{
  "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": 2.61431884765625,
      "accuracy": 0.4765625,
      "epoch": 0,
      "step": 0
    },
    {
      "epoch": 0.0025,
      "grad_norm": 42.59496307373047,
      "learning_rate": 1.5000000000000002e-07,
      "loss": 0.8092,
      "step": 1
    },
    {
      "Batch Mean": 2.574005126953125,
      "accuracy": 0.578125,
      "epoch": 0.0025,
      "step": 1
    },
    {
      "epoch": 0.005,
      "grad_norm": 46.709930419921875,
      "learning_rate": 3.0000000000000004e-07,
      "loss": 0.7799,
      "step": 2
    },
    {
      "Batch Mean": 2.560516357421875,
      "accuracy": 0.484375,
      "epoch": 0.005,
      "step": 2
    },
    {
      "epoch": 0.0075,
      "grad_norm": 44.51314163208008,
      "learning_rate": 4.5e-07,
      "loss": 0.8043,
      "step": 3
    },
    {
      "Batch Mean": 2.6197509765625,
      "accuracy": 0.5,
      "epoch": 0.0075,
      "step": 3
    },
    {
      "epoch": 0.01,
      "grad_norm": 42.86121368408203,
      "learning_rate": 6.000000000000001e-07,
      "loss": 0.8014,
      "step": 4
    },
    {
      "Batch Mean": 2.565338134765625,
      "accuracy": 0.53125,
      "epoch": 0.01,
      "step": 4
    },
    {
      "epoch": 0.0125,
      "grad_norm": 44.409908294677734,
      "learning_rate": 7.5e-07,
      "loss": 0.7983,
      "step": 5
    },
    {
      "Batch Mean": 2.522857666015625,
      "accuracy": 0.484375,
      "epoch": 0.0125,
      "step": 5
    },
    {
      "epoch": 0.015,
      "grad_norm": 43.77534484863281,
      "learning_rate": 9e-07,
      "loss": 0.7975,
      "step": 6
    },
    {
      "Batch Mean": 2.3875732421875,
      "accuracy": 0.4921875,
      "epoch": 0.015,
      "step": 6
    },
    {
      "epoch": 0.0175,
      "grad_norm": 41.981407165527344,
      "learning_rate": 1.05e-06,
      "loss": 0.7951,
      "step": 7
    },
    {
      "Batch Mean": 2.317047119140625,
      "accuracy": 0.515625,
      "epoch": 0.0175,
      "step": 7
    },
    {
      "epoch": 0.02,
      "grad_norm": 40.53751754760742,
      "learning_rate": 1.2000000000000002e-06,
      "loss": 0.791,
      "step": 8
    },
    {
      "Batch Mean": 1.757080078125,
      "accuracy": 0.5078125,
      "epoch": 0.02,
      "step": 8
    },
    {
      "epoch": 0.0225,
      "grad_norm": 31.105939865112305,
      "learning_rate": 1.35e-06,
      "loss": 0.7571,
      "step": 9
    },
    {
      "Batch Mean": 1.6220855712890625,
      "accuracy": 0.578125,
      "epoch": 0.0225,
      "step": 9
    },
    {
      "epoch": 0.025,
      "grad_norm": 33.61161804199219,
      "learning_rate": 1.5e-06,
      "loss": 0.7234,
      "step": 10
    },
    {
      "Batch Mean": 1.400146484375,
      "accuracy": 0.53125,
      "epoch": 0.025,
      "step": 10
    },
    {
      "epoch": 0.0275,
      "grad_norm": 34.554622650146484,
      "learning_rate": 1.65e-06,
      "loss": 0.725,
      "step": 11
    },
    {
      "Batch Mean": -0.43697381019592285,
      "accuracy": 0.5234375,
      "epoch": 0.0275,
      "step": 11
    },
    {
      "epoch": 0.03,
      "grad_norm": 12.222326278686523,
      "learning_rate": 1.8e-06,
      "loss": 0.6986,
      "step": 12
    },
    {
      "Batch Mean": -0.8854951858520508,
      "accuracy": 0.5234375,
      "epoch": 0.03,
      "step": 12
    },
    {
      "epoch": 0.0325,
      "grad_norm": 16.398786544799805,
      "learning_rate": 1.95e-06,
      "loss": 0.7081,
      "step": 13
    },
    {
      "Batch Mean": -0.9508838653564453,
      "accuracy": 0.640625,
      "epoch": 0.0325,
      "step": 13
    },
    {
      "epoch": 0.035,
      "grad_norm": 19.590002059936523,
      "learning_rate": 2.1e-06,
      "loss": 0.6784,
      "step": 14
    },
    {
      "Batch Mean": -1.0560526847839355,
      "accuracy": 0.4921875,
      "epoch": 0.035,
      "step": 14
    },
    {
      "epoch": 0.0375,
      "grad_norm": 22.669742584228516,
      "learning_rate": 2.25e-06,
      "loss": 0.7235,
      "step": 15
    },
    {
      "Batch Mean": -1.1405725479125977,
      "accuracy": 0.625,
      "epoch": 0.0375,
      "step": 15
    },
    {
      "epoch": 0.04,
      "grad_norm": 22.189626693725586,
      "learning_rate": 2.4000000000000003e-06,
      "loss": 0.6714,
      "step": 16
    },
    {
      "Batch Mean": -1.1522831916809082,
      "accuracy": 0.640625,
      "epoch": 0.04,
      "step": 16
    },
    {
      "epoch": 0.0425,
      "grad_norm": 19.9060115814209,
      "learning_rate": 2.55e-06,
      "loss": 0.6757,
      "step": 17
    },
    {
      "Batch Mean": -1.1666946411132812,
      "accuracy": 0.53125,
      "epoch": 0.0425,
      "step": 17
    },
    {
      "epoch": 0.045,
      "grad_norm": 23.661216735839844,
      "learning_rate": 2.7e-06,
      "loss": 0.7002,
      "step": 18
    },
    {
      "Batch Mean": -0.999359130859375,
      "accuracy": 0.6640625,
      "epoch": 0.045,
      "step": 18
    },
    {
      "epoch": 0.0475,
      "grad_norm": 15.433859825134277,
      "learning_rate": 2.85e-06,
      "loss": 0.6471,
      "step": 19
    },
    {
      "Batch Mean": -0.6625549793243408,
      "accuracy": 0.6171875,
      "epoch": 0.0475,
      "step": 19
    },
    {
      "epoch": 0.05,
      "grad_norm": 13.135390281677246,
      "learning_rate": 3e-06,
      "loss": 0.662,
      "step": 20
    },
    {
      "Batch Mean": -0.13487493991851807,
      "accuracy": 0.6171875,
      "epoch": 0.05,
      "step": 20
    },
    {
      "epoch": 0.0525,
      "grad_norm": 7.336737632751465,
      "learning_rate": 2.992105263157895e-06,
      "loss": 0.629,
      "step": 21
    },
    {
      "Batch Mean": 0.2938040494918823,
      "accuracy": 0.6640625,
      "epoch": 0.0525,
      "step": 21
    },
    {
      "epoch": 0.055,
      "grad_norm": 9.586155891418457,
      "learning_rate": 2.9842105263157896e-06,
      "loss": 0.6225,
      "step": 22
    },
    {
      "Batch Mean": 0.6780391931533813,
      "accuracy": 0.7421875,
      "epoch": 0.055,
      "step": 22
    },
    {
      "epoch": 0.0575,
      "grad_norm": 14.825817108154297,
      "learning_rate": 2.9763157894736843e-06,
      "loss": 0.607,
      "step": 23
    },
    {
      "Batch Mean": 0.9421095848083496,
      "accuracy": 0.609375,
      "epoch": 0.0575,
      "step": 23
    },
    {
      "epoch": 0.06,
      "grad_norm": 16.7529354095459,
      "learning_rate": 2.968421052631579e-06,
      "loss": 0.6542,
      "step": 24
    },
    {
      "Batch Mean": 1.0722179412841797,
      "accuracy": 0.71875,
      "epoch": 0.06,
      "step": 24
    },
    {
      "epoch": 0.0625,
      "grad_norm": 18.940032958984375,
      "learning_rate": 2.960526315789474e-06,
      "loss": 0.5882,
      "step": 25
    },
    {
      "Batch Mean": 0.8396664261817932,
      "accuracy": 0.6484375,
      "epoch": 0.0625,
      "step": 25
    },
    {
      "epoch": 0.065,
      "grad_norm": 14.680763244628906,
      "learning_rate": 2.9526315789473685e-06,
