665,
      "step": 266
    },
    {
      "Batch Mean": -3.899505615234375,
      "accuracy": 0.90625,
      "epoch": 0.665,
      "step": 266
    },
    {
      "Batch Mean": -3.084259033203125,
      "accuracy": 0.875,
      "epoch": 0.665,
      "step": 266
    },
    {
      "epoch": 0.6675,
      "grad_norm": 6.894853591918945,
      "learning_rate": 1.05e-06,
      "loss": 0.31,
      "step": 267
    },
    {
      "Batch Mean": -4.47222900390625,
      "accuracy": 0.84375,
      "epoch": 0.6675,
      "step": 267
    },
    {
      "Batch Mean": -3.929351806640625,
      "accuracy": 0.6875,
      "epoch": 0.6675,
      "step": 267
    },
    {
      "Batch Mean": -3.6689300537109375,
      "accuracy": 0.875,
      "epoch": 0.6675,
      "step": 267
    },
    {
      "Batch Mean": -3.8176116943359375,
      "accuracy": 0.90625,
      "epoch": 0.6675,
      "step": 267
    },
    {
      "epoch": 0.67,
      "grad_norm": 6.852884292602539,
      "learning_rate": 1.0421052631578949e-06,
      "loss": 0.3472,
      "step": 268
    },
    {
      "Batch Mean": -2.9620208740234375,
      "accuracy": 0.78125,
      "epoch": 0.67,
      "step": 268
    },
    {
      "Batch Mean": -2.3269195556640625,
      "accuracy": 0.84375,
      "epoch": 0.67,
      "step": 268
    },
    {
      "Batch Mean": -2.8743438720703125,
      "accuracy": 0.90625,
      "epoch": 0.67,
      "step": 268
    },
    {
      "Batch Mean": -2.6758041381835938,
      "accuracy": 0.8125,
      "epoch": 0.67,
      "step": 268
    },
    {
      "epoch": 0.6725,
      "grad_norm": 7.455336093902588,
      "learning_rate": 1.0342105263157896e-06,
      "loss": 0.3242,
      "step": 269
    },
    {
      "Batch Mean": -2.933929443359375,
      "accuracy": 0.84375,
      "epoch": 0.6725,
      "step": 269
    },
    {
      "Batch Mean": -2.581918716430664,
      "accuracy": 0.84375,
      "epoch": 0.6725,
      "step": 269
    },
    {
      "Batch Mean": -3.84466552734375,
      "accuracy": 0.875,
      "epoch": 0.6725,
      "step": 269
    },
    {
      "Batch Mean": -2.2253456115722656,
      "accuracy": 0.75,
      "epoch": 0.6725,
      "step": 269
    },
    {
      "epoch": 0.675,
      "grad_norm": 7.862179756164551,
      "learning_rate": 1.0263157894736843e-06,
      "loss": 0.3216,
      "step": 270
    },
    {
      "Batch Mean": -3.59698486328125,
      "accuracy": 0.875,
      "epoch": 0.675,
      "step": 270
    },
    {
      "Batch Mean": -3.780181884765625,
      "accuracy": 0.84375,
      "epoch": 0.675,
      "step": 270
    },
    {
      "Batch Mean": -3.2669754028320312,
      "accuracy": 0.875,
      "epoch": 0.675,
      "step": 270
    },
    {
      "Batch Mean": -2.4827651977539062,
      "accuracy": 0.90625,
      "epoch": 0.675,
      "step": 270
    },
    {
      "epoch": 0.6775,
      "grad_norm": 6.436431884765625,
      "learning_rate": 1.0184210526315789e-06,
      "loss": 0.2718,
      "step": 271
    },
    {
      "Batch Mean": -2.3623504638671875,
      "accuracy": 0.9375,
      "epoch": 0.6775,
      "step": 271
    },
    {
      "Batch Mean": -3.0159378051757812,
      "accuracy": 0.78125,
      "epoch": 0.6775,
      "step": 271
    },
    {
      "Batch Mean": -3.6621856689453125,
      "accuracy": 0.9375,
      "epoch": 0.6775,
      "step": 271
    },
    {
      "Batch Mean": -2.6310272216796875,
      "accuracy": 0.78125,
      "epoch": 0.6775,
      "step": 271
    },
    {
      "epoch": 0.68,
      "grad_norm": 7.512716293334961,
      "learning_rate": 1.0105263157894736e-06,
      "loss": 0.279,
      "step": 272
    },
    {
      "Batch Mean": -2.4361724853515625,
      "accuracy": 0.84375,
      "epoch": 0.68,
      "step": 272
    },
    {
      "Batch Mean": -2.6321334838867188,
      "accuracy": 0.90625,
      "epoch": 0.68,
      "step": 272
    },
    {
      "Batch Mean": -2.1671600341796875,
      "accuracy": 0.8125,
      "epoch": 0.68,
      "step": 272
    },
    {
      "Batch Mean": -2.6406402587890625,
      "accuracy": 0.8125,
      "epoch": 0.68,
      "step": 272
    },
    {
      "epoch": 0.6825,
      "grad_norm": 8.025399208068848,
      "learning_rate": 1.0026315789473684e-06,
      "loss": 0.3236,
      "step": 273
    },
    {
      "Batch Mean": -2.757819175720215,
