{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.5,
  "eval_steps": 500,
  "global_step": 200,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "Batch Mean": -3.4444580078125,
      "accuracy": 0.40625,
      "epoch": 0,
      "step": 0
    },
    {
      "epoch": 0.0025,
      "grad_norm": 43.31455612182617,
      "learning_rate": 1.5000000000000002e-07,
      "loss": 0.9077,
      "step": 1
    },
    {
      "Batch Mean": -3.43255615234375,
      "accuracy": 0.4921875,
      "epoch": 0.0025,
      "step": 1
    },
    {
      "epoch": 0.005,
      "grad_norm": 50.06682205200195,
      "learning_rate": 3.0000000000000004e-07,
      "loss": 0.8907,
      "step": 2
    },
    {
      "Batch Mean": -3.44775390625,
      "accuracy": 0.453125,
      "epoch": 0.005,
      "step": 2
    },
    {
      "epoch": 0.0075,
      "grad_norm": 47.389522552490234,
      "learning_rate": 4.5e-07,
      "loss": 0.8899,
      "step": 3
    },
    {
      "Batch Mean": -3.4224853515625,
      "accuracy": 0.4375,
      "epoch": 0.0075,
      "step": 3
    },
    {
      "epoch": 0.01,
      "grad_norm": 49.24280548095703,
      "learning_rate": 6.000000000000001e-07,
      "loss": 0.8833,
      "step": 4
    },
    {
      "Batch Mean": -3.39697265625,
      "accuracy": 0.5234375,
      "epoch": 0.01,
      "step": 4
    },
    {
      "epoch": 0.0125,
      "grad_norm": 44.11930847167969,
      "learning_rate": 7.5e-07,
      "loss": 0.8661,
      "step": 5
    },
    {
      "Batch Mean": -3.38104248046875,
      "accuracy": 0.5078125,
      "epoch": 0.0125,
      "step": 5
    },
    {
      "epoch": 0.015,
      "grad_norm": 44.9741325378418,
      "learning_rate": 9e-07,
      "loss": 0.8805,
      "step": 6
    },
    {
      "Batch Mean": -3.19671630859375,
      "accuracy": 0.5078125,
      "epoch": 0.015,
      "step": 6
    },
    {
      "epoch": 0.0175,
      "grad_norm": 42.32276153564453,
      "learning_rate": 1.05e-06,
      "loss": 0.8547,
      "step": 7
    },
    {
      "Batch Mean": -3.1279296875,
      "accuracy": 0.4921875,
      "epoch": 0.0175,
      "step": 7
    },
    {
      "epoch": 0.02,
      "grad_norm": 44.22520065307617,
      "learning_rate": 1.2000000000000002e-06,
      "loss": 0.8464,
      "step": 8
    },
    {
      "Batch Mean": -2.520538330078125,
      "accuracy": 0.5,
      "epoch": 0.02,
      "step": 8
    },
    {
      "epoch": 0.0225,
      "grad_norm": 36.41325378417969,
      "learning_rate": 1.35e-06,
      "loss": 0.7946,
      "step": 9
    },
    {
      "Batch Mean": -2.413787841796875,
      "accuracy": 0.515625,
      "epoch": 0.0225,
      "step": 9
    },
    {
      "epoch": 0.025,
      "grad_norm": 31.74874496459961,
      "learning_rate": 1.5e-06,
      "loss": 0.7939,
      "step": 10
    },
    {
      "Batch Mean": -2.218231201171875,
      "accuracy": 0.4921875,
      "epoch": 0.025,
      "step": 10
    },
    {
      "epoch": 0.0275,
      "grad_norm": 34.12691879272461,
      "learning_rate": 1.65e-06,
      "loss": 0.7505,
      "step": 11
    },
    {
      "Batch Mean": -0.2393415868282318,
      "accuracy": 0.53125,
      "epoch": 0.0275,
      "step": 11
    },
    {
      "epoch": 0.03,
      "grad_norm": 10.481990814208984,
      "learning_rate": 1.8e-06,
      "loss": 0.6779,
      "step": 12
    },
    {
      "Batch Mean": 0.2733480930328369,
      "accuracy": 0.5703125,
      "epoch": 0.03,
      "step": 12
    },
    {
      "epoch": 0.0325,
      "grad_norm": 8.39962387084961,
      "learning_rate": 1.95e-06,
      "loss": 0.6728,
      "step": 13
    },
    {
      "Batch Mean": 0.5041149854660034,
      "accuracy": 0.671875,
      "epoch": 0.0325,
      "step": 13
    },
    {
      "epoch": 0.035,
      "grad_norm": 15.906760215759277,
      "learning_rate": 2.1e-06,
      "loss": 0.6679,
      "step": 14
    },
    {
      "Batch Mean": 0.8821840286254883,
      "accuracy": 0.578125,
      "epoch": 0.035,
      "step": 14
    },
    {
      "epoch": 0.0375,
      "grad_norm": 19.03310775756836,
      "learning_rate": 2.25e-06,
      "loss": 0.7024,
      "step": 15
    },
    {
      "Batch Mean": 1.1891746520996094,
      "accuracy": 0.5859375,
      "epoch": 0.0375,
      "step": 15
    },
    {
      "epoch": 0.04,
      "grad_norm": 20.89366340637207,
      "learning_rate": 2.4000000000000003e-06,
      "loss": 0.6868,
      "step": 16
    },
    {
      "Batch Mean": 2.1533203125,
      "accuracy": 0.671875,
      "epoch": 0.04,
      "step": 16
    },
    {
      "epoch": 0.0425,
      "grad_norm": 34.633541107177734,
      "learning_rate": 2.55e-06,
      "loss": 0.7297,
      "step": 17
    },
    {
      "Batch Mean": 2.5652008056640625,
      "accuracy": 0.6484375,
      "epoch": 0.0425,
      "step": 17
    },
    {
      "epoch": 0.045,
      "grad_norm": 41.46820068359375,
      "learning_rate": 2.7e-06,
      "loss": 0.7387,
      "step": 18
    },
    {
      "Batch Mean": 2.4411849975585938,
      "accuracy": 0.75,
      "epoch": 0.045,
      "step": 18
    },
    {
      "epoch": 0.0475,
      "grad_norm": 42.9193000793457,
      "learning_rate": 2.85e-06,
      "loss": 0.6813,
      "step": 19
    },
    {
      "Batch Mean": 2.1285018920898438,
      "accuracy": 0.703125,
      "epoch": 0.0475,
      "step": 19
    },
    {
      "epoch": 0.05,
      "grad_norm": 37.16952133178711,
      "learning_rate": 3e-06,
      "loss": 0.672,
      "step": 20
    },
    {
      "Batch Mean": 1.4230481386184692,
      "accuracy": 0.671875,
      "epoch": 0.05,
      "step": 20
    },
    {
      "epoch": 0.0525,
      "grad_norm": 27.2924747467041,
      "learning_rate": 2.992105263157895e-06,
      "loss": 0.647,
      "step": 21
    },
    {
      "Batch Mean": 0.5414033532142639,
      "accuracy": 0.6953125,
      "epoch": 0.0525,
      "step": 21
    },
    {
      "epoch": 0.055,
      "grad_norm": 11.927323341369629,
      "learning_rate": 2.9842105263157896e-06,
      "loss": 0.5981,
      "step": 22
    },
    {
      "Batch Mean": -0.36795324087142944,
      "accuracy": 0.7265625,
      "epoch": 0.055,
      "step": 22
    },
    {
      "epoch": 0.0575,
      "grad_norm": 13.455949783325195,
      "learning_rate": 2.9763157894736843e-06,
      "loss": 0.5763,
      "step": 23
    },
    {
      "Batch Mean": -1.0874066352844238,
      "accuracy": 0.7421875,
      "epoch": 0.0575,
      "step": 23
    },
    {
      "epoch": 0.06,
      "grad_norm": 22.382274627685547,
      "learning_rate": 2.968421052631579e-06,
      "loss": 0.5763,
      "step": 24
    },
    {
      "Batch Mean": -1.103028416633606,
      "accuracy": 0.6953125,
      "epoch": 0.06,
      "step": 24
    },
    {
      "epoch": 0.0625,
      "grad_norm": 19.854228973388672,
      "learning_rate": 2.960526315789474e-06,
      "loss": 0.6065,
      "step": 25
    },
    {
      "Batch Mean": -0.8232302069664001,
      "accuracy": 0.734375,
      "epoch": 0.0625,
      "step": 25
    },
    {
      "epoch": 0.065,