      "loss": 0.6463,
      "step": 26
    },
    {
      "Batch Mean": 0.2986793518066406,
      "accuracy": 0.671875,
      "epoch": 0.065,
      "step": 26
    },
    {
      "epoch": 0.0675,
      "grad_norm": 7.3362345695495605,
      "learning_rate": 2.9447368421052633e-06,
      "loss": 0.5923,
      "step": 27
    },
    {
      "Batch Mean": -0.29196763038635254,
      "accuracy": 0.734375,
      "epoch": 0.0675,
      "step": 27
    },
    {
      "epoch": 0.07,
      "grad_norm": 8.053507804870605,
      "learning_rate": 2.936842105263158e-06,
      "loss": 0.5261,
      "step": 28
    },
    {
      "Batch Mean": -0.8161113262176514,
      "accuracy": 0.7109375,
      "epoch": 0.07,
      "step": 28
    },
    {
      "epoch": 0.0725,
      "grad_norm": 14.036270141601562,
      "learning_rate": 2.9289473684210528e-06,
      "loss": 0.5482,
      "step": 29
    },
    {
      "Batch Mean": -0.7824737429618835,
      "accuracy": 0.75,
      "epoch": 0.0725,
      "step": 29
    },
    {
      "epoch": 0.075,
      "grad_norm": 13.984892845153809,
      "learning_rate": 2.9210526315789475e-06,
      "loss": 0.5939,
      "step": 30
    },
    {
      "Batch Mean": -0.4927825927734375,
      "accuracy": 0.734375,
      "epoch": 0.075,
      "step": 30
    },
    {
      "epoch": 0.0775,
      "grad_norm": 10.751565933227539,
      "learning_rate": 2.9131578947368423e-06,
      "loss": 0.5901,
      "step": 31
    },
    {
      "Batch Mean": 0.15497040748596191,
      "accuracy": 0.6796875,
      "epoch": 0.0775,
      "step": 31
    },
    {
      "epoch": 0.08,
      "grad_norm": 8.30102252960205,
      "learning_rate": 2.905263157894737e-06,
      "loss": 0.5731,
      "step": 32
    },
    {
      "Batch Mean": 0.16421844065189362,
      "accuracy": 0.7578125,
      "epoch": 0.08,
      "step": 32
    },
    {
      "epoch": 0.0825,
      "grad_norm": 9.087366104125977,
      "learning_rate": 2.8973684210526318e-06,
      "loss": 0.4958,
      "step": 33
    },
    {
      "Batch Mean": 0.3596491813659668,
      "accuracy": 0.859375,
      "epoch": 0.0825,
      "step": 33
    },
    {
      "epoch": 0.085,
      "grad_norm": 9.984371185302734,
      "learning_rate": 2.8894736842105265e-06,
      "loss": 0.4321,
      "step": 34
    },
    {
      "Batch Mean": 0.2128266543149948,
      "accuracy": 0.734375,
      "epoch": 0.085,
      "step": 34
    },
    {
      "epoch": 0.0875,
      "grad_norm": 7.400077819824219,
      "learning_rate": 2.8815789473684213e-06,
      "loss": 0.5178,
      "step": 35
    },
    {
      "Batch Mean": 0.10449030995368958,
      "accuracy": 0.7578125,
      "epoch": 0.0875,
      "step": 35
    },
    {
      "epoch": 0.09,
      "grad_norm": 6.669457912445068,
      "learning_rate": 2.873684210526316e-06,
      "loss": 0.4914,
      "step": 36
    },
    {
      "Batch Mean": -0.24382781982421875,
      "accuracy": 0.6953125,
      "epoch": 0.09,
      "step": 36
    },
    {
      "epoch": 0.0925,
      "grad_norm": 10.831747055053711,
      "learning_rate": 2.8657894736842103e-06,
      "loss": 0.5305,
      "step": 37
    },
    {
      "Batch Mean": -0.29154396057128906,
      "accuracy": 0.7109375,
      "epoch": 0.0925,
      "step": 37
    },
    {
      "epoch": 0.095,
      "grad_norm": 11.060967445373535,
      "learning_rate": 2.857894736842105e-06,
      "loss": 0.5589,
      "step": 38
    },
    {
      "Batch Mean": -0.07830595970153809,
      "accuracy": 0.765625,
      "epoch": 0.095,
      "step": 38
    },
    {
      "epoch": 0.0975,
      "grad_norm": 8.426444053649902,
      "learning_rate": 2.85e-06,
      "loss": 0.4663,
      "step": 39
    },
    {
      "Batch Mean": -0.09856069087982178,
      "accuracy": 0.7890625,
      "epoch": 0.0975,
      "step": 39
    },
    {
      "epoch": 0.1,
      "grad_norm": 11.046717643737793,
      "learning_rate": 2.8421052631578946e-06,
      "loss": 0.455,
      "step": 40
    },
    {
      "Batch Mean": 0.29904642701148987,
      "accuracy": 0.8125,
      "epoch": 0.1,
      "step": 40
    },
    {
      "epoch": 0.1025,
      "grad_norm": 11.232060432434082,
      "learning_rate": 2.8342105263157897e-06,
      "loss": 0.437,
      "step": 41
    },
    {
      "Batch Mean": 0.43527090549468994,
      "accuracy": 0.78125,
      "epoch": 0.1025,
      "step": 41
    },
    {
      "epoch": 0.105,
      "grad_norm": 12.555906295776367,
      "learning_rate": 2.8263157894736845e-06,
      "loss": 0.5316,
      "step": 42
    },
    {
      "Batch Mean": 0.02639901638031006,
      "accuracy": 0.8046875,
      "epoch": 0.105,
      "step": 42
    },
    {
      "epoch": 0.1075,
      "grad_norm": 9.500948905944824,
      "learning_rate": 2.8184210526315792e-06,
      "loss": 0.4874,
      "step": 43
    },
    {
      "Batch Mean": -0.3524761199951172,
      "accuracy": 0.7421875,
      "epoch": 0.1075,
      "step": 43
    },
    {
      "epoch": 0.11,
      "grad_norm": 10.417457580566406,
      "learning_rate": 2.810526315789474e-06,
      "loss": 0.5011,
      "step": 44
    },
    {
      "Batch Mean": -0.32391130924224854,
      "accuracy": 0.8125,
      "epoch": 0.11,
      "step": 44
    },
    {
      "epoch": 0.1125,
      "grad_norm": 9.867128372192383,
      "learning_rate": 2.8026315789473687e-06,
      "loss": 0.4908,
      "step": 45
    },
    {
      "Batch Mean": -0.42060422897338867,
      "accuracy": 0.8125,
      "epoch": 0.1125,
      "step": 45
    },
    {
      "epoch": 0.115,
      "grad_norm": 10.613035202026367,
      "learning_rate": 2.7947368421052635e-06,
      "loss": 0.4642,
      "step": 46
    },
    {
      "Batch Mean": 0.04145359992980957,
      "accuracy": 0.7890625,
      "epoch": 0.115,
      "step": 46
    },
    {
      "epoch": 0.1175,
      "grad_norm": 12.13752555847168,
      "learning_rate": 2.7868421052631578e-06,
      "loss": 0.4943,
      "step": 47
    },
    {
      "Batch Mean": 0.24553179740905762,
      "accuracy": 0.8046875,
      "epoch": 0.1175,
      "step": 47
    },
    {
      "epoch": 0.12,
      "grad_norm": 11.048128128051758,
      "learning_rate": 2.7789473684210525e-06,
      "loss": 0.3946,
      "step": 48
    },
    {
      "Batch Mean": -0.05052506923675537,
      "accuracy": 0.796875,
      "epoch": 0.12,
      "step": 48
    },
    {
      "epoch": 0.1225,
      "grad_norm": 9.053565979003906,
      "learning_rate": 2.7710526315789473e-06,
      "loss": 0.4474,
      "step": 49
    },
    {
      "Batch Mean": -0.43336963653564453,
      "accuracy": 0.7265625,
      "epoch": 0.1225,
      "step": 49
    },
    {
      "epoch": 0.125,
      "grad_norm": 15.04668140411377,
      "learning_rate": 2.763157894736842e-06,
      "loss": 0.5496,
      "step": 50
    },
    {
      "Batch Mean": -0.29952335357666016,
      "accuracy": 0.765625,
      "epoch": 0.125,
      "step": 50
    },
    {
      "epoch": 0.1275,
      "grad_norm": 11.302435874938965,
      "learning_rate": 2.7552631578947368e-06,