      "accuracy": 0.75,
      "epoch": 0.6825,
      "step": 273
    },
    {
      "Batch Mean": -3.958160400390625,
      "accuracy": 0.90625,
      "epoch": 0.6825,
      "step": 273
    },
    {
      "Batch Mean": -2.927225112915039,
      "accuracy": 0.8125,
      "epoch": 0.6825,
      "step": 273
    },
    {
      "Batch Mean": -2.726654052734375,
      "accuracy": 0.8125,
      "epoch": 0.6825,
      "step": 273
    },
    {
      "epoch": 0.685,
      "grad_norm": 7.602370262145996,
      "learning_rate": 9.947368421052631e-07,
      "loss": 0.3538,
      "step": 274
    },
    {
      "Batch Mean": -2.7696990966796875,
      "accuracy": 0.8125,
      "epoch": 0.685,
      "step": 274
    },
    {
      "Batch Mean": -3.3070907592773438,
      "accuracy": 0.8125,
      "epoch": 0.685,
      "step": 274
    },
    {
      "Batch Mean": -2.14569091796875,
      "accuracy": 0.875,
      "epoch": 0.685,
      "step": 274
    },
    {
      "Batch Mean": -2.2299423217773438,
      "accuracy": 0.90625,
      "epoch": 0.685,
      "step": 274
    },
    {
      "epoch": 0.6875,
      "grad_norm": 7.543124675750732,
      "learning_rate": 9.86842105263158e-07,
      "loss": 0.3338,
      "step": 275
    },
    {
      "Batch Mean": -2.778106689453125,
      "accuracy": 0.90625,
      "epoch": 0.6875,
      "step": 275
    },
    {
      "Batch Mean": -2.34002685546875,
      "accuracy": 0.875,
      "epoch": 0.6875,
      "step": 275
    },
    {
      "Batch Mean": -2.372264862060547,
      "accuracy": 0.78125,
      "epoch": 0.6875,
      "step": 275
    },
    {
      "Batch Mean": -3.2867889404296875,
      "accuracy": 0.78125,
      "epoch": 0.6875,
      "step": 275
    },
    {
      "epoch": 0.69,
      "grad_norm": 7.420588493347168,
      "learning_rate": 9.789473684210526e-07,
      "loss": 0.3661,
      "step": 276
    },
    {
      "Batch Mean": -1.7787017822265625,
      "accuracy": 0.8125,
      "epoch": 0.69,
      "step": 276
    },
    {
      "Batch Mean": -2.858257293701172,
      "accuracy": 0.8125,
      "epoch": 0.69,
      "step": 276
    },
    {
      "Batch Mean": -3.493408203125,
      "accuracy": 0.84375,
      "epoch": 0.69,
      "step": 276
    },
    {
      "Batch Mean": -2.6409072875976562,
      "accuracy": 0.8125,
      "epoch": 0.69,
      "step": 276
    },
    {
      "epoch": 0.6925,
      "grad_norm": 9.459863662719727,
      "learning_rate": 9.710526315789474e-07,
      "loss": 0.4359,
      "step": 277
    },
    {
      "Batch Mean": -2.4638671875,
      "accuracy": 0.8125,
      "epoch": 0.6925,
      "step": 277
    },
    {
      "Batch Mean": -2.5003557205200195,
      "accuracy": 0.875,
      "epoch": 0.6925,
      "step": 277
    },
    {
      "Batch Mean": -3.018157958984375,
      "accuracy": 0.875,
      "epoch": 0.6925,
      "step": 277
    },
    {
      "Batch Mean": -3.19573974609375,
      "accuracy": 0.78125,
      "epoch": 0.6925,
      "step": 277
    },
    {
      "epoch": 0.695,
      "grad_norm": 7.942737102508545,
      "learning_rate": 9.63157894736842e-07,
      "loss": 0.2998,
      "step": 278
    },
    {
      "Batch Mean": -3.3094139099121094,
      "accuracy": 0.8125,
      "epoch": 0.695,
      "step": 278
    },
    {
      "Batch Mean": -2.495997905731201,
      "accuracy": 0.65625,
      "epoch": 0.695,
      "step": 278
    },
    {
      "Batch Mean": -2.2965316772460938,
      "accuracy": 0.90625,
      "epoch": 0.695,
      "step": 278
    },
    {
      "Batch Mean": -2.8788223266601562,
      "accuracy": 0.8125,
      "epoch": 0.695,
      "step": 278
    },
    {
      "epoch": 0.6975,
      "grad_norm": 8.83698558807373,
      "learning_rate": 9.552631578947368e-07,
      "loss": 0.3886,
      "step": 279
    },
    {
      "Batch Mean": -2.6972923278808594,
      "accuracy": 0.84375,
      "epoch": 0.6975,
      "step": 279
    },
    {
      "Batch Mean": -3.4521484375,
      "accuracy": 0.9375,
      "epoch": 0.6975,
      "step": 279
    },
    {
      "Batch Mean": -2.5295982360839844,
      "accuracy": 0.84375,
      "epoch": 0.6975,
      "step": 279
    },
    {
      "Batch Mean": -3.016693115234375,
      "accuracy": 0.78125,
      "epoch": 0.6975,
      "step": 279
    },
    {
      "epoch": 0.7,