      "grad_norm": 16.642824172973633,
      "learning_rate": 2.9526315789473685e-06,
      "loss": 0.5282,
      "step": 26
    },
    {
      "Batch Mean": -0.11966800689697266,
      "accuracy": 0.6875,
      "epoch": 0.065,
      "step": 26
    },
    {
      "epoch": 0.0675,
      "grad_norm": 8.563494682312012,
      "learning_rate": 2.9447368421052633e-06,
      "loss": 0.5348,
      "step": 27
    },
    {
      "Batch Mean": 0.5674030780792236,
      "accuracy": 0.671875,
      "epoch": 0.0675,
      "step": 27
    },
    {
      "epoch": 0.07,
      "grad_norm": 13.211753845214844,
      "learning_rate": 2.936842105263158e-06,
      "loss": 0.5812,
      "step": 28
    },
    {
      "Batch Mean": 1.1245126724243164,
      "accuracy": 0.6171875,
      "epoch": 0.07,
      "step": 28
    },
    {
      "epoch": 0.0725,
      "grad_norm": 22.8548583984375,
      "learning_rate": 2.9289473684210528e-06,
      "loss": 0.7336,
      "step": 29
    },
    {
      "Batch Mean": 1.2212319374084473,
      "accuracy": 0.6796875,
      "epoch": 0.0725,
      "step": 29
    },
    {
      "epoch": 0.075,
      "grad_norm": 21.96833038330078,
      "learning_rate": 2.9210526315789475e-06,
      "loss": 0.6167,
      "step": 30
    },
    {
      "Batch Mean": 0.7509702444076538,
      "accuracy": 0.796875,
      "epoch": 0.075,
      "step": 30
    },
    {
      "epoch": 0.0775,
      "grad_norm": 12.47424030303955,
      "learning_rate": 2.9131578947368423e-06,
      "loss": 0.463,
      "step": 31
    },
    {
      "Batch Mean": 0.2412339150905609,
      "accuracy": 0.7578125,
      "epoch": 0.0775,
      "step": 31
    },
    {
      "epoch": 0.08,
      "grad_norm": 9.769195556640625,
      "learning_rate": 2.905263157894737e-06,
      "loss": 0.5412,
      "step": 32
    },
    {
      "Batch Mean": -0.5602660179138184,
      "accuracy": 0.7109375,
      "epoch": 0.08,
      "step": 32
    },
    {
      "epoch": 0.0825,
      "grad_norm": 11.323891639709473,
      "learning_rate": 2.8973684210526318e-06,
      "loss": 0.5601,
      "step": 33
    },
    {
      "Batch Mean": -0.8669416904449463,
      "accuracy": 0.7734375,
      "epoch": 0.0825,
      "step": 33
    },
    {
      "epoch": 0.085,
      "grad_norm": 15.461758613586426,
      "learning_rate": 2.8894736842105265e-06,
      "loss": 0.5147,
      "step": 34
    },
    {
      "Batch Mean": -0.7635477781295776,
      "accuracy": 0.75,
      "epoch": 0.085,
      "step": 34
    },
    {
      "epoch": 0.0875,
      "grad_norm": 14.736496925354004,
      "learning_rate": 2.8815789473684213e-06,
      "loss": 0.5421,
      "step": 35
    },
    {
      "Batch Mean": -0.5088728070259094,
      "accuracy": 0.734375,
      "epoch": 0.0875,
      "step": 35
    },
    {
      "epoch": 0.09,
      "grad_norm": 11.982501029968262,
      "learning_rate": 2.873684210526316e-06,
      "loss": 0.5234,
      "step": 36
    },
    {
      "Batch Mean": -0.03705340623855591,
      "accuracy": 0.7578125,
      "epoch": 0.09,
      "step": 36
    },
    {
      "epoch": 0.0925,
      "grad_norm": 7.854848861694336,
      "learning_rate": 2.8657894736842103e-06,
      "loss": 0.4996,
      "step": 37
    },
    {
      "Batch Mean": 0.5426892042160034,
      "accuracy": 0.734375,
      "epoch": 0.0925,
      "step": 37
    },
    {
      "epoch": 0.095,
      "grad_norm": 10.834790229797363,
      "learning_rate": 2.857894736842105e-06,
      "loss": 0.5187,
      "step": 38
    },
    {
      "Batch Mean": 0.4622654914855957,
      "accuracy": 0.7734375,
      "epoch": 0.095,
      "step": 38
    },
    {
      "epoch": 0.0975,
      "grad_norm": 10.764225959777832,
      "learning_rate": 2.85e-06,
      "loss": 0.4723,
      "step": 39
    },
    {
      "Batch Mean": 0.18012571334838867,
      "accuracy": 0.7890625,
      "epoch": 0.0975,
      "step": 39
    },
    {
      "epoch": 0.1,
      "grad_norm": 9.500924110412598,
      "learning_rate": 2.8421052631578946e-06,
      "loss": 0.4545,
      "step": 40
    },
    {
      "Batch Mean": -0.05992317199707031,
      "accuracy": 0.7890625,
      "epoch": 0.1,
      "step": 40
    },
    {
      "epoch": 0.1025,
      "grad_norm": 8.079994201660156,
      "learning_rate": 2.8342105263157897e-06,
      "loss": 0.49,
      "step": 41
    },
    {
      "Batch Mean": 0.04392993450164795,
      "accuracy": 0.8515625,
      "epoch": 0.1025,
      "step": 41
    },
    {
      "epoch": 0.105,
      "grad_norm": 10.184918403625488,
      "learning_rate": 2.8263157894736845e-06,
      "loss": 0.3789,
      "step": 42
    },
    {
      "Batch Mean": -0.2515716552734375,
      "accuracy": 0.8046875,
      "epoch": 0.105,
      "step": 42
    },
    {
      "epoch": 0.1075,
      "grad_norm": 9.894248008728027,
      "learning_rate": 2.8184210526315792e-06,
      "loss": 0.435,
      "step": 43
    },
    {
      "Batch Mean": 0.15310335159301758,
      "accuracy": 0.7734375,
      "epoch": 0.1075,
      "step": 43
    },
    {
      "epoch": 0.11,
      "grad_norm": 10.671575546264648,
      "learning_rate": 2.810526315789474e-06,
      "loss": 0.4291,
      "step": 44
    },
    {
      "Batch Mean": 0.2563772201538086,
      "accuracy": 0.765625,
      "epoch": 0.11,
      "step": 44
    },
    {
      "epoch": 0.1125,
      "grad_norm": 10.690945625305176,
      "learning_rate": 2.8026315789473687e-06,
      "loss": 0.4812,
      "step": 45
    },
    {
      "Batch Mean": 0.046463966369628906,
      "accuracy": 0.875,
      "epoch": 0.1125,
      "step": 45
    },
    {
      "epoch": 0.115,
      "grad_norm": 9.616545677185059,
      "learning_rate": 2.7947368421052635e-06,
      "loss": 0.3082,
      "step": 46
    },
    {
      "Batch Mean": 0.11115455627441406,
      "accuracy": 0.75,
      "epoch": 0.115,
      "step": 46
    },
    {
      "epoch": 0.1175,
      "grad_norm": 15.726456642150879,
      "learning_rate": 2.7868421052631578e-06,
      "loss": 0.5576,
      "step": 47
    },
    {
      "Batch Mean": 0.2878689765930176,
      "accuracy": 0.8125,
      "epoch": 0.1175,
      "step": 47
    },
    {
      "epoch": 0.12,
      "grad_norm": 11.675898551940918,
      "learning_rate": 2.7789473684210525e-06,
      "loss": 0.472,
      "step": 48
    },
    {
      "Batch Mean": -0.25928783416748047,
      "accuracy": 0.75,
      "epoch": 0.12,
      "step": 48
    },
    {
      "epoch": 0.1225,
      "grad_norm": 15.394309997558594,
      "learning_rate": 2.7710526315789473e-06,
      "loss": 0.5011,
      "step": 49
    },
    {
      "Batch Mean": -0.2885439097881317,
      "accuracy": 0.703125,
      "epoch": 0.1225,
      "step": 49
    },
    {
      "epoch": 0.125,
      "grad_norm": 15.69314956665039,
      "learning_rate": 2.763157894736842e-06,
      "loss": 0.5974,
      "step": 50
    },
    {
      "Batch Mean": -0.8389434814453125,
      "accuracy": 0.8046875,
      "epoch": 0.125,
      "step": 50
    },
    {
      "epoch": 0.1275,
      "grad_norm": 16.744606018066406,