      "loss": 0.4337,
      "step": 51
    },
    {
      "Batch Mean": -0.01873302459716797,
      "accuracy": 0.828125,
      "epoch": 0.1275,
      "step": 51
    },
    {
      "epoch": 0.13,
      "grad_norm": 10.582301139831543,
      "learning_rate": 2.7473684210526315e-06,
      "loss": 0.4073,
      "step": 52
    },
    {
      "Batch Mean": 0.4117751717567444,
      "accuracy": 0.8046875,
      "epoch": 0.13,
      "step": 52
    },
    {
      "epoch": 0.1325,
      "grad_norm": 9.776220321655273,
      "learning_rate": 2.7394736842105263e-06,
      "loss": 0.4235,
      "step": 53
    },
    {
      "Batch Mean": 0.5990171432495117,
      "accuracy": 0.78125,
      "epoch": 0.1325,
      "step": 53
    },
    {
      "epoch": 0.135,
      "grad_norm": 12.647154808044434,
      "learning_rate": 2.7315789473684214e-06,
      "loss": 0.4858,
      "step": 54
    },
    {
      "Batch Mean": 0.2538492679595947,
      "accuracy": 0.7265625,
      "epoch": 0.135,
      "step": 54
    },
    {
      "epoch": 0.1375,
      "grad_norm": 11.564103126525879,
      "learning_rate": 2.723684210526316e-06,
      "loss": 0.5233,
      "step": 55
    },
    {
      "Batch Mean": 0.17059040069580078,
      "accuracy": 0.7578125,
      "epoch": 0.1375,
      "step": 55
    },
    {
      "epoch": 0.14,
      "grad_norm": 10.589515686035156,
      "learning_rate": 2.715789473684211e-06,
      "loss": 0.4762,
      "step": 56
    },
    {
      "Batch Mean": -0.3642357587814331,
      "accuracy": 0.78125,
      "epoch": 0.14,
      "step": 56
    },
    {
      "epoch": 0.1425,
      "grad_norm": 10.727502822875977,
      "learning_rate": 2.7078947368421052e-06,
      "loss": 0.5289,
      "step": 57
    },
    {
      "Batch Mean": -0.5980481505393982,
      "accuracy": 0.8203125,
      "epoch": 0.1425,
      "step": 57
    },
    {
      "epoch": 0.145,
      "grad_norm": 8.915611267089844,
      "learning_rate": 2.7e-06,
      "loss": 0.4038,
      "step": 58
    },
    {
      "Batch Mean": -0.46013855934143066,
      "accuracy": 0.859375,
      "epoch": 0.145,
      "step": 58
    },
    {
      "epoch": 0.1475,
      "grad_norm": 9.178686141967773,
      "learning_rate": 2.6921052631578947e-06,
      "loss": 0.3708,
      "step": 59
    },
    {
      "Batch Mean": -0.1478586494922638,
      "accuracy": 0.84375,
      "epoch": 0.1475,
      "step": 59
    },
    {
      "epoch": 0.15,
      "grad_norm": 6.339688301086426,
      "learning_rate": 2.6842105263157895e-06,
      "loss": 0.4036,
      "step": 60
    },
    {
      "Batch Mean": 0.003389716148376465,
      "accuracy": 0.7578125,
      "epoch": 0.15,
      "step": 60
    },
    {
      "epoch": 0.1525,
      "grad_norm": 6.763300895690918,
      "learning_rate": 2.6763157894736842e-06,
      "loss": 0.5163,
      "step": 61
    },
    {
      "Batch Mean": 0.4018087387084961,
      "accuracy": 0.7734375,
      "epoch": 0.1525,
      "step": 61
    },
    {
      "epoch": 0.155,
      "grad_norm": 8.631525039672852,
      "learning_rate": 2.668421052631579e-06,
      "loss": 0.4219,
      "step": 62
    },
    {
      "Batch Mean": 0.342004656791687,
      "accuracy": 0.734375,
      "epoch": 0.155,
      "step": 62
    },
    {
      "epoch": 0.1575,
      "grad_norm": 9.190362930297852,
      "learning_rate": 2.6605263157894737e-06,
      "loss": 0.4708,
      "step": 63
    },
    {
      "Batch Mean": 0.5178697109222412,
      "accuracy": 0.8125,
      "epoch": 0.1575,
      "step": 63
    },
    {
      "epoch": 0.16,
      "grad_norm": 10.207452774047852,
      "learning_rate": 2.6526315789473685e-06,
      "loss": 0.4458,
      "step": 64
    },
    {
      "Batch Mean": 0.03012150526046753,
      "accuracy": 0.8125,
      "epoch": 0.16,
      "step": 64
    },
    {
      "epoch": 0.1625,
      "grad_norm": 8.050507545471191,
      "learning_rate": 2.644736842105263e-06,
      "loss": 0.4344,
      "step": 65
    },
    {
      "Batch Mean": -0.3691895604133606,
      "accuracy": 0.78125,
      "epoch": 0.1625,
      "step": 65
    },
    {
      "epoch": 0.165,
      "grad_norm": 11.163555145263672,
      "learning_rate": 2.636842105263158e-06,
      "loss": 0.4276,
      "step": 66
    },
    {
      "Batch Mean": -0.2624788284301758,
      "accuracy": 0.8125,
      "epoch": 0.165,
      "step": 66
    },
    {
      "epoch": 0.1675,
      "grad_norm": 9.523059844970703,
      "learning_rate": 2.6289473684210527e-06,
      "loss": 0.4001,
      "step": 67
    },
    {
      "Batch Mean": -0.0055138468742370605,
      "accuracy": 0.8046875,
      "epoch": 0.1675,
      "step": 67
    },
    {
      "epoch": 0.17,
      "grad_norm": 9.094910621643066,
      "learning_rate": 2.6210526315789474e-06,
      "loss": 0.4024,
      "step": 68
    },
    {
      "Batch Mean": 0.2572704553604126,
      "accuracy": 0.7734375,
      "epoch": 0.17,
      "step": 68
    },
    {
      "epoch": 0.1725,
      "grad_norm": 10.00645923614502,
      "learning_rate": 2.613157894736842e-06,
      "loss": 0.4802,
      "step": 69
    },
    {
      "Batch Mean": 0.5112218856811523,
      "accuracy": 0.8046875,
      "epoch": 0.1725,
      "step": 69
    },
    {
      "epoch": 0.175,
      "grad_norm": 11.776649475097656,
      "learning_rate": 2.605263157894737e-06,
      "loss": 0.4194,
      "step": 70
    },
    {
      "Batch Mean": 0.29661768674850464,
      "accuracy": 0.8125,
      "epoch": 0.175,
      "step": 70
    },
    {
      "epoch": 0.1775,
      "grad_norm": 9.701162338256836,
      "learning_rate": 2.5973684210526317e-06,
      "loss": 0.4356,
      "step": 71
    },
    {
      "Batch Mean": 0.049837589263916016,
      "accuracy": 0.859375,
      "epoch": 0.1775,
      "step": 71
    },
    {
      "epoch": 0.18,
      "grad_norm": 8.692646980285645,
      "learning_rate": 2.5894736842105264e-06,
      "loss": 0.2881,
      "step": 72
    },
    {
      "Batch Mean": -0.3786022663116455,
      "accuracy": 0.8125,
      "epoch": 0.18,
      "step": 72
    },
    {
      "epoch": 0.1825,
      "grad_norm": 10.834145545959473,
      "learning_rate": 2.581578947368421e-06,
      "loss": 0.4654,
      "step": 73
    },
    {
      "Batch Mean": -0.20977401733398438,
      "accuracy": 0.8515625,
      "epoch": 0.1825,
      "step": 73
    },
    {
      "epoch": 0.185,
      "grad_norm": 7.816598892211914,
      "learning_rate": 2.573684210526316e-06,
      "loss": 0.4101,
      "step": 74
    },
    {
      "Batch Mean": -0.6599991321563721,
      "accuracy": 0.8203125,
      "epoch": 0.185,
      "step": 74
    },
    {
      "epoch": 0.1875,
      "grad_norm": 13.089007377624512,
      "learning_rate": 2.5657894736842107e-06,
      "loss": 0.427,
      "step": 75
    },
    {
      "Batch Mean": -0.37617337703704834,
      "accuracy": 0.7734375,
      "epoch": 0.1875,
      "step": 75
    },
    {
      "epoch": 0.19,
      "grad_norm": 12.075210571289062,
      "learning_rate": 2.5578947368421054e-06,