      "grad_norm": 7.647303104400635,
      "learning_rate": 9.473684210526316e-07,
      "loss": 0.2891,
      "step": 280
    },
    {
      "Batch Mean": -3.5346527099609375,
      "accuracy": 0.8125,
      "epoch": 0.7,
      "step": 280
    },
    {
      "Batch Mean": -3.563547134399414,
      "accuracy": 0.71875,
      "epoch": 0.7,
      "step": 280
    },
    {
      "Batch Mean": -3.076751708984375,
      "accuracy": 0.9375,
      "epoch": 0.7,
      "step": 280
    },
    {
      "Batch Mean": -2.9289207458496094,
      "accuracy": 0.84375,
      "epoch": 0.7,
      "step": 280
    },
    {
      "epoch": 0.7025,
      "grad_norm": 8.476974487304688,
      "learning_rate": 9.394736842105262e-07,
      "loss": 0.3541,
      "step": 281
    },
    {
      "Batch Mean": -2.8744544982910156,
      "accuracy": 0.96875,
      "epoch": 0.7025,
      "step": 281
    },
    {
      "Batch Mean": -3.2609710693359375,
      "accuracy": 0.8125,
      "epoch": 0.7025,
      "step": 281
    },
    {
      "Batch Mean": -2.6938629150390625,
      "accuracy": 0.90625,
      "epoch": 0.7025,
      "step": 281
    },
    {
      "Batch Mean": -3.2952957153320312,
      "accuracy": 0.75,
      "epoch": 0.7025,
      "step": 281
    },
    {
      "epoch": 0.705,
      "grad_norm": 8.024421691894531,
      "learning_rate": 9.315789473684212e-07,
      "loss": 0.3344,
      "step": 282
    },
    {
      "Batch Mean": -2.7999496459960938,
      "accuracy": 0.90625,
      "epoch": 0.705,
      "step": 282
    },
    {
      "Batch Mean": -2.3118896484375,
      "accuracy": 0.75,
      "epoch": 0.705,
      "step": 282
    },
    {
      "Batch Mean": -3.025981903076172,
      "accuracy": 0.90625,
      "epoch": 0.705,
      "step": 282
    },
    {
      "Batch Mean": -3.867767333984375,
      "accuracy": 0.71875,
      "epoch": 0.705,
      "step": 282
    },
    {
      "epoch": 0.7075,
      "grad_norm": 8.309261322021484,
      "learning_rate": 9.236842105263158e-07,
      "loss": 0.4361,
      "step": 283
    },
    {
      "Batch Mean": -2.7422637939453125,
      "accuracy": 0.90625,
      "epoch": 0.7075,
      "step": 283
    },
    {
      "Batch Mean": -3.6142730712890625,
      "accuracy": 0.75,
      "epoch": 0.7075,
      "step": 283
    },
    {
      "Batch Mean": -3.0303115844726562,
      "accuracy": 0.84375,
      "epoch": 0.7075,
      "step": 283
    },
    {
      "Batch Mean": -3.4545669555664062,
      "accuracy": 0.9375,
      "epoch": 0.7075,
      "step": 283
    },
    {
      "epoch": 0.71,
      "grad_norm": 6.4524455070495605,
      "learning_rate": 9.157894736842106e-07,
      "loss": 0.2779,
      "step": 284
    },
    {
      "Batch Mean": -3.4946441650390625,
      "accuracy": 0.78125,
      "epoch": 0.71,
      "step": 284
    },
    {
      "Batch Mean": -2.9883852005004883,
      "accuracy": 0.96875,
      "epoch": 0.71,
      "step": 284
    },
    {
      "Batch Mean": -3.276304244995117,
      "accuracy": 0.875,
      "epoch": 0.71,
      "step": 284
    },
    {
      "Batch Mean": -3.5324325561523438,
      "accuracy": 0.9375,
      "epoch": 0.71,
      "step": 284
    },
    {
      "epoch": 0.7125,
      "grad_norm": 6.129977226257324,
      "learning_rate": 9.078947368421053e-07,
      "loss": 0.2836,
      "step": 285
    },
    {
      "Batch Mean": -3.501049041748047,
      "accuracy": 0.90625,
      "epoch": 0.7125,
      "step": 285
    },
    {
      "Batch Mean": -3.31304931640625,
      "accuracy": 0.8125,
      "epoch": 0.7125,
      "step": 285
    },
    {
      "Batch Mean": -3.850574493408203,
      "accuracy": 0.9375,
      "epoch": 0.7125,
      "step": 285
    },
    {
      "Batch Mean": -2.6171798706054688,
      "accuracy": 0.90625,
      "epoch": 0.7125,
      "step": 285
    },
    {
      "epoch": 0.715,
      "grad_norm": 6.9938459396362305,
      "learning_rate": 9e-07,
      "loss": 0.2995,
      "step": 286
    },
    {
      "Batch Mean": -3.8350830078125,
      "accuracy": 0.75,
      "epoch": 0.715,
      "step": 286
    },
    {
      "Batch Mean": -2.8526458740234375,
      "accuracy": 0.84375,
      "epoch": 0.715,
      "step": 286
    },
    {
      "Batch Mean": -3.446859836578369,
      "accuracy": 0.875,
      "epoch": 0.715,