      "learning_rate": 2.7552631578947368e-06,
      "loss": 0.5151,
      "step": 51
    },
    {
      "Batch Mean": -0.6685854196548462,
      "accuracy": 0.703125,
      "epoch": 0.1275,
      "step": 51
    },
    {
      "epoch": 0.13,
      "grad_norm": 15.530059814453125,
      "learning_rate": 2.7473684210526315e-06,
      "loss": 0.5502,
      "step": 52
    },
    {
      "Batch Mean": -0.5919044017791748,
      "accuracy": 0.71875,
      "epoch": 0.13,
      "step": 52
    },
    {
      "epoch": 0.1325,
      "grad_norm": 15.14562702178955,
      "learning_rate": 2.7394736842105263e-06,
      "loss": 0.5305,
      "step": 53
    },
    {
      "Batch Mean": -0.11876988410949707,
      "accuracy": 0.75,
      "epoch": 0.1325,
      "step": 53
    },
    {
      "epoch": 0.135,
      "grad_norm": 7.011972427368164,
      "learning_rate": 2.7315789473684214e-06,
      "loss": 0.4802,
      "step": 54
    },
    {
      "Batch Mean": 0.09675300121307373,
      "accuracy": 0.7578125,
      "epoch": 0.135,
      "step": 54
    },
    {
      "epoch": 0.1375,
      "grad_norm": 8.37578010559082,
      "learning_rate": 2.723684210526316e-06,
      "loss": 0.4354,
      "step": 55
    },
    {
      "Batch Mean": 0.3877286911010742,
      "accuracy": 0.8046875,
      "epoch": 0.1375,
      "step": 55
    },
    {
      "epoch": 0.14,
      "grad_norm": 8.380095481872559,
      "learning_rate": 2.715789473684211e-06,
      "loss": 0.4114,
      "step": 56
    },
    {
      "Batch Mean": 0.7472854852676392,
      "accuracy": 0.8046875,
      "epoch": 0.14,
      "step": 56
    },
    {
      "epoch": 0.1425,
      "grad_norm": 12.794017791748047,
      "learning_rate": 2.7078947368421052e-06,
      "loss": 0.465,
      "step": 57
    },
    {
      "Batch Mean": 0.5131087303161621,
      "accuracy": 0.8046875,
      "epoch": 0.1425,
      "step": 57
    },
    {
      "epoch": 0.145,
      "grad_norm": 8.367005348205566,
      "learning_rate": 2.7e-06,
      "loss": 0.4407,
      "step": 58
    },
    {
      "Batch Mean": 0.5741372108459473,
      "accuracy": 0.8515625,
      "epoch": 0.145,
      "step": 58
    },
    {
      "epoch": 0.1475,
      "grad_norm": 9.066951751708984,
      "learning_rate": 2.6921052631578947e-06,
      "loss": 0.4468,
      "step": 59
    },
    {
      "Batch Mean": 0.169053316116333,
      "accuracy": 0.734375,
      "epoch": 0.1475,
      "step": 59
    },
    {
      "epoch": 0.15,
      "grad_norm": 6.888144016265869,
      "learning_rate": 2.6842105263157895e-06,
      "loss": 0.4849,
      "step": 60
    },
    {
      "Batch Mean": -0.18010663986206055,
      "accuracy": 0.765625,
      "epoch": 0.15,
      "step": 60
    },
    {
      "epoch": 0.1525,
      "grad_norm": 6.846613883972168,
      "learning_rate": 2.6763157894736842e-06,
      "loss": 0.4643,
      "step": 61
    },
    {
      "Batch Mean": -0.4085197448730469,
      "accuracy": 0.8359375,
      "epoch": 0.1525,
      "step": 61
    },
    {
      "epoch": 0.155,
      "grad_norm": 7.677695274353027,
      "learning_rate": 2.668421052631579e-06,
      "loss": 0.3926,
      "step": 62
    },
    {
      "Batch Mean": -0.48792409896850586,
      "accuracy": 0.828125,
      "epoch": 0.155,
      "step": 62
    },
    {
      "epoch": 0.1575,
      "grad_norm": 8.656214714050293,
      "learning_rate": 2.6605263157894737e-06,
      "loss": 0.4116,
      "step": 63
    },
    {
      "Batch Mean": -0.19100165367126465,
      "accuracy": 0.7734375,
      "epoch": 0.1575,
      "step": 63
    },
    {
      "epoch": 0.16,
      "grad_norm": 9.927976608276367,
      "learning_rate": 2.6526315789473685e-06,
      "loss": 0.4395,
      "step": 64
    },
    {
      "Batch Mean": 0.025106430053710938,
      "accuracy": 0.8046875,
      "epoch": 0.16,
      "step": 64
    },
    {
      "epoch": 0.1625,
      "grad_norm": 8.97470474243164,
      "learning_rate": 2.644736842105263e-06,
      "loss": 0.4632,
      "step": 65
    },
    {
      "Batch Mean": 0.4497801661491394,
      "accuracy": 0.796875,
      "epoch": 0.1625,
      "step": 65
    },
    {
      "epoch": 0.165,
      "grad_norm": 11.160425186157227,
      "learning_rate": 2.636842105263158e-06,
      "loss": 0.5073,
      "step": 66
    },
    {
      "Batch Mean": 0.6494235992431641,
      "accuracy": 0.8203125,
      "epoch": 0.165,
      "step": 66
    },
    {
      "epoch": 0.1675,
      "grad_norm": 11.464187622070312,
      "learning_rate": 2.6289473684210527e-06,
      "loss": 0.4467,
      "step": 67
    },
    {
      "Batch Mean": -0.17097759246826172,
      "accuracy": 0.828125,
      "epoch": 0.1675,
      "step": 67
    },
    {
      "epoch": 0.17,
      "grad_norm": 8.820326805114746,
      "learning_rate": 2.6210526315789474e-06,
      "loss": 0.4543,
      "step": 68
    },
    {
      "Batch Mean": -0.1554449200630188,
      "accuracy": 0.8046875,
      "epoch": 0.17,
      "step": 68
    },
    {
      "epoch": 0.1725,
      "grad_norm": 9.248348236083984,
      "learning_rate": 2.613157894736842e-06,
      "loss": 0.3796,
      "step": 69
    },
    {
      "Batch Mean": -0.21274375915527344,
      "accuracy": 0.75,
      "epoch": 0.1725,
      "step": 69
    },
    {
      "epoch": 0.175,
      "grad_norm": 8.15241813659668,
      "learning_rate": 2.605263157894737e-06,
      "loss": 0.4445,
      "step": 70
    },
    {
      "Batch Mean": -0.06371593475341797,
      "accuracy": 0.765625,
      "epoch": 0.175,
      "step": 70
    },
    {
      "epoch": 0.1775,
      "grad_norm": 6.543805122375488,
      "learning_rate": 2.5973684210526317e-06,
      "loss": 0.4376,
      "step": 71
    },
    {
      "Batch Mean": -0.10743597149848938,
      "accuracy": 0.7265625,
      "epoch": 0.1775,
      "step": 71
    },
    {
      "epoch": 0.18,
      "grad_norm": 7.594844341278076,
      "learning_rate": 2.5894736842105264e-06,
      "loss": 0.4363,
      "step": 72
    },
    {
      "Batch Mean": 0.08045387268066406,
      "accuracy": 0.84375,
      "epoch": 0.18,
      "step": 72
    },
    {
      "epoch": 0.1825,
      "grad_norm": 7.977687835693359,
      "learning_rate": 2.581578947368421e-06,
      "loss": 0.3616,
      "step": 73
    },
    {
      "Batch Mean": -0.11765336990356445,
      "accuracy": 0.796875,
      "epoch": 0.1825,
      "step": 73
    },
    {
      "epoch": 0.185,
      "grad_norm": 6.521571636199951,
      "learning_rate": 2.573684210526316e-06,
      "loss": 0.4001,
      "step": 74
    },
    {
      "Batch Mean": 0.0607762336730957,
      "accuracy": 0.703125,
      "epoch": 0.185,
      "step": 74
    },
    {
      "epoch": 0.1875,
      "grad_norm": 8.15623950958252,
      "learning_rate": 2.5657894736842107e-06,
      "loss": 0.4941,
      "step": 75
    },
    {
      "Batch Mean": 0.4910193681716919,
      "accuracy": 0.8046875,
      "epoch": 0.1875,
      "step": 75
    },
    {
      "epoch": 0.19,
      "grad_norm": 9.602405548095703,
      "learning_rate": 2.5578947368421054e-06,