      "loss": 0.4797,
      "step": 76
    },
    {
      "Batch Mean": 0.1281442642211914,
      "accuracy": 0.8046875,
      "epoch": 0.19,
      "step": 76
    },
    {
      "epoch": 0.1925,
      "grad_norm": 9.70462703704834,
      "learning_rate": 2.55e-06,
      "loss": 0.4425,
      "step": 77
    },
    {
      "Batch Mean": 0.6071650981903076,
      "accuracy": 0.828125,
      "epoch": 0.1925,
      "step": 77
    },
    {
      "epoch": 0.195,
      "grad_norm": 14.395353317260742,
      "learning_rate": 2.542105263157895e-06,
      "loss": 0.4016,
      "step": 78
    },
    {
      "Batch Mean": 0.34730714559555054,
      "accuracy": 0.8203125,
      "epoch": 0.195,
      "step": 78
    },
    {
      "epoch": 0.1975,
      "grad_norm": 9.122536659240723,
      "learning_rate": 2.5342105263157892e-06,
      "loss": 0.3994,
      "step": 79
    },
    {
      "Batch Mean": 0.36483922600746155,
      "accuracy": 0.828125,
      "epoch": 0.1975,
      "step": 79
    },
    {
      "epoch": 0.2,
      "grad_norm": 10.289307594299316,
      "learning_rate": 2.526315789473684e-06,
      "loss": 0.368,
      "step": 80
    },
    {
      "Batch Mean": -0.20039799809455872,
      "accuracy": 0.8125,
      "epoch": 0.2,
      "step": 80
    },
    {
      "epoch": 0.2025,
      "grad_norm": 7.813342094421387,
      "learning_rate": 2.5184210526315787e-06,
      "loss": 0.3679,
      "step": 81
    },
    {
      "Batch Mean": -0.1629079282283783,
      "accuracy": 0.8203125,
      "epoch": 0.2025,
      "step": 81
    },
    {
      "epoch": 0.205,
      "grad_norm": 7.504952430725098,
      "learning_rate": 2.510526315789474e-06,
      "loss": 0.3823,
      "step": 82
    },
    {
      "Batch Mean": -0.07863587141036987,
      "accuracy": 0.8046875,
      "epoch": 0.205,
      "step": 82
    },
    {
      "epoch": 0.2075,
      "grad_norm": 7.721461296081543,
      "learning_rate": 2.5026315789473686e-06,
      "loss": 0.3967,
      "step": 83
    },
    {
      "Batch Mean": 0.17194491624832153,
      "accuracy": 0.8125,
      "epoch": 0.2075,
      "step": 83
    },
    {
      "epoch": 0.21,
      "grad_norm": 8.062063217163086,
      "learning_rate": 2.4947368421052634e-06,
      "loss": 0.4195,
      "step": 84
    },
    {
      "Batch Mean": 0.11748838424682617,
      "accuracy": 0.8359375,
      "epoch": 0.21,
      "step": 84
    },
    {
      "epoch": 0.2125,
      "grad_norm": 8.320355415344238,
      "learning_rate": 2.486842105263158e-06,
      "loss": 0.3864,
      "step": 85
    },
    {
      "Batch Mean": -0.40801382064819336,
      "accuracy": 0.7734375,
      "epoch": 0.2125,
      "step": 85
    },
    {
      "epoch": 0.215,
      "grad_norm": 14.8066987991333,
      "learning_rate": 2.478947368421053e-06,
      "loss": 0.4639,
      "step": 86
    },
    {
      "Batch Mean": -0.35811758041381836,
      "accuracy": 0.7734375,
      "epoch": 0.215,
      "step": 86
    },
    {
      "epoch": 0.2175,
      "grad_norm": 14.074177742004395,
      "learning_rate": 2.4710526315789476e-06,
      "loss": 0.4999,
      "step": 87
    },
    {
      "Batch Mean": 0.278584361076355,
      "accuracy": 0.7734375,
      "epoch": 0.2175,
      "step": 87
    },
    {
      "epoch": 0.22,
      "grad_norm": 10.59004020690918,
      "learning_rate": 2.4631578947368424e-06,
      "loss": 0.404,
      "step": 88
    },
    {
      "Batch Mean": -0.21054387092590332,
      "accuracy": 0.84375,
      "epoch": 0.22,
      "step": 88
    },
    {
      "epoch": 0.2225,
      "grad_norm": 8.488351821899414,
      "learning_rate": 2.4552631578947367e-06,
      "loss": 0.3517,
      "step": 89
    },
    {
      "Batch Mean": -0.22967231273651123,
      "accuracy": 0.84375,
      "epoch": 0.2225,
      "step": 89
    },
    {
      "epoch": 0.225,
      "grad_norm": 10.347502708435059,
      "learning_rate": 2.4473684210526314e-06,
      "loss": 0.3382,
      "step": 90
    },
    {
      "Batch Mean": 0.1892259418964386,
      "accuracy": 0.7890625,
      "epoch": 0.225,
      "step": 90
    },
    {
      "epoch": 0.2275,
      "grad_norm": 10.092534065246582,
      "learning_rate": 2.439473684210526e-06,
      "loss": 0.4586,
      "step": 91
    },
    {
      "Batch Mean": 0.516247570514679,
      "accuracy": 0.8828125,
      "epoch": 0.2275,
      "step": 91
    },
    {
      "epoch": 0.23,
      "grad_norm": 13.944212913513184,
      "learning_rate": 2.431578947368421e-06,
      "loss": 0.3067,
      "step": 92
    },
    {
      "Batch Mean": 0.056681275367736816,
      "accuracy": 0.7890625,
      "epoch": 0.23,
      "step": 92
    },
    {
      "epoch": 0.2325,
      "grad_norm": 8.53646183013916,
      "learning_rate": 2.4236842105263157e-06,
      "loss": 0.3946,
      "step": 93
    },
    {
      "Batch Mean": -0.47129684686660767,
      "accuracy": 0.8359375,
      "epoch": 0.2325,
      "step": 93
    },
    {
      "epoch": 0.235,
      "grad_norm": 11.441537857055664,
      "learning_rate": 2.4157894736842104e-06,
      "loss": 0.3891,
      "step": 94
    },
    {
      "Batch Mean": -0.3262195587158203,
      "accuracy": 0.84375,
      "epoch": 0.235,
      "step": 94
    },
    {
      "epoch": 0.2375,
      "grad_norm": 10.751629829406738,
      "learning_rate": 2.4078947368421056e-06,
      "loss": 0.3578,
      "step": 95
    },
    {
      "Batch Mean": -0.6593484878540039,
      "accuracy": 0.828125,
      "epoch": 0.2375,
      "step": 95
    },
    {
      "epoch": 0.24,
      "grad_norm": 15.914350509643555,
      "learning_rate": 2.4000000000000003e-06,
      "loss": 0.4204,
      "step": 96
    },
    {
      "Batch Mean": -0.32147085666656494,
      "accuracy": 0.8359375,
      "epoch": 0.24,
      "step": 96
    },
    {
      "epoch": 0.2425,
      "grad_norm": 11.814691543579102,
      "learning_rate": 2.392105263157895e-06,
      "loss": 0.3892,
      "step": 97
    },
    {
      "Batch Mean": 0.49291136860847473,
      "accuracy": 0.875,
      "epoch": 0.2425,
      "step": 97
    },
    {
      "epoch": 0.245,
      "grad_norm": 11.86365795135498,
      "learning_rate": 2.38421052631579e-06,
      "loss": 0.3264,
      "step": 98
    },
    {
      "Batch Mean": 0.49174070358276367,
      "accuracy": 0.8515625,
      "epoch": 0.245,
      "step": 98
    },
    {
      "epoch": 0.2475,
      "grad_norm": 10.874544143676758,
      "learning_rate": 2.376315789473684e-06,
      "loss": 0.3847,
      "step": 99
    },
    {
      "Batch Mean": 0.6919900178909302,
      "accuracy": 0.796875,
      "epoch": 0.2475,
      "step": 99
    },
    {
      "epoch": 0.25,
      "grad_norm": 15.105185508728027,
      "learning_rate": 2.368421052631579e-06,
      "loss": 0.4066,
      "step": 100
    },
    {
      "Batch Mean": 0.061200618743896484,
      "accuracy": 0.7734375,
      "epoch": 0.25,
      "step": 100
    },
    {
      "epoch": 0.2525,
      "grad_norm": 8.795661926269531,
      "learning_rate": 2.3605263157894736e-06,