      "step": 286
    },
    {
      "Batch Mean": -3.232410430908203,
      "accuracy": 0.90625,
      "epoch": 0.715,
      "step": 286
    },
    {
      "epoch": 0.7175,
      "grad_norm": 8.538209915161133,
      "learning_rate": 8.921052631578947e-07,
      "loss": 0.3235,
      "step": 287
    },
    {
      "Batch Mean": -3.7576637268066406,
      "accuracy": 0.78125,
      "epoch": 0.7175,
      "step": 287
    },
    {
      "Batch Mean": -3.164440155029297,
      "accuracy": 0.90625,
      "epoch": 0.7175,
      "step": 287
    },
    {
      "Batch Mean": -3.6189346313476562,
      "accuracy": 0.875,
      "epoch": 0.7175,
      "step": 287
    },
    {
      "Batch Mean": -3.303243637084961,
      "accuracy": 0.75,
      "epoch": 0.7175,
      "step": 287
    },
    {
      "epoch": 0.72,
      "grad_norm": 8.347675323486328,
      "learning_rate": 8.842105263157895e-07,
      "loss": 0.3859,
      "step": 288
    },
    {
      "Batch Mean": -3.249699831008911,
      "accuracy": 0.875,
      "epoch": 0.72,
      "step": 288
    },
    {
      "Batch Mean": -3.5829315185546875,
      "accuracy": 0.90625,
      "epoch": 0.72,
      "step": 288
    },
    {
      "Batch Mean": -3.14984130859375,
      "accuracy": 0.90625,
      "epoch": 0.72,
      "step": 288
    },
    {
      "Batch Mean": -3.806365966796875,
      "accuracy": 0.84375,
      "epoch": 0.72,
      "step": 288
    },
    {
      "epoch": 0.7225,
      "grad_norm": 6.650171279907227,
      "learning_rate": 8.763157894736843e-07,
      "loss": 0.2523,
      "step": 289
    },
    {
      "Batch Mean": -3.68951416015625,
      "accuracy": 0.90625,
      "epoch": 0.7225,
      "step": 289
    },
    {
      "Batch Mean": -3.31353759765625,
      "accuracy": 0.875,
      "epoch": 0.7225,
      "step": 289
    },
    {
      "Batch Mean": -3.738893508911133,
      "accuracy": 0.90625,
      "epoch": 0.7225,
      "step": 289
    },
    {
      "Batch Mean": -3.700286865234375,
      "accuracy": 0.75,
      "epoch": 0.7225,
      "step": 289
    },
    {
      "epoch": 0.725,
      "grad_norm": 6.943315029144287,
      "learning_rate": 8.68421052631579e-07,
      "loss": 0.3015,
      "step": 290
    },
    {
      "Batch Mean": -3.887226104736328,
      "accuracy": 0.84375,
      "epoch": 0.725,
      "step": 290
    },
    {
      "Batch Mean": -2.4902381896972656,
      "accuracy": 0.875,
      "epoch": 0.725,
      "step": 290
    },
    {
      "Batch Mean": -3.1428375244140625,
      "accuracy": 0.8125,
      "epoch": 0.725,
      "step": 290
    },
    {
      "Batch Mean": -2.7413482666015625,
      "accuracy": 0.875,
      "epoch": 0.725,
      "step": 290
    },
    {
      "epoch": 0.7275,
      "grad_norm": 7.108777046203613,
      "learning_rate": 8.605263157894737e-07,
      "loss": 0.3567,
      "step": 291
    },
    {
      "Batch Mean": -3.3324928283691406,
      "accuracy": 0.78125,
      "epoch": 0.7275,
      "step": 291
    },
    {
      "Batch Mean": -3.8308334350585938,
      "accuracy": 0.90625,
      "epoch": 0.7275,
      "step": 291
    },
    {
      "Batch Mean": -4.030181884765625,
      "accuracy": 0.84375,
      "epoch": 0.7275,
      "step": 291
    },
    {
      "Batch Mean": -3.4538726806640625,
      "accuracy": 0.84375,
      "epoch": 0.7275,
      "step": 291
    },
    {
      "epoch": 0.73,
      "grad_norm": 8.231657981872559,
      "learning_rate": 8.526315789473684e-07,
      "loss": 0.3771,
      "step": 292
    },
    {
      "Batch Mean": -3.4822425842285156,
      "accuracy": 0.75,
      "epoch": 0.73,
      "step": 292
    },
    {
      "Batch Mean": -4.32940673828125,
      "accuracy": 0.8125,
      "epoch": 0.73,
      "step": 292
    },
    {
      "Batch Mean": -4.422943115234375,
      "accuracy": 0.71875,
      "epoch": 0.73,
      "step": 292
    },
    {
      "Batch Mean": -4.202728271484375,
      "accuracy": 0.875,
      "epoch": 0.73,
      "step": 292
    },
    {
      "epoch": 0.7325,
      "grad_norm": 7.5367560386657715,
      "learning_rate": 8.447368421052632e-07,
      "loss": 0.4246,
      "step": 293
    },
    {
      "Batch Mean": -3.4039154052734375,
      "accuracy": 0.78125,
      "epoch": 0.7325,
      "step": 293
    },
    {
      "Batch Mean": -3.3543243408203125,