      "loss": 0.4224,
      "step": 76
    },
    {
      "Batch Mean": 0.06563162803649902,
      "accuracy": 0.8046875,
      "epoch": 0.19,
      "step": 76
    },
    {
      "epoch": 0.1925,
      "grad_norm": 6.577213764190674,
      "learning_rate": 2.55e-06,
      "loss": 0.4047,
      "step": 77
    },
    {
      "Batch Mean": -0.08942317962646484,
      "accuracy": 0.8046875,
      "epoch": 0.1925,
      "step": 77
    },
    {
      "epoch": 0.195,
      "grad_norm": 16.202590942382812,
      "learning_rate": 2.542105263157895e-06,
      "loss": 0.4418,
      "step": 78
    },
    {
      "Batch Mean": -0.37958288192749023,
      "accuracy": 0.8125,
      "epoch": 0.195,
      "step": 78
    },
    {
      "epoch": 0.1975,
      "grad_norm": 23.45298194885254,
      "learning_rate": 2.5342105263157892e-06,
      "loss": 0.391,
      "step": 79
    },
    {
      "Batch Mean": -0.5664916038513184,
      "accuracy": 0.8125,
      "epoch": 0.1975,
      "step": 79
    },
    {
      "epoch": 0.2,
      "grad_norm": 10.679176330566406,
      "learning_rate": 2.526315789473684e-06,
      "loss": 0.4402,
      "step": 80
    },
    {
      "Batch Mean": -0.31339550018310547,
      "accuracy": 0.8359375,
      "epoch": 0.2,
      "step": 80
    },
    {
      "epoch": 0.2025,
      "grad_norm": 16.459426879882812,
      "learning_rate": 2.5184210526315787e-06,
      "loss": 0.3765,
      "step": 81
    },
    {
      "Batch Mean": 0.20376300811767578,
      "accuracy": 0.78125,
      "epoch": 0.2025,
      "step": 81
    },
    {
      "epoch": 0.205,
      "grad_norm": 8.887246131896973,
      "learning_rate": 2.510526315789474e-06,
      "loss": 0.4509,
      "step": 82
    },
    {
      "Batch Mean": -0.054105013608932495,
      "accuracy": 0.796875,
      "epoch": 0.205,
      "step": 82
    },
    {
      "epoch": 0.2075,
      "grad_norm": 8.116570472717285,
      "learning_rate": 2.5026315789473686e-06,
      "loss": 0.3906,
      "step": 83
    },
    {
      "Batch Mean": 0.4383058547973633,
      "accuracy": 0.875,
      "epoch": 0.2075,
      "step": 83
    },
    {
      "epoch": 0.21,
      "grad_norm": 9.300606727600098,
      "learning_rate": 2.4947368421052634e-06,
      "loss": 0.3099,
      "step": 84
    },
    {
      "Batch Mean": 0.4242062568664551,
      "accuracy": 0.7890625,
      "epoch": 0.21,
      "step": 84
    },
    {
      "epoch": 0.2125,
      "grad_norm": 8.7870512008667,
      "learning_rate": 2.486842105263158e-06,
      "loss": 0.3996,
      "step": 85
    },
    {
      "Batch Mean": 0.22194945812225342,
      "accuracy": 0.78125,
      "epoch": 0.2125,
      "step": 85
    },
    {
      "epoch": 0.215,
      "grad_norm": 8.691889762878418,
      "learning_rate": 2.478947368421053e-06,
      "loss": 0.4512,
      "step": 86
    },
    {
      "Batch Mean": -0.14107918739318848,
      "accuracy": 0.8359375,
      "epoch": 0.215,
      "step": 86
    },
    {
      "epoch": 0.2175,
      "grad_norm": 7.0445146560668945,
      "learning_rate": 2.4710526315789476e-06,
      "loss": 0.3771,
      "step": 87
    },
    {
      "Batch Mean": -0.35906583070755005,
      "accuracy": 0.765625,
      "epoch": 0.2175,
      "step": 87
    },
    {
      "epoch": 0.22,
      "grad_norm": 9.609421730041504,
      "learning_rate": 2.4631578947368424e-06,
      "loss": 0.4558,
      "step": 88
    },
    {
      "Batch Mean": -0.45313501358032227,
      "accuracy": 0.828125,
      "epoch": 0.22,
      "step": 88
    },
    {
      "epoch": 0.2225,
      "grad_norm": 9.091865539550781,
      "learning_rate": 2.4552631578947367e-06,
      "loss": 0.4292,
      "step": 89
    },
    {
      "Batch Mean": -0.35767626762390137,
      "accuracy": 0.8359375,
      "epoch": 0.2225,
      "step": 89
    },
    {
      "epoch": 0.225,
      "grad_norm": 7.811399459838867,
      "learning_rate": 2.4473684210526314e-06,
      "loss": 0.3823,
      "step": 90
    },
    {
      "Batch Mean": 0.0022423267364501953,
      "accuracy": 0.8046875,
      "epoch": 0.225,
      "step": 90
    },
    {
      "epoch": 0.2275,
      "grad_norm": 7.804200172424316,
      "learning_rate": 2.439473684210526e-06,
      "loss": 0.4079,
      "step": 91
    },
    {
      "Batch Mean": 0.17678523063659668,
      "accuracy": 0.8046875,
      "epoch": 0.2275,
      "step": 91
    },
    {
      "epoch": 0.23,
      "grad_norm": 7.715360641479492,
      "learning_rate": 2.431578947368421e-06,
      "loss": 0.391,
      "step": 92
    },
    {
      "Batch Mean": -0.22452354431152344,
      "accuracy": 0.8203125,
      "epoch": 0.23,
      "step": 92
    },
    {
      "epoch": 0.2325,
      "grad_norm": 7.197687149047852,
      "learning_rate": 2.4236842105263157e-06,
      "loss": 0.3864,
      "step": 93
    },
    {
      "Batch Mean": 0.30264759063720703,
      "accuracy": 0.84375,
      "epoch": 0.2325,
      "step": 93
    },
    {
      "epoch": 0.235,
      "grad_norm": 7.754411697387695,
      "learning_rate": 2.4157894736842104e-06,
      "loss": 0.3939,
      "step": 94
    },
    {
      "Batch Mean": -0.2516747713088989,
      "accuracy": 0.7890625,
      "epoch": 0.235,
      "step": 94
    },
    {
      "epoch": 0.2375,
      "grad_norm": 8.55639362335205,
      "learning_rate": 2.4078947368421056e-06,
      "loss": 0.4671,
      "step": 95
    },
    {
      "Batch Mean": -0.15928316116333008,
      "accuracy": 0.8046875,
      "epoch": 0.2375,
      "step": 95
    },
    {
      "epoch": 0.24,
      "grad_norm": 7.463867664337158,
      "learning_rate": 2.4000000000000003e-06,
      "loss": 0.471,
      "step": 96
    },
    {
      "Batch Mean": -0.0638132095336914,
      "accuracy": 0.8125,
      "epoch": 0.24,
      "step": 96
    },
    {
      "epoch": 0.2425,
      "grad_norm": 6.797260761260986,
      "learning_rate": 2.392105263157895e-06,
      "loss": 0.3892,
      "step": 97
    },
    {
      "Batch Mean": -0.08082282543182373,
      "accuracy": 0.8203125,
      "epoch": 0.2425,
      "step": 97
    },
    {
      "epoch": 0.245,
      "grad_norm": 7.096107482910156,
      "learning_rate": 2.38421052631579e-06,
      "loss": 0.3942,
      "step": 98
    },
    {
      "Batch Mean": 0.3185189962387085,
      "accuracy": 0.796875,
      "epoch": 0.245,
      "step": 98
    },
    {
      "epoch": 0.2475,
      "grad_norm": 9.447708129882812,
      "learning_rate": 2.376315789473684e-06,
      "loss": 0.5037,
      "step": 99
    },
    {
      "Batch Mean": 0.26651012897491455,
      "accuracy": 0.890625,
      "epoch": 0.2475,
      "step": 99
    },
    {
      "epoch": 0.25,
      "grad_norm": 9.391190528869629,
      "learning_rate": 2.368421052631579e-06,
      "loss": 0.3352,
      "step": 100
    },
    {
      "Batch Mean": -0.0040132105350494385,
      "accuracy": 0.7578125,
      "epoch": 0.25,
      "step": 100
    },
    {
      "epoch": 0.2525,
      "grad_norm": 7.999260425567627,
      "learning_rate": 2.3605263157894736e-06,