      "loss": 0.4193,
      "step": 101
    },
    {
      "Batch Mean": -0.4929161071777344,
      "accuracy": 0.7890625,
      "epoch": 0.2525,
      "step": 101
    },
    {
      "epoch": 0.255,
      "grad_norm": 12.120612144470215,
      "learning_rate": 2.3526315789473684e-06,
      "loss": 0.4833,
      "step": 102
    },
    {
      "Batch Mean": -0.48431068658828735,
      "accuracy": 0.859375,
      "epoch": 0.255,
      "step": 102
    },
    {
      "epoch": 0.2575,
      "grad_norm": 10.263632774353027,
      "learning_rate": 2.344736842105263e-06,
      "loss": 0.3397,
      "step": 103
    },
    {
      "Batch Mean": -0.2162787914276123,
      "accuracy": 0.8203125,
      "epoch": 0.2575,
      "step": 103
    },
    {
      "epoch": 0.26,
      "grad_norm": 7.323156356811523,
      "learning_rate": 2.336842105263158e-06,
      "loss": 0.3688,
      "step": 104
    },
    {
      "Batch Mean": 0.2203059196472168,
      "accuracy": 0.8515625,
      "epoch": 0.26,
      "step": 104
    },
    {
      "epoch": 0.2625,
      "grad_norm": 7.128310203552246,
      "learning_rate": 2.3289473684210526e-06,
      "loss": 0.3454,
      "step": 105
    },
    {
      "Batch Mean": 0.19034957885742188,
      "accuracy": 0.8828125,
      "epoch": 0.2625,
      "step": 105
    },
    {
      "epoch": 0.265,
      "grad_norm": 7.388917446136475,
      "learning_rate": 2.3210526315789473e-06,
      "loss": 0.3317,
      "step": 106
    },
    {
      "Batch Mean": -0.0413057804107666,
      "accuracy": 0.796875,
      "epoch": 0.265,
      "step": 106
    },
    {
      "epoch": 0.2675,
      "grad_norm": 9.412615776062012,
      "learning_rate": 2.313157894736842e-06,
      "loss": 0.4393,
      "step": 107
    },
    {
      "Batch Mean": 0.473476767539978,
      "accuracy": 0.875,
      "epoch": 0.2675,
      "step": 107
    },
    {
      "epoch": 0.27,
      "grad_norm": 11.040026664733887,
      "learning_rate": 2.305263157894737e-06,
      "loss": 0.35,
      "step": 108
    },
    {
      "Batch Mean": 0.37884294986724854,
      "accuracy": 0.875,
      "epoch": 0.27,
      "step": 108
    },
    {
      "epoch": 0.2725,
      "grad_norm": 12.01086711883545,
      "learning_rate": 2.2973684210526316e-06,
      "loss": 0.3001,
      "step": 109
    },
    {
      "Batch Mean": 0.4409904479980469,
      "accuracy": 0.8828125,
      "epoch": 0.2725,
      "step": 109
    },
    {
      "epoch": 0.275,
      "grad_norm": 12.842291831970215,
      "learning_rate": 2.2894736842105263e-06,
      "loss": 0.2767,
      "step": 110
    },
    {
      "Batch Mean": -0.12960529327392578,
      "accuracy": 0.7890625,
      "epoch": 0.275,
      "step": 110
    },
    {
      "epoch": 0.2775,
      "grad_norm": 11.811309814453125,
      "learning_rate": 2.281578947368421e-06,
      "loss": 0.4361,
      "step": 111
    },
    {
      "Batch Mean": -0.3953084945678711,
      "accuracy": 0.7578125,
      "epoch": 0.2775,
      "step": 111
    },
    {
      "epoch": 0.28,
      "grad_norm": 14.014053344726562,
      "learning_rate": 2.273684210526316e-06,
      "loss": 0.5185,
      "step": 112
    },
    {
      "Batch Mean": -0.16724836826324463,
      "accuracy": 0.859375,
      "epoch": 0.28,
      "step": 112
    },
    {
      "epoch": 0.2825,
      "grad_norm": 7.608469486236572,
      "learning_rate": 2.2657894736842106e-06,
      "loss": 0.3033,
      "step": 113
    },
    {
      "Batch Mean": -0.36876964569091797,
      "accuracy": 0.8203125,
      "epoch": 0.2825,
      "step": 113
    },
    {
      "epoch": 0.285,
      "grad_norm": 11.263389587402344,
      "learning_rate": 2.2578947368421053e-06,
      "loss": 0.4069,
      "step": 114
    },
    {
      "Batch Mean": 0.22658658027648926,
      "accuracy": 0.7734375,
      "epoch": 0.285,
      "step": 114
    },
    {
      "epoch": 0.2875,
      "grad_norm": 11.426595687866211,
      "learning_rate": 2.25e-06,
      "loss": 0.4412,
      "step": 115
    },
    {
      "Batch Mean": 0.5346584320068359,
      "accuracy": 0.8359375,
      "epoch": 0.2875,
      "step": 115
    },
    {
      "epoch": 0.29,
      "grad_norm": 11.347992897033691,
      "learning_rate": 2.242105263157895e-06,
      "loss": 0.3996,
      "step": 116
    },
    {
      "Batch Mean": -0.40567731857299805,
      "accuracy": 0.8671875,
      "epoch": 0.29,
      "step": 116
    },
    {
      "epoch": 0.2925,
      "grad_norm": 9.157360076904297,
      "learning_rate": 2.2342105263157895e-06,
      "loss": 0.3318,
      "step": 117
    },
    {
      "Batch Mean": -0.1567370891571045,
      "accuracy": 0.8046875,
      "epoch": 0.2925,
      "step": 117
    },
    {
      "epoch": 0.295,
      "grad_norm": 12.001374244689941,
      "learning_rate": 2.2263157894736843e-06,
      "loss": 0.4401,
      "step": 118
    },
    {
      "Batch Mean": -0.357755184173584,
      "accuracy": 0.8671875,
      "epoch": 0.295,
      "step": 118
    },
    {
      "epoch": 0.2975,
      "grad_norm": 8.508570671081543,
      "learning_rate": 2.218421052631579e-06,
      "loss": 0.3305,
      "step": 119
    },
    {
      "Batch Mean": -0.35381609201431274,
      "accuracy": 0.8046875,
      "epoch": 0.2975,
      "step": 119
    },
    {
      "epoch": 0.3,
      "grad_norm": 8.094682693481445,
      "learning_rate": 2.2105263157894738e-06,
      "loss": 0.3984,
      "step": 120
    },
    {
      "Batch Mean": 0.22805237770080566,
      "accuracy": 0.8515625,
      "epoch": 0.3,
      "step": 120
    },
    {
      "epoch": 0.3025,
      "grad_norm": 9.788799285888672,
      "learning_rate": 2.2026315789473685e-06,
      "loss": 0.3478,
      "step": 121
    },
    {
      "Batch Mean": 0.5038647651672363,
      "accuracy": 0.859375,
      "epoch": 0.3025,
      "step": 121
    },
    {
      "epoch": 0.305,
      "grad_norm": 9.84559440612793,
      "learning_rate": 2.1947368421052633e-06,
      "loss": 0.3561,
      "step": 122
    },
    {
      "Batch Mean": 0.7104835510253906,
      "accuracy": 0.828125,
      "epoch": 0.305,
      "step": 122
    },
    {
      "epoch": 0.3075,
      "grad_norm": 11.43686580657959,
      "learning_rate": 2.186842105263158e-06,
      "loss": 0.3686,
      "step": 123
    },
    {
      "Batch Mean": -0.007034778594970703,
      "accuracy": 0.796875,
      "epoch": 0.3075,
      "step": 123
    },
    {
      "epoch": 0.31,
      "grad_norm": 8.770352363586426,
      "learning_rate": 2.1789473684210528e-06,
      "loss": 0.3772,
      "step": 124
    },
    {
      "Batch Mean": 0.036529541015625,
      "accuracy": 0.7890625,
      "epoch": 0.31,
      "step": 124
    },
    {
      "epoch": 0.3125,
      "grad_norm": 13.17721939086914,
      "learning_rate": 2.1710526315789475e-06,
      "loss": 0.5148,
      "step": 125
    },
    {
      "Batch Mean": -0.30262577533721924,
      "accuracy": 0.875,
      "epoch": 0.3125,
      "step": 125
    },
    {
      "epoch": 0.315,