      "accuracy": 0.78125,
      "epoch": 0.7325,
      "step": 293
    },
    {
      "Batch Mean": -3.695354461669922,
      "accuracy": 0.8125,
      "epoch": 0.7325,
      "step": 293
    },
    {
      "Batch Mean": -3.03192138671875,
      "accuracy": 0.875,
      "epoch": 0.7325,
      "step": 293
    },
    {
      "epoch": 0.735,
      "grad_norm": 7.1592583656311035,
      "learning_rate": 8.368421052631578e-07,
      "loss": 0.3243,
      "step": 294
    },
    {
      "Batch Mean": -3.390155792236328,
      "accuracy": 0.84375,
      "epoch": 0.735,
      "step": 294
    },
    {
      "Batch Mean": -3.437049388885498,
      "accuracy": 0.90625,
      "epoch": 0.735,
      "step": 294
    },
    {
      "Batch Mean": -3.310375213623047,
      "accuracy": 0.90625,
      "epoch": 0.735,
      "step": 294
    },
    {
      "Batch Mean": -3.1389846801757812,
      "accuracy": 0.71875,
      "epoch": 0.735,
      "step": 294
    },
    {
      "epoch": 0.7375,
      "grad_norm": 7.694336891174316,
      "learning_rate": 8.289473684210528e-07,
      "loss": 0.3502,
      "step": 295
    },
    {
      "Batch Mean": -3.582538604736328,
      "accuracy": 0.90625,
      "epoch": 0.7375,
      "step": 295
    },
    {
      "Batch Mean": -3.919708251953125,
      "accuracy": 0.75,
      "epoch": 0.7375,
      "step": 295
    },
    {
      "Batch Mean": -3.33111572265625,
      "accuracy": 0.78125,
      "epoch": 0.7375,
      "step": 295
    },
    {
      "Batch Mean": -4.291971206665039,
      "accuracy": 0.90625,
      "epoch": 0.7375,
      "step": 295
    },
    {
      "epoch": 0.74,
      "grad_norm": 6.617946147918701,
      "learning_rate": 8.210526315789474e-07,
      "loss": 0.3229,
      "step": 296
    },
    {
      "Batch Mean": -2.8900604248046875,
      "accuracy": 0.84375,
      "epoch": 0.74,
      "step": 296
    },
    {
      "Batch Mean": -3.4481353759765625,
      "accuracy": 0.8125,
      "epoch": 0.74,
      "step": 296
    },
    {
      "Batch Mean": -3.49090576171875,
      "accuracy": 0.875,
      "epoch": 0.74,
      "step": 296
    },
    {
      "Batch Mean": -3.353851318359375,
      "accuracy": 0.90625,
      "epoch": 0.74,
      "step": 296
    },
    {
      "epoch": 0.7425,
      "grad_norm": 6.788773059844971,
      "learning_rate": 8.131578947368422e-07,
      "loss": 0.3667,
      "step": 297
    },
    {
      "Batch Mean": -3.618988037109375,
      "accuracy": 0.8125,
      "epoch": 0.7425,
      "step": 297
    },
    {
      "Batch Mean": -3.5942840576171875,
      "accuracy": 0.90625,
      "epoch": 0.7425,
      "step": 297
    },
    {
      "Batch Mean": -4.110076904296875,
      "accuracy": 0.8125,
      "epoch": 0.7425,
      "step": 297
    },
    {
      "Batch Mean": -3.9837026596069336,
      "accuracy": 0.8125,
      "epoch": 0.7425,
      "step": 297
    },
    {
      "epoch": 0.745,
      "grad_norm": 6.459832191467285,
      "learning_rate": 8.052631578947369e-07,
      "loss": 0.3202,
      "step": 298
    },
    {
      "Batch Mean": -3.5243988037109375,
      "accuracy": 0.84375,
      "epoch": 0.745,
      "step": 298
    },
    {
      "Batch Mean": -3.4703292846679688,
      "accuracy": 0.78125,
      "epoch": 0.745,
      "step": 298
    },
    {
      "Batch Mean": -3.6161651611328125,
      "accuracy": 0.875,
      "epoch": 0.745,
      "step": 298
    },
    {
      "Batch Mean": -3.5712432861328125,
      "accuracy": 0.90625,
      "epoch": 0.745,
      "step": 298
    },
    {
      "epoch": 0.7475,
      "grad_norm": 6.366523265838623,
      "learning_rate": 7.973684210526315e-07,
      "loss": 0.3409,
      "step": 299
    },
    {
      "Batch Mean": -3.39794921875,
      "accuracy": 0.78125,
      "epoch": 0.7475,
      "step": 299
    },
    {
      "Batch Mean": -4.16961669921875,
      "accuracy": 0.78125,
      "epoch": 0.7475,
      "step": 299
    },
    {
      "Batch Mean": -3.1703014373779297,
      "accuracy": 0.75,
      "epoch": 0.7475,
      "step": 299
    },
    {
      "Batch Mean": -2.985076904296875,
      "accuracy": 0.9375,
      "epoch": 0.7475,
      "step": 299
    },
    {
      "epoch": 0.75,
      "grad_norm": 6.7196125984191895,
      "learning_rate": 7.894736842105263e-07,
      "loss": 0.3255,