      "loss": 0.4583,
      "step": 101
    },
    {
      "Batch Mean": 0.13753342628479004,
      "accuracy": 0.8125,
      "epoch": 0.2525,
      "step": 101
    },
    {
      "epoch": 0.255,
      "grad_norm": 7.398046493530273,
      "learning_rate": 2.3526315789473684e-06,
      "loss": 0.3847,
      "step": 102
    },
    {
      "Batch Mean": 0.15256333351135254,
      "accuracy": 0.8671875,
      "epoch": 0.255,
      "step": 102
    },
    {
      "epoch": 0.2575,
      "grad_norm": 7.636544227600098,
      "learning_rate": 2.344736842105263e-06,
      "loss": 0.332,
      "step": 103
    },
    {
      "Batch Mean": -0.2140512466430664,
      "accuracy": 0.75,
      "epoch": 0.2575,
      "step": 103
    },
    {
      "epoch": 0.26,
      "grad_norm": 8.101052284240723,
      "learning_rate": 2.336842105263158e-06,
      "loss": 0.4434,
      "step": 104
    },
    {
      "Batch Mean": -0.2993035316467285,
      "accuracy": 0.8359375,
      "epoch": 0.26,
      "step": 104
    },
    {
      "epoch": 0.2625,
      "grad_norm": 7.990828514099121,
      "learning_rate": 2.3289473684210526e-06,
      "loss": 0.3701,
      "step": 105
    },
    {
      "Batch Mean": -0.0038167238235473633,
      "accuracy": 0.796875,
      "epoch": 0.2625,
      "step": 105
    },
    {
      "epoch": 0.265,
      "grad_norm": 6.86246919631958,
      "learning_rate": 2.3210526315789473e-06,
      "loss": 0.3546,
      "step": 106
    },
    {
      "Batch Mean": 0.4233437776565552,
      "accuracy": 0.796875,
      "epoch": 0.265,
      "step": 106
    },
    {
      "epoch": 0.2675,
      "grad_norm": 9.27191162109375,
      "learning_rate": 2.313157894736842e-06,
      "loss": 0.414,
      "step": 107
    },
    {
      "Batch Mean": 0.5853117108345032,
      "accuracy": 0.8046875,
      "epoch": 0.2675,
      "step": 107
    },
    {
      "epoch": 0.27,
      "grad_norm": 12.894753456115723,
      "learning_rate": 2.305263157894737e-06,
      "loss": 0.4258,
      "step": 108
    },
    {
      "Batch Mean": 0.07786870002746582,
      "accuracy": 0.8203125,
      "epoch": 0.27,
      "step": 108
    },
    {
      "epoch": 0.2725,
      "grad_norm": 10.551774024963379,
      "learning_rate": 2.2973684210526316e-06,
      "loss": 0.455,
      "step": 109
    },
    {
      "Batch Mean": 0.003017425537109375,
      "accuracy": 0.8515625,
      "epoch": 0.2725,
      "step": 109
    },
    {
      "epoch": 0.275,
      "grad_norm": 8.059015274047852,
      "learning_rate": 2.2894736842105263e-06,
      "loss": 0.3559,
      "step": 110
    },
    {
      "Batch Mean": 0.08839225769042969,
      "accuracy": 0.859375,
      "epoch": 0.275,
      "step": 110
    },
    {
      "epoch": 0.2775,
      "grad_norm": 7.2062530517578125,
      "learning_rate": 2.281578947368421e-06,
      "loss": 0.2765,
      "step": 111
    },
    {
      "Batch Mean": -0.4452195167541504,
      "accuracy": 0.890625,
      "epoch": 0.2775,
      "step": 111
    },
    {
      "epoch": 0.28,
      "grad_norm": 9.328097343444824,
      "learning_rate": 2.273684210526316e-06,
      "loss": 0.2535,
      "step": 112
    },
    {
      "Batch Mean": -0.2987861633300781,
      "accuracy": 0.828125,
      "epoch": 0.28,
      "step": 112
    },
    {
      "epoch": 0.2825,
      "grad_norm": 10.32376766204834,
      "learning_rate": 2.2657894736842106e-06,
      "loss": 0.4152,
      "step": 113
    },
    {
      "Batch Mean": -0.05189800262451172,
      "accuracy": 0.875,
      "epoch": 0.2825,
      "step": 113
    },
    {
      "epoch": 0.285,
      "grad_norm": 7.734317779541016,
      "learning_rate": 2.2578947368421053e-06,
      "loss": 0.2874,
      "step": 114
    },
    {
      "Batch Mean": 0.8154248595237732,
      "accuracy": 0.7578125,
      "epoch": 0.285,
      "step": 114
    },
    {
      "epoch": 0.2875,
      "grad_norm": 16.42246437072754,
      "learning_rate": 2.25e-06,
      "loss": 0.517,
      "step": 115
    },
    {
      "Batch Mean": 0.2764568328857422,
      "accuracy": 0.8203125,
      "epoch": 0.2875,
      "step": 115
    },
    {
      "epoch": 0.29,
      "grad_norm": 10.174060821533203,
      "learning_rate": 2.242105263157895e-06,
      "loss": 0.3843,
      "step": 116
    },
    {
      "Batch Mean": -0.10035037994384766,
      "accuracy": 0.84375,
      "epoch": 0.29,
      "step": 116
    },
    {
      "epoch": 0.2925,
      "grad_norm": 7.6288161277771,
      "learning_rate": 2.2342105263157895e-06,
      "loss": 0.3509,
      "step": 117
    },
    {
      "Batch Mean": -0.22602462768554688,
      "accuracy": 0.84375,
      "epoch": 0.2925,
      "step": 117
    },
    {
      "epoch": 0.295,
      "grad_norm": 8.613261222839355,
      "learning_rate": 2.2263157894736843e-06,
      "loss": 0.3731,
      "step": 118
    },
    {
      "Batch Mean": -0.09805738925933838,
      "accuracy": 0.7265625,
      "epoch": 0.295,
      "step": 118
    },
    {
      "epoch": 0.2975,
      "grad_norm": 10.688722610473633,
      "learning_rate": 2.218421052631579e-06,
      "loss": 0.4856,
      "step": 119
    },
    {
      "Batch Mean": -0.21628332138061523,
      "accuracy": 0.8125,
      "epoch": 0.2975,
      "step": 119
    },
    {
      "epoch": 0.3,
      "grad_norm": 7.811151027679443,
      "learning_rate": 2.2105263157894738e-06,
      "loss": 0.3613,
      "step": 120
    },
    {
      "Batch Mean": -0.4057881832122803,
      "accuracy": 0.796875,
      "epoch": 0.3,
      "step": 120
    },
    {
      "epoch": 0.3025,
      "grad_norm": 8.88042163848877,
      "learning_rate": 2.2026315789473685e-06,
      "loss": 0.3821,
      "step": 121
    },
    {
      "Batch Mean": -0.3958141803741455,
      "accuracy": 0.84375,
      "epoch": 0.3025,
      "step": 121
    },
    {
      "epoch": 0.305,
      "grad_norm": 8.513867378234863,
      "learning_rate": 2.1947368421052633e-06,
      "loss": 0.3592,
      "step": 122
    },
    {
      "Batch Mean": 0.11790943145751953,
      "accuracy": 0.8671875,
      "epoch": 0.305,
      "step": 122
    },
    {
      "epoch": 0.3075,
      "grad_norm": 6.011488437652588,
      "learning_rate": 2.186842105263158e-06,
      "loss": 0.3249,
      "step": 123
    },
    {
      "Batch Mean": 0.7885537147521973,
      "accuracy": 0.84375,
      "epoch": 0.3075,
      "step": 123
    },
    {
      "epoch": 0.31,
      "grad_norm": 13.359249114990234,
      "learning_rate": 2.1789473684210528e-06,
      "loss": 0.3776,
      "step": 124
    },
    {
      "Batch Mean": 0.906917929649353,
      "accuracy": 0.8125,
      "epoch": 0.31,
      "step": 124
    },
    {
      "epoch": 0.3125,
      "grad_norm": 13.69727897644043,
      "learning_rate": 2.1710526315789475e-06,
      "loss": 0.4525,
      "step": 125
    },
    {
      "Batch Mean": 0.7685656547546387,
      "accuracy": 0.8046875,
      "epoch": 0.3125,
      "step": 125
    },
    {
      "epoch": 0.315,
      "grad_norm": 13.353426933288574,
      "learning_rate": 2.1631578947368423e-06,