      "grad_norm": 8.691415786743164,
      "learning_rate": 2.1631578947368423e-06,
      "loss": 0.3671,
      "step": 126
    },
    {
      "Batch Mean": -0.12425446510314941,
      "accuracy": 0.8515625,
      "epoch": 0.315,
      "step": 126
    },
    {
      "epoch": 0.3175,
      "grad_norm": 8.049582481384277,
      "learning_rate": 2.155263157894737e-06,
      "loss": 0.3522,
      "step": 127
    },
    {
      "Batch Mean": -0.4603520631790161,
      "accuracy": 0.765625,
      "epoch": 0.3175,
      "step": 127
    },
    {
      "epoch": 0.32,
      "grad_norm": 10.966066360473633,
      "learning_rate": 2.1473684210526317e-06,
      "loss": 0.4656,
      "step": 128
    },
    {
      "Batch Mean": 0.0716085433959961,
      "accuracy": 0.796875,
      "epoch": 0.32,
      "step": 128
    },
    {
      "epoch": 0.3225,
      "grad_norm": 7.68458890914917,
      "learning_rate": 2.1394736842105265e-06,
      "loss": 0.3586,
      "step": 129
    },
    {
      "Batch Mean": 0.22112131118774414,
      "accuracy": 0.7890625,
      "epoch": 0.3225,
      "step": 129
    },
    {
      "epoch": 0.325,
      "grad_norm": 8.140663146972656,
      "learning_rate": 2.1315789473684212e-06,
      "loss": 0.4103,
      "step": 130
    },
    {
      "Batch Mean": 0.1299583911895752,
      "accuracy": 0.8125,
      "epoch": 0.325,
      "step": 130
    },
    {
      "epoch": 0.3275,
      "grad_norm": 7.390633583068848,
      "learning_rate": 2.123684210526316e-06,
      "loss": 0.3914,
      "step": 131
    },
    {
      "Batch Mean": 0.608335018157959,
      "accuracy": 0.921875,
      "epoch": 0.3275,
      "step": 131
    },
    {
      "epoch": 0.33,
      "grad_norm": 13.193960189819336,
      "learning_rate": 2.1157894736842103e-06,
      "loss": 0.2749,
      "step": 132
    },
    {
      "Batch Mean": 0.15748775005340576,
      "accuracy": 0.7578125,
      "epoch": 0.33,
      "step": 132
    },
    {
      "epoch": 0.3325,
      "grad_norm": 8.155619621276855,
      "learning_rate": 2.107894736842105e-06,
      "loss": 0.4138,
      "step": 133
    },
    {
      "Batch Mean": -0.010571569204330444,
      "accuracy": 0.796875,
      "epoch": 0.3325,
      "step": 133
    },
    {
      "epoch": 0.335,
      "grad_norm": 7.174466609954834,
      "learning_rate": 2.1e-06,
      "loss": 0.3603,
      "step": 134
    },
    {
      "Batch Mean": -0.17839515209197998,
      "accuracy": 0.8125,
      "epoch": 0.335,
      "step": 134
    },
    {
      "epoch": 0.3375,
      "grad_norm": 9.18850326538086,
      "learning_rate": 2.0921052631578945e-06,
      "loss": 0.3665,
      "step": 135
    },
    {
      "Batch Mean": -0.0516209602355957,
      "accuracy": 0.7578125,
      "epoch": 0.3375,
      "step": 135
    },
    {
      "epoch": 0.34,
      "grad_norm": 8.749617576599121,
      "learning_rate": 2.0842105263157897e-06,
      "loss": 0.4511,
      "step": 136
    },
    {
      "Batch Mean": -0.33418262004852295,
      "accuracy": 0.796875,
      "epoch": 0.34,
      "step": 136
    },
    {
      "epoch": 0.3425,
      "grad_norm": 10.268221855163574,
      "learning_rate": 2.0763157894736845e-06,
      "loss": 0.4395,
      "step": 137
    },
    {
      "Batch Mean": -0.208038330078125,
      "accuracy": 0.8515625,
      "epoch": 0.3425,
      "step": 137
    },
    {
      "epoch": 0.345,
      "grad_norm": 6.962666034698486,
      "learning_rate": 2.068421052631579e-06,
      "loss": 0.3682,
      "step": 138
    },
    {
      "Batch Mean": 0.1473088264465332,
      "accuracy": 0.8046875,
      "epoch": 0.345,
      "step": 138
    },
    {
      "epoch": 0.3475,
      "grad_norm": 8.331689834594727,
      "learning_rate": 2.060526315789474e-06,
      "loss": 0.3881,
      "step": 139
    },
    {
      "Batch Mean": 0.13666605949401855,
      "accuracy": 0.7734375,
      "epoch": 0.3475,
      "step": 139
    },
    {
      "epoch": 0.35,
      "grad_norm": 8.181924819946289,
      "learning_rate": 2.0526315789473687e-06,
      "loss": 0.4522,
      "step": 140
    },
    {
      "Batch Mean": -0.06362247467041016,
      "accuracy": 0.828125,
      "epoch": 0.35,
      "step": 140
    },
    {
      "epoch": 0.3525,
      "grad_norm": 7.695260047912598,
      "learning_rate": 2.0447368421052634e-06,
      "loss": 0.3219,
      "step": 141
    },
    {
      "Batch Mean": -0.1371777057647705,
      "accuracy": 0.8203125,
      "epoch": 0.3525,
      "step": 141
    },
    {
      "epoch": 0.355,
      "grad_norm": 7.281587600708008,
      "learning_rate": 2.0368421052631578e-06,
      "loss": 0.3907,
      "step": 142
    },
    {
      "Batch Mean": 0.5550127029418945,
      "accuracy": 0.84375,
      "epoch": 0.355,
      "step": 142
    },
    {
      "epoch": 0.3575,
      "grad_norm": 10.004101753234863,
      "learning_rate": 2.0289473684210525e-06,
      "loss": 0.3703,
      "step": 143
    },
    {
      "Batch Mean": 0.5781314373016357,
      "accuracy": 0.8828125,
      "epoch": 0.3575,
      "step": 143
    },
    {
      "epoch": 0.36,
      "grad_norm": 10.783740997314453,
      "learning_rate": 2.0210526315789473e-06,
      "loss": 0.3023,
      "step": 144
    },
    {
      "Batch Mean": -0.06548458337783813,
      "accuracy": 0.796875,
      "epoch": 0.36,
      "step": 144
    },
    {
      "epoch": 0.3625,
      "grad_norm": 8.395689010620117,
      "learning_rate": 2.013157894736842e-06,
      "loss": 0.3807,
      "step": 145
    },
    {
      "Batch Mean": -0.28943490982055664,
      "accuracy": 0.8125,
      "epoch": 0.3625,
      "step": 145
    },
    {
      "epoch": 0.365,
      "grad_norm": 8.962904930114746,
      "learning_rate": 2.0052631578947367e-06,
      "loss": 0.4011,
      "step": 146
    },
    {
      "Batch Mean": -0.23036813735961914,
      "accuracy": 0.8359375,
      "epoch": 0.365,
      "step": 146
    },
    {
      "epoch": 0.3675,
      "grad_norm": 8.533259391784668,
      "learning_rate": 1.9973684210526315e-06,
      "loss": 0.4078,
      "step": 147
    },
    {
      "Batch Mean": 0.04428744316101074,
      "accuracy": 0.8203125,
      "epoch": 0.3675,
      "step": 147
    },
    {
      "epoch": 0.37,
      "grad_norm": 9.401562690734863,
      "learning_rate": 1.9894736842105262e-06,
      "loss": 0.4861,
      "step": 148
    },
    {
      "Batch Mean": -0.631324052810669,
      "accuracy": 0.84375,
      "epoch": 0.37,
      "step": 148
    },
    {
      "epoch": 0.3725,
      "grad_norm": 11.169927597045898,
      "learning_rate": 1.9815789473684214e-06,
      "loss": 0.353,
      "step": 149
    },
    {
      "Batch Mean": -0.36252474784851074,
      "accuracy": 0.875,
      "epoch": 0.3725,
      "step": 149
    },
    {
      "epoch": 0.375,
      "grad_norm": 8.18213939666748,
      "learning_rate": 1.973684210526316e-06,
      "loss": 0.3192,
      "step": 150
    },
    {
      "Batch Mean": 0.1312960386276245,
      "accuracy": 0.8515625,
      "epoch": 0.375,
      "step": 150