      "step": 300
    },
    {
      "Batch Mean": -3.1924591064453125,
      "accuracy": 0.875,
      "epoch": 0.75,
      "step": 300
    },
    {
      "Batch Mean": -2.8484268188476562,
      "accuracy": 0.90625,
      "epoch": 0.75,
      "step": 300
    },
    {
      "Batch Mean": -2.9788818359375,
      "accuracy": 0.875,
      "epoch": 0.75,
      "step": 300
    },
    {
      "Batch Mean": -3.427490234375,
      "accuracy": 0.78125,
      "epoch": 0.75,
      "step": 300
    },
    {
      "epoch": 0.7525,
      "grad_norm": 6.829746246337891,
      "learning_rate": 7.81578947368421e-07,
      "loss": 0.2915,
      "step": 301
    },
    {
      "Batch Mean": -4.0568695068359375,
      "accuracy": 0.6875,
      "epoch": 0.7525,
      "step": 301
    },
    {
      "Batch Mean": -3.538299560546875,
      "accuracy": 0.84375,
      "epoch": 0.7525,
      "step": 301
    },
    {
      "Batch Mean": -4.1434783935546875,
      "accuracy": 0.96875,
      "epoch": 0.7525,
      "step": 301
    },
    {
      "Batch Mean": -3.912379264831543,
      "accuracy": 0.875,
      "epoch": 0.7525,
      "step": 301
    },
    {
      "epoch": 0.755,
      "grad_norm": 6.381470680236816,
      "learning_rate": 7.736842105263159e-07,
      "loss": 0.3439,
      "step": 302
    },
    {
      "Batch Mean": -3.84674072265625,
      "accuracy": 0.84375,
      "epoch": 0.755,
      "step": 302
    },
    {
      "Batch Mean": -3.388824462890625,
      "accuracy": 0.9375,
      "epoch": 0.755,
      "step": 302
    },
    {
      "Batch Mean": -3.8634033203125,
      "accuracy": 0.8125,
      "epoch": 0.755,
      "step": 302
    },
    {
      "Batch Mean": -3.41217041015625,
      "accuracy": 0.78125,
      "epoch": 0.755,
      "step": 302
    },
    {
      "epoch": 0.7575,
      "grad_norm": 7.850198268890381,
      "learning_rate": 7.657894736842106e-07,
      "loss": 0.4118,
      "step": 303
    },
    {
      "Batch Mean": -3.5252227783203125,
      "accuracy": 0.78125,
      "epoch": 0.7575,
      "step": 303
    },
    {
      "Batch Mean": -2.8803558349609375,
      "accuracy": 0.90625,
      "epoch": 0.7575,
      "step": 303
    },
    {
      "Batch Mean": -3.93988037109375,
      "accuracy": 0.9375,
      "epoch": 0.7575,
      "step": 303
    },
    {
      "Batch Mean": -3.7777557373046875,
      "accuracy": 0.84375,
      "epoch": 0.7575,
      "step": 303
    },
    {
      "epoch": 0.76,
      "grad_norm": 6.541781425476074,
      "learning_rate": 7.578947368421053e-07,
      "loss": 0.2979,
      "step": 304
    },
    {
      "Batch Mean": -3.21771240234375,
      "accuracy": 0.78125,
      "epoch": 0.76,
      "step": 304
    },
    {
      "Batch Mean": -2.7990989685058594,
      "accuracy": 0.84375,
      "epoch": 0.76,
      "step": 304
    },
    {
      "Batch Mean": -2.48345947265625,
      "accuracy": 1.0,
      "epoch": 0.76,
      "step": 304
    },
    {
      "Batch Mean": -2.9448890686035156,
      "accuracy": 0.875,
      "epoch": 0.76,
      "step": 304
    },
    {
      "epoch": 0.7625,
      "grad_norm": 6.998548984527588,
      "learning_rate": 7.5e-07,
      "loss": 0.2772,
      "step": 305
    },
    {
      "Batch Mean": -3.30224609375,
      "accuracy": 0.84375,
      "epoch": 0.7625,
      "step": 305
    },
    {
      "Batch Mean": -4.053798675537109,
      "accuracy": 0.65625,
      "epoch": 0.7625,
      "step": 305
    },
    {
      "Batch Mean": -3.5540008544921875,
      "accuracy": 0.84375,
      "epoch": 0.7625,
      "step": 305
    },
    {
      "Batch Mean": -2.8664073944091797,
      "accuracy": 0.90625,
      "epoch": 0.7625,
      "step": 305
    },
    {
      "epoch": 0.765,
      "grad_norm": 7.624053001403809,
      "learning_rate": 7.421052631578948e-07,
      "loss": 0.4125,
      "step": 306
    },
    {
      "Batch Mean": -3.087240219116211,
      "accuracy": 0.75,
      "epoch": 0.765,
      "step": 306
    },
    {
      "Batch Mean": -2.1680727005004883,
      "accuracy": 0.71875,
      "epoch": 0.765,
      "step": 306
    },
    {
      "Batch Mean": -2.568817138671875,
      "accuracy": 0.90625,
      "epoch": 0.765,
      "step": 306
    },
    {
      "Batch Mean": -2.698699951171875,
      "accuracy": 0.6875,
      "epoch": 0.765,
      "step": 306