      "loss": 0.3873,
      "step": 126
    },
    {
      "Batch Mean": 0.31209874153137207,
      "accuracy": 0.8515625,
      "epoch": 0.315,
      "step": 126
    },
    {
      "epoch": 0.3175,
      "grad_norm": 8.501708030700684,
      "learning_rate": 2.155263157894737e-06,
      "loss": 0.3643,
      "step": 127
    },
    {
      "Batch Mean": -0.0774112343788147,
      "accuracy": 0.8515625,
      "epoch": 0.3175,
      "step": 127
    },
    {
      "epoch": 0.32,
      "grad_norm": 6.79693078994751,
      "learning_rate": 2.1473684210526317e-06,
      "loss": 0.3361,
      "step": 128
    },
    {
      "Batch Mean": -1.1531860828399658,
      "accuracy": 0.875,
      "epoch": 0.32,
      "step": 128
    },
    {
      "epoch": 0.3225,
      "grad_norm": 18.78434181213379,
      "learning_rate": 2.1394736842105265e-06,
      "loss": 0.381,
      "step": 129
    },
    {
      "Batch Mean": -0.9209874868392944,
      "accuracy": 0.828125,
      "epoch": 0.3225,
      "step": 129
    },
    {
      "epoch": 0.325,
      "grad_norm": 15.549153327941895,
      "learning_rate": 2.1315789473684212e-06,
      "loss": 0.352,
      "step": 130
    },
    {
      "Batch Mean": -0.5993038415908813,
      "accuracy": 0.7734375,
      "epoch": 0.325,
      "step": 130
    },
    {
      "epoch": 0.3275,
      "grad_norm": 11.017681121826172,
      "learning_rate": 2.123684210526316e-06,
      "loss": 0.4731,
      "step": 131
    },
    {
      "Batch Mean": -0.5436639785766602,
      "accuracy": 0.859375,
      "epoch": 0.3275,
      "step": 131
    },
    {
      "epoch": 0.33,
      "grad_norm": 9.474353790283203,
      "learning_rate": 2.1157894736842103e-06,
      "loss": 0.3224,
      "step": 132
    },
    {
      "Batch Mean": 0.03529167175292969,
      "accuracy": 0.7890625,
      "epoch": 0.33,
      "step": 132
    },
    {
      "epoch": 0.3325,
      "grad_norm": 9.17381763458252,
      "learning_rate": 2.107894736842105e-06,
      "loss": 0.4091,
      "step": 133
    },
    {
      "Batch Mean": 0.37068402767181396,
      "accuracy": 0.7890625,
      "epoch": 0.3325,
      "step": 133
    },
    {
      "epoch": 0.335,
      "grad_norm": 8.768671035766602,
      "learning_rate": 2.1e-06,
      "loss": 0.4094,
      "step": 134
    },
    {
      "Batch Mean": 0.8418741226196289,
      "accuracy": 0.8203125,
      "epoch": 0.335,
      "step": 134
    },
    {
      "epoch": 0.3375,
      "grad_norm": 13.19222354888916,
      "learning_rate": 2.0921052631578945e-06,
      "loss": 0.4361,
      "step": 135
    },
    {
      "Batch Mean": 0.38561856746673584,
      "accuracy": 0.828125,
      "epoch": 0.3375,
      "step": 135
    },
    {
      "epoch": 0.34,
      "grad_norm": 7.658647537231445,
      "learning_rate": 2.0842105263157897e-06,
      "loss": 0.3619,
      "step": 136
    },
    {
      "Batch Mean": 0.24846503138542175,
      "accuracy": 0.8203125,
      "epoch": 0.34,
      "step": 136
    },
    {
      "epoch": 0.3425,
      "grad_norm": 6.886051654815674,
      "learning_rate": 2.0763157894736845e-06,
      "loss": 0.3778,
      "step": 137
    },
    {
      "Batch Mean": -0.3311891555786133,
      "accuracy": 0.8203125,
      "epoch": 0.3425,
      "step": 137
    },
    {
      "epoch": 0.345,
      "grad_norm": 7.944254398345947,
      "learning_rate": 2.068421052631579e-06,
      "loss": 0.3948,
      "step": 138
    },
    {
      "Batch Mean": -0.568354606628418,
      "accuracy": 0.84375,
      "epoch": 0.345,
      "step": 138
    },
    {
      "epoch": 0.3475,
      "grad_norm": 9.763813018798828,
      "learning_rate": 2.060526315789474e-06,
      "loss": 0.3355,
      "step": 139
    },
    {
      "Batch Mean": -0.39496636390686035,
      "accuracy": 0.8671875,
      "epoch": 0.3475,
      "step": 139
    },
    {
      "epoch": 0.35,
      "grad_norm": 7.516551971435547,
      "learning_rate": 2.0526315789473687e-06,
      "loss": 0.3501,
      "step": 140
    },
    {
      "Batch Mean": -0.15581655502319336,
      "accuracy": 0.828125,
      "epoch": 0.35,
      "step": 140
    },
    {
      "epoch": 0.3525,
      "grad_norm": 8.155888557434082,
      "learning_rate": 2.0447368421052634e-06,
      "loss": 0.3773,
      "step": 141
    },
    {
      "Batch Mean": 0.44981956481933594,
      "accuracy": 0.7578125,
      "epoch": 0.3525,
      "step": 141
    },
    {
      "epoch": 0.355,
      "grad_norm": 9.395730018615723,
      "learning_rate": 2.0368421052631578e-06,
      "loss": 0.4603,
      "step": 142
    },
    {
      "Batch Mean": 0.33932405710220337,
      "accuracy": 0.8125,
      "epoch": 0.355,
      "step": 142
    },
    {
      "epoch": 0.3575,
      "grad_norm": 8.501273155212402,
      "learning_rate": 2.0289473684210525e-06,
      "loss": 0.4167,
      "step": 143
    },
    {
      "Batch Mean": 0.31684207916259766,
      "accuracy": 0.875,
      "epoch": 0.3575,
      "step": 143
    },
    {
      "epoch": 0.36,
      "grad_norm": 7.186644554138184,
      "learning_rate": 2.0210526315789473e-06,
      "loss": 0.3359,
      "step": 144
    },
    {
      "Batch Mean": 0.14882159233093262,
      "accuracy": 0.8359375,
      "epoch": 0.36,
      "step": 144
    },
    {
      "epoch": 0.3625,
      "grad_norm": 7.065666675567627,
      "learning_rate": 2.013157894736842e-06,
      "loss": 0.3773,
      "step": 145
    },
    {
      "Batch Mean": 0.010605573654174805,
      "accuracy": 0.8671875,
      "epoch": 0.3625,
      "step": 145
    },
    {
      "epoch": 0.365,
      "grad_norm": 6.343169212341309,
      "learning_rate": 2.0052631578947367e-06,
      "loss": 0.3441,
      "step": 146
    },
    {
      "Batch Mean": 0.13873529434204102,
      "accuracy": 0.828125,
      "epoch": 0.365,
      "step": 146
    },
    {
      "epoch": 0.3675,
      "grad_norm": 7.586203098297119,
      "learning_rate": 1.9973684210526315e-06,
      "loss": 0.3723,
      "step": 147
    },
    {
      "Batch Mean": -0.6648737192153931,
      "accuracy": 0.7890625,
      "epoch": 0.3675,
      "step": 147
    },
    {
      "epoch": 0.37,
      "grad_norm": 11.366660118103027,
      "learning_rate": 1.9894736842105262e-06,
      "loss": 0.4372,
      "step": 148
    },
    {
      "Batch Mean": -0.3633718490600586,
      "accuracy": 0.8046875,
      "epoch": 0.37,
      "step": 148
    },
    {
      "epoch": 0.3725,
      "grad_norm": 7.831110000610352,
      "learning_rate": 1.9815789473684214e-06,
      "loss": 0.4074,
      "step": 149
    },
    {
      "Batch Mean": -0.12644743919372559,
      "accuracy": 0.8828125,
      "epoch": 0.3725,
      "step": 149
    },
    {
      "epoch": 0.375,
      "grad_norm": 6.3676371574401855,
      "learning_rate": 1.973684210526316e-06,
      "loss": 0.304,
      "step": 150
    },
    {
      "Batch Mean": 0.36059999465942383,
      "accuracy": 0.8125,
      "epoch": 0.375,
      "step": 150
    },
    {
      "epoch": 0.3775,
      "grad_norm": 8.31554889678955,