    },
    {
      "epoch": 0.3775,
      "grad_norm": 7.884857654571533,
      "learning_rate": 1.965789473684211e-06,
      "loss": 0.3448,
      "step": 151
    },
    {
      "Batch Mean": -0.07881736755371094,
      "accuracy": 0.75,
      "epoch": 0.3775,
      "step": 151
    },
    {
      "epoch": 0.38,
      "grad_norm": 11.124417304992676,
      "learning_rate": 1.9578947368421052e-06,
      "loss": 0.5138,
      "step": 152
    },
    {
      "Batch Mean": 0.44060420989990234,
      "accuracy": 0.84375,
      "epoch": 0.38,
      "step": 152
    },
    {
      "epoch": 0.3825,
      "grad_norm": 8.98283576965332,
      "learning_rate": 1.95e-06,
      "loss": 0.3417,
      "step": 153
    },
    {
      "Batch Mean": 0.09381070733070374,
      "accuracy": 0.84375,
      "epoch": 0.3825,
      "step": 153
    },
    {
      "epoch": 0.385,
      "grad_norm": 6.34003210067749,
      "learning_rate": 1.9421052631578947e-06,
      "loss": 0.3136,
      "step": 154
    },
    {
      "Batch Mean": -0.005020737648010254,
      "accuracy": 0.8359375,
      "epoch": 0.385,
      "step": 154
    },
    {
      "epoch": 0.3875,
      "grad_norm": 6.930245399475098,
      "learning_rate": 1.9342105263157895e-06,
      "loss": 0.3927,
      "step": 155
    },
    {
      "Batch Mean": 0.38732171058654785,
      "accuracy": 0.859375,
      "epoch": 0.3875,
      "step": 155
    },
    {
      "epoch": 0.39,
      "grad_norm": 7.609610080718994,
      "learning_rate": 1.926315789473684e-06,
      "loss": 0.3484,
      "step": 156
    },
    {
      "Batch Mean": 0.1457352638244629,
      "accuracy": 0.859375,
      "epoch": 0.39,
      "step": 156
    },
    {
      "epoch": 0.3925,
      "grad_norm": 6.132632732391357,
      "learning_rate": 1.918421052631579e-06,
      "loss": 0.3313,
      "step": 157
    },
    {
      "Batch Mean": -0.12105239927768707,
      "accuracy": 0.859375,
      "epoch": 0.3925,
      "step": 157
    },
    {
      "epoch": 0.395,
      "grad_norm": 6.948753356933594,
      "learning_rate": 1.9105263157894737e-06,
      "loss": 0.3103,
      "step": 158
    },
    {
      "Batch Mean": -0.2733367681503296,
      "accuracy": 0.8125,
      "epoch": 0.395,
      "step": 158
    },
    {
      "epoch": 0.3975,
      "grad_norm": 8.32777214050293,
      "learning_rate": 1.9026315789473684e-06,
      "loss": 0.3927,
      "step": 159
    },
    {
      "Batch Mean": -0.2845277786254883,
      "accuracy": 0.8046875,
      "epoch": 0.3975,
      "step": 159
    },
    {
      "epoch": 0.4,
      "grad_norm": 8.584951400756836,
      "learning_rate": 1.8947368421052632e-06,
      "loss": 0.3756,
      "step": 160
    },
    {
      "Batch Mean": 0.45987796783447266,
      "accuracy": 0.8828125,
      "epoch": 0.4,
      "step": 160
    },
    {
      "epoch": 0.4025,
      "grad_norm": 11.069204330444336,
      "learning_rate": 1.8868421052631577e-06,
      "loss": 0.3221,
      "step": 161
    },
    {
      "Batch Mean": 0.05443763732910156,
      "accuracy": 0.8046875,
      "epoch": 0.4025,
      "step": 161
    },
    {
      "epoch": 0.405,
      "grad_norm": 8.764575958251953,
      "learning_rate": 1.8789473684210525e-06,
      "loss": 0.3762,
      "step": 162
    },
    {
      "Batch Mean": 0.0884392112493515,
      "accuracy": 0.8203125,
      "epoch": 0.405,
      "step": 162
    },
    {
      "epoch": 0.4075,
      "grad_norm": 8.510726928710938,
      "learning_rate": 1.8710526315789476e-06,
      "loss": 0.3937,
      "step": 163
    },
    {
      "Batch Mean": 0.06999784708023071,
      "accuracy": 0.859375,
      "epoch": 0.4075,
      "step": 163
    },
    {
      "epoch": 0.41,
      "grad_norm": 7.868905067443848,
      "learning_rate": 1.8631578947368424e-06,
      "loss": 0.2751,
      "step": 164
    },
    {
      "Batch Mean": 0.36394214630126953,
      "accuracy": 0.828125,
      "epoch": 0.41,
      "step": 164
    },
    {
      "epoch": 0.4125,
      "grad_norm": 12.59380054473877,
      "learning_rate": 1.855263157894737e-06,
      "loss": 0.399,
      "step": 165
    },
    {
      "Batch Mean": -0.03789353370666504,
      "accuracy": 0.8203125,
      "epoch": 0.4125,
      "step": 165
    },
    {
      "epoch": 0.415,
      "grad_norm": 8.086974143981934,
      "learning_rate": 1.8473684210526317e-06,
      "loss": 0.3637,
      "step": 166
    },
    {
      "Batch Mean": 0.06824040412902832,
      "accuracy": 0.8515625,
      "epoch": 0.415,
      "step": 166
    },
    {
      "epoch": 0.4175,
      "grad_norm": 9.723653793334961,
      "learning_rate": 1.8394736842105264e-06,
      "loss": 0.394,
      "step": 167
    },
    {
      "Batch Mean": -0.5270748138427734,
      "accuracy": 0.8203125,
      "epoch": 0.4175,
      "step": 167
    },
    {
      "epoch": 0.42,
      "grad_norm": 13.736504554748535,
      "learning_rate": 1.8315789473684211e-06,
      "loss": 0.429,
      "step": 168
    },
    {
      "Batch Mean": -0.12757277488708496,
      "accuracy": 0.8359375,
      "epoch": 0.42,
      "step": 168
    },
    {
      "epoch": 0.4225,
      "grad_norm": 8.577059745788574,
      "learning_rate": 1.8236842105263159e-06,
      "loss": 0.3599,
      "step": 169
    },
    {
      "Batch Mean": -0.024989277124404907,
      "accuracy": 0.8203125,
      "epoch": 0.4225,
      "step": 169
    },
    {
      "epoch": 0.425,
      "grad_norm": 8.25197982788086,
      "learning_rate": 1.8157894736842106e-06,
      "loss": 0.4185,
      "step": 170
    },
    {
      "Batch Mean": 0.03246608376502991,
      "accuracy": 0.875,
      "epoch": 0.425,
      "step": 170
    },
    {
      "epoch": 0.4275,
      "grad_norm": 5.9294819831848145,
      "learning_rate": 1.8078947368421052e-06,
      "loss": 0.2621,
      "step": 171
    },
    {
      "Batch Mean": 0.26308655738830566,
      "accuracy": 0.828125,
      "epoch": 0.4275,
      "step": 171
    },
    {
      "epoch": 0.43,
      "grad_norm": 7.570163249969482,
      "learning_rate": 1.8e-06,
      "loss": 0.3638,
      "step": 172
    },
    {
      "Batch Mean": -0.25519466400146484,
      "accuracy": 0.828125,
      "epoch": 0.43,
      "step": 172
    },
    {
      "epoch": 0.4325,
      "grad_norm": 7.611836910247803,
      "learning_rate": 1.7921052631578947e-06,
      "loss": 0.3867,
      "step": 173
    },
    {
      "Batch Mean": -0.01856723427772522,
      "accuracy": 0.8515625,
      "epoch": 0.4325,
      "step": 173
    },
    {
      "epoch": 0.435,
      "grad_norm": 7.200859546661377,
      "learning_rate": 1.7842105263157894e-06,
      "loss": 0.3596,
      "step": 174
    },
    {
      "Batch Mean": -0.06785082817077637,
      "accuracy": 0.8125,
      "epoch": 0.435,
      "step": 174
    },
    {
      "epoch": 0.4375,
      "grad_norm": 8.669882774353027,
      "learning_rate": 1.7763157894736842e-06,
      "loss": 0.4487,
      "step": 175
    },
    {
      "Batch Mean": -0.3832893371582031,
      "accuracy": 0.828125,