    },
    {
      "epoch": 0.7675,
      "grad_norm": 9.134117126464844,
      "learning_rate": 7.342105263157895e-07,
      "loss": 0.4755,
      "step": 307
    },
    {
      "Batch Mean": -3.3601226806640625,
      "accuracy": 0.71875,
      "epoch": 0.7675,
      "step": 307
    },
    {
      "Batch Mean": -3.3802719116210938,
      "accuracy": 0.84375,
      "epoch": 0.7675,
      "step": 307
    },
    {
      "Batch Mean": -3.57061767578125,
      "accuracy": 0.8125,
      "epoch": 0.7675,
      "step": 307
    },
    {
      "Batch Mean": -3.086669921875,
      "accuracy": 0.875,
      "epoch": 0.7675,
      "step": 307
    },
    {
      "epoch": 0.77,
      "grad_norm": 7.370970249176025,
      "learning_rate": 7.263157894736843e-07,
      "loss": 0.3724,
      "step": 308
    },
    {
      "Batch Mean": -3.758331298828125,
      "accuracy": 0.78125,
      "epoch": 0.77,
      "step": 308
    },
    {
      "Batch Mean": -3.1384429931640625,
      "accuracy": 0.875,
      "epoch": 0.77,
      "step": 308
    },
    {
      "Batch Mean": -3.933807373046875,
      "accuracy": 0.84375,
      "epoch": 0.77,
      "step": 308
    },
    {
      "Batch Mean": -1.7305450439453125,
      "accuracy": 0.875,
      "epoch": 0.77,
      "step": 308
    },
    {
      "epoch": 0.7725,
      "grad_norm": 6.031234264373779,
      "learning_rate": 7.18421052631579e-07,
      "loss": 0.3028,
      "step": 309
    },
    {
      "Batch Mean": -2.9827880859375,
      "accuracy": 0.875,
      "epoch": 0.7725,
      "step": 309
    },
    {
      "Batch Mean": -2.880035400390625,
      "accuracy": 0.8125,
      "epoch": 0.7725,
      "step": 309
    },
    {
      "Batch Mean": -3.112762451171875,
      "accuracy": 0.9375,
      "epoch": 0.7725,
      "step": 309
    },
    {
      "Batch Mean": -3.1557464599609375,
      "accuracy": 0.84375,
      "epoch": 0.7725,
      "step": 309
    },
    {
      "epoch": 0.775,
      "grad_norm": 6.867088794708252,
      "learning_rate": 7.105263157894736e-07,
      "loss": 0.3139,
      "step": 310
    },
    {
      "Batch Mean": -2.5328445434570312,
      "accuracy": 0.90625,
      "epoch": 0.775,
      "step": 310
    },
    {
      "Batch Mean": -2.6245956420898438,
      "accuracy": 0.875,
      "epoch": 0.775,
      "step": 310
    },
    {
      "Batch Mean": -2.5271682739257812,
      "accuracy": 0.78125,
      "epoch": 0.775,
      "step": 310
    },
    {
      "Batch Mean": -2.7974395751953125,
      "accuracy": 0.78125,
      "epoch": 0.775,
      "step": 310
    },
    {
      "epoch": 0.7775,
      "grad_norm": 6.109833717346191,
      "learning_rate": 7.026315789473685e-07,
      "loss": 0.2965,
      "step": 311
    },
    {
      "Batch Mean": -3.696605682373047,
      "accuracy": 0.78125,
      "epoch": 0.7775,
      "step": 311
    },
    {
      "Batch Mean": -2.560495376586914,
      "accuracy": 0.84375,
      "epoch": 0.7775,
      "step": 311
    },
    {
      "Batch Mean": -2.85882568359375,
      "accuracy": 0.84375,
      "epoch": 0.7775,
      "step": 311
    },
    {
      "Batch Mean": -3.4066162109375,
      "accuracy": 0.78125,
      "epoch": 0.7775,
      "step": 311
    },
    {
      "epoch": 0.78,
      "grad_norm": 8.343950271606445,
      "learning_rate": 6.947368421052631e-07,
      "loss": 0.3525,
      "step": 312
    },
    {
      "Batch Mean": -3.6371707916259766,
      "accuracy": 0.90625,
      "epoch": 0.78,
      "step": 312
    },
    {
      "Batch Mean": -3.43695068359375,
      "accuracy": 0.84375,
      "epoch": 0.78,
      "step": 312
    },
    {
      "Batch Mean": -2.969696044921875,
      "accuracy": 0.875,
      "epoch": 0.78,
      "step": 312
    },
    {
      "Batch Mean": -2.048126220703125,
      "accuracy": 0.6875,
      "epoch": 0.78,
      "step": 312
    },
    {
      "epoch": 0.7825,
      "grad_norm": 7.288078308105469,
      "learning_rate": 6.868421052631579e-07,
      "loss": 0.3558,
      "step": 313
    },
    {
      "Batch Mean": -1.7478790283203125,
      "accuracy": 0.90625,
      "epoch": 0.7825,
      "step": 313
    },
    {
      "Batch Mean": -3.667694091796875,
      "accuracy": 0.8125,
      "epoch": 0.7825,
      "step": 313
    },
    {
      "Batch Mean": -2.718048095703125,