      "learning_rate": 1.965789473684211e-06,
      "loss": 0.3811,
      "step": 151
    },
    {
      "Batch Mean": 0.3979158401489258,
      "accuracy": 0.8125,
      "epoch": 0.3775,
      "step": 151
    },
    {
      "epoch": 0.38,
      "grad_norm": 9.095989227294922,
      "learning_rate": 1.9578947368421052e-06,
      "loss": 0.4376,
      "step": 152
    },
    {
      "Batch Mean": 0.443440318107605,
      "accuracy": 0.8359375,
      "epoch": 0.38,
      "step": 152
    },
    {
      "epoch": 0.3825,
      "grad_norm": 8.482221603393555,
      "learning_rate": 1.95e-06,
      "loss": 0.3351,
      "step": 153
    },
    {
      "Batch Mean": 0.3431110382080078,
      "accuracy": 0.828125,
      "epoch": 0.3825,
      "step": 153
    },
    {
      "epoch": 0.385,
      "grad_norm": 8.352450370788574,
      "learning_rate": 1.9421052631578947e-06,
      "loss": 0.3627,
      "step": 154
    },
    {
      "Batch Mean": 0.17504891753196716,
      "accuracy": 0.8828125,
      "epoch": 0.385,
      "step": 154
    },
    {
      "epoch": 0.3875,
      "grad_norm": 6.995707035064697,
      "learning_rate": 1.9342105263157895e-06,
      "loss": 0.323,
      "step": 155
    },
    {
      "Batch Mean": -0.1753711700439453,
      "accuracy": 0.8515625,
      "epoch": 0.3875,
      "step": 155
    },
    {
      "epoch": 0.39,
      "grad_norm": 7.297128677368164,
      "learning_rate": 1.926315789473684e-06,
      "loss": 0.3426,
      "step": 156
    },
    {
      "Batch Mean": -0.17636048793792725,
      "accuracy": 0.8515625,
      "epoch": 0.39,
      "step": 156
    },
    {
      "epoch": 0.3925,
      "grad_norm": 6.796414375305176,
      "learning_rate": 1.918421052631579e-06,
      "loss": 0.3263,
      "step": 157
    },
    {
      "Batch Mean": -0.7188338041305542,
      "accuracy": 0.875,
      "epoch": 0.3925,
      "step": 157
    },
    {
      "epoch": 0.395,
      "grad_norm": 10.371746063232422,
      "learning_rate": 1.9105263157894737e-06,
      "loss": 0.3046,
      "step": 158
    },
    {
      "Batch Mean": -0.5050258636474609,
      "accuracy": 0.8671875,
      "epoch": 0.395,
      "step": 158
    },
    {
      "epoch": 0.3975,
      "grad_norm": 8.59885311126709,
      "learning_rate": 1.9026315789473684e-06,
      "loss": 0.3257,
      "step": 159
    },
    {
      "Batch Mean": 0.15757006406784058,
      "accuracy": 0.875,
      "epoch": 0.3975,
      "step": 159
    },
    {
      "epoch": 0.4,
      "grad_norm": 6.61456298828125,
      "learning_rate": 1.8947368421052632e-06,
      "loss": 0.2748,
      "step": 160
    },
    {
      "Batch Mean": 0.4738607406616211,
      "accuracy": 0.8359375,
      "epoch": 0.4,
      "step": 160
    },
    {
      "epoch": 0.4025,
      "grad_norm": 9.277711868286133,
      "learning_rate": 1.8868421052631577e-06,
      "loss": 0.3558,
      "step": 161
    },
    {
      "Batch Mean": 0.9098443984985352,
      "accuracy": 0.8515625,
      "epoch": 0.4025,
      "step": 161
    },
    {
      "epoch": 0.405,
      "grad_norm": 14.112342834472656,
      "learning_rate": 1.8789473684210525e-06,
      "loss": 0.3479,
      "step": 162
    },
    {
      "Batch Mean": -0.09790003299713135,
      "accuracy": 0.8515625,
      "epoch": 0.405,
      "step": 162
    },
    {
      "epoch": 0.4075,
      "grad_norm": 7.988643169403076,
      "learning_rate": 1.8710526315789476e-06,
      "loss": 0.3329,
      "step": 163
    },
    {
      "Batch Mean": -0.3589191436767578,
      "accuracy": 0.875,
      "epoch": 0.4075,
      "step": 163
    },
    {
      "epoch": 0.41,
      "grad_norm": 8.374390602111816,
      "learning_rate": 1.8631578947368424e-06,
      "loss": 0.3404,
      "step": 164
    },
    {
      "Batch Mean": -0.2895526885986328,
      "accuracy": 0.8671875,
      "epoch": 0.41,
      "step": 164
    },
    {
      "epoch": 0.4125,
      "grad_norm": 10.125029563903809,
      "learning_rate": 1.855263157894737e-06,
      "loss": 0.3828,
      "step": 165
    },
    {
      "Batch Mean": -0.7614960670471191,
      "accuracy": 0.8515625,
      "epoch": 0.4125,
      "step": 165
    },
    {
      "epoch": 0.415,
      "grad_norm": 12.611261367797852,
      "learning_rate": 1.8473684210526317e-06,
      "loss": 0.3182,
      "step": 166
    },
    {
      "Batch Mean": 0.047823190689086914,
      "accuracy": 0.84375,
      "epoch": 0.415,
      "step": 166
    },
    {
      "epoch": 0.4175,
      "grad_norm": 7.5092644691467285,
      "learning_rate": 1.8394736842105264e-06,
      "loss": 0.3277,
      "step": 167
    },
    {
      "Batch Mean": 0.3487553596496582,
      "accuracy": 0.8828125,
      "epoch": 0.4175,
      "step": 167
    },
    {
      "epoch": 0.42,
      "grad_norm": 7.762388229370117,
      "learning_rate": 1.8315789473684211e-06,
      "loss": 0.3021,
      "step": 168
    },
    {
      "Batch Mean": -0.34416115283966064,
      "accuracy": 0.7890625,
      "epoch": 0.42,
      "step": 168
    },
    {
      "epoch": 0.4225,
      "grad_norm": 9.338401794433594,
      "learning_rate": 1.8236842105263159e-06,
      "loss": 0.4031,
      "step": 169
    },
    {
      "Batch Mean": 0.013556480407714844,
      "accuracy": 0.8515625,
      "epoch": 0.4225,
      "step": 169
    },
    {
      "epoch": 0.425,
      "grad_norm": 8.208139419555664,
      "learning_rate": 1.8157894736842106e-06,
      "loss": 0.367,
      "step": 170
    },
    {
      "Batch Mean": -0.1292668581008911,
      "accuracy": 0.8828125,
      "epoch": 0.425,
      "step": 170
    },
    {
      "epoch": 0.4275,
      "grad_norm": 6.575423717498779,
      "learning_rate": 1.8078947368421052e-06,
      "loss": 0.2778,
      "step": 171
    },
    {
      "Batch Mean": -0.5196864604949951,
      "accuracy": 0.828125,
      "epoch": 0.4275,
      "step": 171
    },
    {
      "epoch": 0.43,
      "grad_norm": 9.9501953125,
      "learning_rate": 1.8e-06,
      "loss": 0.388,
      "step": 172
    },
    {
      "Batch Mean": -0.10827267169952393,
      "accuracy": 0.8515625,
      "epoch": 0.43,
      "step": 172
    },
    {
      "epoch": 0.4325,
      "grad_norm": 6.746122360229492,
      "learning_rate": 1.7921052631578947e-06,
      "loss": 0.3337,
      "step": 173
    },
    {
      "Batch Mean": -0.20117001235485077,
      "accuracy": 0.859375,
      "epoch": 0.4325,
      "step": 173
    },
    {
      "epoch": 0.435,
      "grad_norm": 7.591915130615234,
      "learning_rate": 1.7842105263157894e-06,
      "loss": 0.3576,
      "step": 174
    },
    {
      "Batch Mean": 0.4478938579559326,
      "accuracy": 0.8125,
      "epoch": 0.435,
      "step": 174
    },
    {
      "epoch": 0.4375,
      "grad_norm": 8.740865707397461,
      "learning_rate": 1.7763157894736842e-06,
      "loss": 0.3836,
      "step": 175
    },
    {
      "Batch Mean": 0.1370692253112793,
      "accuracy": 0.8359375,
      "epoch": 0.4375,
      "step": 175
    },
    {
      "epoch": 0.44,
      "grad_norm": 6.480626583099365,