      "epoch": 0.4375,
      "step": 175
    },
    {
      "epoch": 0.44,
      "grad_norm": 8.320089340209961,
      "learning_rate": 1.768421052631579e-06,
      "loss": 0.4209,
      "step": 176
    },
    {
      "Batch Mean": -0.08297932147979736,
      "accuracy": 0.8203125,
      "epoch": 0.44,
      "step": 176
    },
    {
      "epoch": 0.4425,
      "grad_norm": 7.819711208343506,
      "learning_rate": 1.7605263157894739e-06,
      "loss": 0.3551,
      "step": 177
    },
    {
      "Batch Mean": 0.37918123602867126,
      "accuracy": 0.8125,
      "epoch": 0.4425,
      "step": 177
    },
    {
      "epoch": 0.445,
      "grad_norm": 8.0424165725708,
      "learning_rate": 1.7526315789473686e-06,
      "loss": 0.3594,
      "step": 178
    },
    {
      "Batch Mean": 0.05135989189147949,
      "accuracy": 0.890625,
      "epoch": 0.445,
      "step": 178
    },
    {
      "epoch": 0.4475,
      "grad_norm": 6.7139811515808105,
      "learning_rate": 1.7447368421052633e-06,
      "loss": 0.3117,
      "step": 179
    },
    {
      "Batch Mean": 0.06295812129974365,
      "accuracy": 0.828125,
      "epoch": 0.4475,
      "step": 179
    },
    {
      "epoch": 0.45,
      "grad_norm": 7.755770683288574,
      "learning_rate": 1.736842105263158e-06,
      "loss": 0.396,
      "step": 180
    },
    {
      "Batch Mean": -0.19158035516738892,
      "accuracy": 0.8671875,
      "epoch": 0.45,
      "step": 180
    },
    {
      "epoch": 0.4525,
      "grad_norm": 6.9720778465271,
      "learning_rate": 1.7289473684210526e-06,
      "loss": 0.3034,
      "step": 181
    },
    {
      "Batch Mean": 0.045849740505218506,
      "accuracy": 0.859375,
      "epoch": 0.4525,
      "step": 181
    },
    {
      "epoch": 0.455,
      "grad_norm": 6.858092784881592,
      "learning_rate": 1.7210526315789474e-06,
      "loss": 0.3629,
      "step": 182
    },
    {
      "Batch Mean": 0.46304988861083984,
      "accuracy": 0.8125,
      "epoch": 0.455,
      "step": 182
    },
    {
      "epoch": 0.4575,
      "grad_norm": 8.993224143981934,
      "learning_rate": 1.7131578947368421e-06,
      "loss": 0.3538,
      "step": 183
    },
    {
      "Batch Mean": 0.057231903076171875,
      "accuracy": 0.7890625,
      "epoch": 0.4575,
      "step": 183
    },
    {
      "epoch": 0.46,
      "grad_norm": 11.233847618103027,
      "learning_rate": 1.7052631578947369e-06,
      "loss": 0.4784,
      "step": 184
    },
    {
      "Batch Mean": -0.06488056480884552,
      "accuracy": 0.859375,
      "epoch": 0.46,
      "step": 184
    },
    {
      "epoch": 0.4625,
      "grad_norm": 8.108458518981934,
      "learning_rate": 1.6973684210526316e-06,
      "loss": 0.2988,
      "step": 185
    },
    {
      "Batch Mean": -0.4033060073852539,
      "accuracy": 0.78125,
      "epoch": 0.4625,
      "step": 185
    },
    {
      "epoch": 0.465,
      "grad_norm": 9.097691535949707,
      "learning_rate": 1.6894736842105264e-06,
      "loss": 0.4141,
      "step": 186
    },
    {
      "Batch Mean": -0.41214847564697266,
      "accuracy": 0.8828125,
      "epoch": 0.465,
      "step": 186
    },
    {
      "epoch": 0.4675,
      "grad_norm": 8.145352363586426,
      "learning_rate": 1.6815789473684209e-06,
      "loss": 0.3051,
      "step": 187
    },
    {
      "Batch Mean": 0.14345920085906982,
      "accuracy": 0.8671875,
      "epoch": 0.4675,
      "step": 187
    },
    {
      "epoch": 0.47,
      "grad_norm": 7.163969039916992,
      "learning_rate": 1.6736842105263156e-06,
      "loss": 0.3405,
      "step": 188
    },
    {
      "Batch Mean": 0.25019216537475586,
      "accuracy": 0.8359375,
      "epoch": 0.47,
      "step": 188
    },
    {
      "epoch": 0.4725,
      "grad_norm": 8.14911937713623,
      "learning_rate": 1.6657894736842104e-06,
      "loss": 0.3334,
      "step": 189
    },
    {
      "Batch Mean": 0.030582189559936523,
      "accuracy": 0.8203125,
      "epoch": 0.4725,
      "step": 189
    },
    {
      "epoch": 0.475,
      "grad_norm": 7.993282318115234,
      "learning_rate": 1.6578947368421056e-06,
      "loss": 0.3885,
      "step": 190
    },
    {
      "Batch Mean": 0.14673519134521484,
      "accuracy": 0.8125,
      "epoch": 0.475,
      "step": 190
    },
    {
      "epoch": 0.4775,
      "grad_norm": 9.056065559387207,
      "learning_rate": 1.65e-06,
      "loss": 0.4084,
      "step": 191
    },
    {
      "Batch Mean": 0.3741884231567383,
      "accuracy": 0.875,
      "epoch": 0.4775,
      "step": 191
    },
    {
      "epoch": 0.48,
      "grad_norm": 8.857176780700684,
      "learning_rate": 1.6421052631578948e-06,
      "loss": 0.3182,
      "step": 192
    },
    {
      "Batch Mean": 0.41150104999542236,
      "accuracy": 0.859375,
      "epoch": 0.48,
      "step": 192
    },
    {
      "epoch": 0.4825,
      "grad_norm": 8.556700706481934,
      "learning_rate": 1.6342105263157896e-06,
      "loss": 0.3261,
      "step": 193
    },
    {
      "Batch Mean": -0.4138193130493164,
      "accuracy": 0.8359375,
      "epoch": 0.4825,
      "step": 193
    },
    {
      "epoch": 0.485,
      "grad_norm": 7.911004543304443,
      "learning_rate": 1.6263157894736843e-06,
      "loss": 0.3395,
      "step": 194
    },
    {
      "Batch Mean": -0.7141661643981934,
      "accuracy": 0.765625,
      "epoch": 0.485,
      "step": 194
    },
    {
      "epoch": 0.4875,
      "grad_norm": 10.288289070129395,
      "learning_rate": 1.618421052631579e-06,
      "loss": 0.4338,
      "step": 195
    },
    {
      "Batch Mean": -0.5950565338134766,
      "accuracy": 0.8203125,
      "epoch": 0.4875,
      "step": 195
    },
    {
      "epoch": 0.49,
      "grad_norm": 9.772676467895508,
      "learning_rate": 1.6105263157894738e-06,
      "loss": 0.4012,
      "step": 196
    },
    {
      "Batch Mean": -0.40017926692962646,
      "accuracy": 0.84375,
      "epoch": 0.49,
      "step": 196
    },
    {
      "epoch": 0.4925,
      "grad_norm": 7.894969463348389,
      "learning_rate": 1.6026315789473683e-06,
      "loss": 0.363,
      "step": 197
    },
    {
      "Batch Mean": 0.4873809814453125,
      "accuracy": 0.8203125,
      "epoch": 0.4925,
      "step": 197
    },
    {
      "epoch": 0.495,
      "grad_norm": 9.627370834350586,
      "learning_rate": 1.594736842105263e-06,
      "loss": 0.3737,
      "step": 198
    },
    {
      "Batch Mean": 0.2109694480895996,
      "accuracy": 0.765625,
      "epoch": 0.495,
      "step": 198
    },
    {
      "epoch": 0.4975,
      "grad_norm": 8.528368949890137,
      "learning_rate": 1.5868421052631578e-06,
      "loss": 0.4261,
      "step": 199
    },
    {
      "Batch Mean": 0.3447328805923462,
      "accuracy": 0.8125,
      "epoch": 0.4975,
      "step": 199
    },
    {
      "epoch": 0.5,
      "grad_norm": 8.279525756835938,
      "learning_rate": 1.5789473684210526e-06,
      "loss": 0.3857,
      "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
}