      "accuracy": 0.90625,
      "epoch": 0.7825,
      "step": 313
    },
    {
      "Batch Mean": -2.8699588775634766,
      "accuracy": 0.875,
      "epoch": 0.7825,
      "step": 313
    },
    {
      "epoch": 0.785,
      "grad_norm": 5.961294174194336,
      "learning_rate": 6.789473684210527e-07,
      "loss": 0.2646,
      "step": 314
    },
    {
      "Batch Mean": -3.25496768951416,
      "accuracy": 0.75,
      "epoch": 0.785,
      "step": 314
    },
    {
      "Batch Mean": -2.7613067626953125,
      "accuracy": 0.84375,
      "epoch": 0.785,
      "step": 314
    },
    {
      "Batch Mean": -2.9433860778808594,
      "accuracy": 0.84375,
      "epoch": 0.785,
      "step": 314
    },
    {
      "Batch Mean": -2.6829519271850586,
      "accuracy": 0.875,
      "epoch": 0.785,
      "step": 314
    },
    {
      "epoch": 0.7875,
      "grad_norm": 7.829719066619873,
      "learning_rate": 6.710526315789474e-07,
      "loss": 0.3657,
      "step": 315
    },
    {
      "Batch Mean": -3.2679595947265625,
      "accuracy": 0.875,
      "epoch": 0.7875,
      "step": 315
    },
    {
      "Batch Mean": -2.335235595703125,
      "accuracy": 0.84375,
      "epoch": 0.7875,
      "step": 315
    },
    {
      "Batch Mean": -3.3640594482421875,
      "accuracy": 0.84375,
      "epoch": 0.7875,
      "step": 315
    },
    {
      "Batch Mean": -2.4273529052734375,
      "accuracy": 0.90625,
      "epoch": 0.7875,
      "step": 315
    },
    {
      "epoch": 0.79,
      "grad_norm": 7.626279830932617,
      "learning_rate": 6.631578947368421e-07,
      "loss": 0.3543,
      "step": 316
    },
    {
      "Batch Mean": -2.5838470458984375,
      "accuracy": 0.9375,
      "epoch": 0.79,
      "step": 316
    },
    {
      "Batch Mean": -3.27642822265625,
      "accuracy": 0.78125,
      "epoch": 0.79,
      "step": 316
    },
    {
      "Batch Mean": -2.986175537109375,
      "accuracy": 0.875,
      "epoch": 0.79,
      "step": 316
    },
    {
      "Batch Mean": -1.9454727172851562,
      "accuracy": 0.875,
      "epoch": 0.79,
      "step": 316
    },
    {
      "epoch": 0.7925,
      "grad_norm": 6.644615650177002,
      "learning_rate": 6.552631578947369e-07,
      "loss": 0.2865,
      "step": 317
    },
    {
      "Batch Mean": -2.931793212890625,
      "accuracy": 0.875,
      "epoch": 0.7925,
      "step": 317
    },
    {
      "Batch Mean": -2.5732688903808594,
      "accuracy": 0.875,
      "epoch": 0.7925,
      "step": 317
    },
    {
      "Batch Mean": -2.371246337890625,
      "accuracy": 0.8125,
      "epoch": 0.7925,
      "step": 317
    },
    {
      "Batch Mean": -2.889852523803711,
      "accuracy": 0.8125,
      "epoch": 0.7925,
      "step": 317
    },
    {
      "epoch": 0.795,
      "grad_norm": 7.724899768829346,
      "learning_rate": 6.473684210526316e-07,
      "loss": 0.3715,
      "step": 318
    },
    {
      "Batch Mean": -2.809967041015625,
      "accuracy": 0.875,
      "epoch": 0.795,
      "step": 318
    },
    {
      "Batch Mean": -2.647411346435547,
      "accuracy": 0.8125,
      "epoch": 0.795,
      "step": 318
    },
    {
      "Batch Mean": -3.0574254989624023,
      "accuracy": 0.875,
      "epoch": 0.795,
      "step": 318
    },
    {
      "Batch Mean": -2.959758758544922,
      "accuracy": 0.875,
      "epoch": 0.795,
      "step": 318
    },
    {
      "epoch": 0.7975,
      "grad_norm": 6.657589912414551,
      "learning_rate": 6.394736842105264e-07,
      "loss": 0.329,
      "step": 319
    },
    {
      "Batch Mean": -2.9253692626953125,
      "accuracy": 0.84375,
      "epoch": 0.7975,
      "step": 319
    },
    {
      "Batch Mean": -3.6897506713867188,
      "accuracy": 0.875,
      "epoch": 0.7975,
      "step": 319
    },
    {
      "Batch Mean": -2.2700228691101074,
      "accuracy": 0.78125,
      "epoch": 0.7975,
      "step": 319
    },
    {
      "Batch Mean": -2.148712158203125,
      "accuracy": 0.9375,
      "epoch": 0.7975,
      "step": 319
    },
    {
      "epoch": 0.8,
      "grad_norm": 7.154412746429443,
      "learning_rate": 6.31578947368421e-07,
      "loss": 0.3156,
      "step": 320
    }
  ],
  "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": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}