      "learning_rate": 1.768421052631579e-06,
      "loss": 0.3648,
      "step": 176
    },
    {
      "Batch Mean": 0.47721147537231445,
      "accuracy": 0.859375,
      "epoch": 0.44,
      "step": 176
    },
    {
      "epoch": 0.4425,
      "grad_norm": 9.018311500549316,
      "learning_rate": 1.7605263157894739e-06,
      "loss": 0.4128,
      "step": 177
    },
    {
      "Batch Mean": 0.13465499877929688,
      "accuracy": 0.8671875,
      "epoch": 0.4425,
      "step": 177
    },
    {
      "epoch": 0.445,
      "grad_norm": 6.7764973640441895,
      "learning_rate": 1.7526315789473686e-06,
      "loss": 0.3124,
      "step": 178
    },
    {
      "Batch Mean": 0.25302886962890625,
      "accuracy": 0.8671875,
      "epoch": 0.445,
      "step": 178
    },
    {
      "epoch": 0.4475,
      "grad_norm": 7.894146919250488,
      "learning_rate": 1.7447368421052633e-06,
      "loss": 0.3335,
      "step": 179
    },
    {
      "Batch Mean": -0.5794901847839355,
      "accuracy": 0.7734375,
      "epoch": 0.4475,
      "step": 179
    },
    {
      "epoch": 0.45,
      "grad_norm": 8.826544761657715,
      "learning_rate": 1.736842105263158e-06,
      "loss": 0.3982,
      "step": 180
    },
    {
      "Batch Mean": 0.16424822807312012,
      "accuracy": 0.84375,
      "epoch": 0.45,
      "step": 180
    },
    {
      "epoch": 0.4525,
      "grad_norm": 5.991344451904297,
      "learning_rate": 1.7289473684210526e-06,
      "loss": 0.3408,
      "step": 181
    },
    {
      "Batch Mean": -0.13757801055908203,
      "accuracy": 0.8671875,
      "epoch": 0.4525,
      "step": 181
    },
    {
      "epoch": 0.455,
      "grad_norm": 5.649535179138184,
      "learning_rate": 1.7210526315789474e-06,
      "loss": 0.3081,
      "step": 182
    },
    {
      "Batch Mean": 0.13538789749145508,
      "accuracy": 0.78125,
      "epoch": 0.455,
      "step": 182
    },
    {
      "epoch": 0.4575,
      "grad_norm": 7.989570140838623,
      "learning_rate": 1.7131578947368421e-06,
      "loss": 0.4532,
      "step": 183
    },
    {
      "Batch Mean": -0.09452962875366211,
      "accuracy": 0.796875,
      "epoch": 0.4575,
      "step": 183
    },
    {
      "epoch": 0.46,
      "grad_norm": 6.7350616455078125,
      "learning_rate": 1.7052631578947369e-06,
      "loss": 0.3937,
      "step": 184
    },
    {
      "Batch Mean": -0.19620466232299805,
      "accuracy": 0.8046875,
      "epoch": 0.46,
      "step": 184
    },
    {
      "epoch": 0.4625,
      "grad_norm": 32.979637145996094,
      "learning_rate": 1.6973684210526316e-06,
      "loss": 0.4651,
      "step": 185
    },
    {
      "Batch Mean": 0.203094482421875,
      "accuracy": 0.8203125,
      "epoch": 0.4625,
      "step": 185
    },
    {
      "epoch": 0.465,
      "grad_norm": 6.455933570861816,
      "learning_rate": 1.6894736842105264e-06,
      "loss": 0.3405,
      "step": 186
    },
    {
      "Batch Mean": 0.39686131477355957,
      "accuracy": 0.875,
      "epoch": 0.465,
      "step": 186
    },
    {
      "epoch": 0.4675,
      "grad_norm": 6.76831579208374,
      "learning_rate": 1.6815789473684209e-06,
      "loss": 0.3486,
      "step": 187
    },
    {
      "Batch Mean": -0.008591651916503906,
      "accuracy": 0.859375,
      "epoch": 0.4675,
      "step": 187
    },
    {
      "epoch": 0.47,
      "grad_norm": 5.692453861236572,
      "learning_rate": 1.6736842105263156e-06,
      "loss": 0.3288,
      "step": 188
    },
    {
      "Batch Mean": -0.36408185958862305,
      "accuracy": 0.8046875,
      "epoch": 0.47,
      "step": 188
    },
    {
      "epoch": 0.4725,
      "grad_norm": 7.711721897125244,
      "learning_rate": 1.6657894736842104e-06,
      "loss": 0.4075,
      "step": 189
    },
    {
      "Batch Mean": -0.4406242370605469,
      "accuracy": 0.890625,
      "epoch": 0.4725,
      "step": 189
    },
    {
      "epoch": 0.475,
      "grad_norm": 7.3570756912231445,
      "learning_rate": 1.6578947368421056e-06,
      "loss": 0.3025,
      "step": 190
    },
    {
      "Batch Mean": 0.1726367473602295,
      "accuracy": 0.8125,
      "epoch": 0.475,
      "step": 190
    },
    {
      "epoch": 0.4775,
      "grad_norm": 7.980853080749512,
      "learning_rate": 1.65e-06,
      "loss": 0.3876,
      "step": 191
    },
    {
      "Batch Mean": -0.0961066484451294,
      "accuracy": 0.90625,
      "epoch": 0.4775,
      "step": 191
    },
    {
      "epoch": 0.48,
      "grad_norm": 5.444026947021484,
      "learning_rate": 1.6421052631578948e-06,
      "loss": 0.2777,
      "step": 192
    },
    {
      "Batch Mean": -0.12086796760559082,
      "accuracy": 0.8515625,
      "epoch": 0.48,
      "step": 192
    },
    {
      "epoch": 0.4825,
      "grad_norm": 5.993566989898682,
      "learning_rate": 1.6342105263157896e-06,
      "loss": 0.3327,
      "step": 193
    },
    {
      "Batch Mean": 0.5805015563964844,
      "accuracy": 0.84375,
      "epoch": 0.4825,
      "step": 193
    },
    {
      "epoch": 0.485,
      "grad_norm": 9.61883544921875,
      "learning_rate": 1.6263157894736843e-06,
      "loss": 0.3707,
      "step": 194
    },
    {
      "Batch Mean": 0.8935675024986267,
      "accuracy": 0.8046875,
      "epoch": 0.485,
      "step": 194
    },
    {
      "epoch": 0.4875,
      "grad_norm": 12.503327369689941,
      "learning_rate": 1.618421052631579e-06,
      "loss": 0.3951,
      "step": 195
    },
    {
      "Batch Mean": 0.2793365716934204,
      "accuracy": 0.8046875,
      "epoch": 0.4875,
      "step": 195
    },
    {
      "epoch": 0.49,
      "grad_norm": 7.763684272766113,
      "learning_rate": 1.6105263157894738e-06,
      "loss": 0.3955,
      "step": 196
    },
    {
      "Batch Mean": -0.14763522148132324,
      "accuracy": 0.8359375,
      "epoch": 0.49,
      "step": 196
    },
    {
      "epoch": 0.4925,
      "grad_norm": 7.21512508392334,
      "learning_rate": 1.6026315789473683e-06,
      "loss": 0.3188,
      "step": 197
    },
    {
      "Batch Mean": -0.5799511671066284,
      "accuracy": 0.8671875,
      "epoch": 0.4925,
      "step": 197
    },
    {
      "epoch": 0.495,
      "grad_norm": 8.71732234954834,
      "learning_rate": 1.594736842105263e-06,
      "loss": 0.3139,
      "step": 198
    },
    {
      "Batch Mean": -0.46822547912597656,
      "accuracy": 0.890625,
      "epoch": 0.495,
      "step": 198
    },
    {
      "epoch": 0.4975,
      "grad_norm": 7.613549709320068,
      "learning_rate": 1.5868421052631578e-06,
      "loss": 0.2978,
      "step": 199
    },
    {
      "Batch Mean": -0.370266854763031,
      "accuracy": 0.875,
      "epoch": 0.4975,
      "step": 199
    },
    {
      "epoch": 0.5,
      "grad_norm": 7.669687271118164,
      "learning_rate": 1.5789473684210526e-06,
      "loss": 0.2982,
      "step": 200
    }
  ],
  "logging_steps": 1,
  "max_steps": 400,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 100,
  "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
}