{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.4891101431238334,
  "eval_steps": 10,
  "global_step": 2000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0012445550715619166,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 2.8073,
      "step": 1
    },
    {
      "epoch": 0.002489110143123833,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 2.8285,
      "step": 2
    },
    {
      "epoch": 0.00373366521468575,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 2.7975,
      "step": 3
    },
    {
      "epoch": 0.004978220286247666,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 2.798,
      "step": 4
    },
    {
      "epoch": 0.006222775357809583,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 2.8443,
      "step": 5
    },
    {
      "epoch": 0.0074673304293715,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 2.8074,
      "step": 6
    },
    {
      "epoch": 0.008711885500933417,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 2.7588,
      "step": 7
    },
    {
      "epoch": 0.009956440572495333,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 2.8136,
      "step": 8
    },
    {
      "epoch": 0.01120099564405725,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 2.7397,
      "step": 9
    },
    {
      "epoch": 0.012445550715619166,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 2.7745,
      "step": 10
    },
    {
      "epoch": 0.012445550715619166,
      "eval_loss": 2.8210322856903076,
      "eval_runtime": 43.0476,
      "eval_samples_per_second": 23.23,
      "eval_steps_per_second": 0.976,
      "step": 10
    },
    {
      "epoch": 0.013690105787181083,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 2.8275,
      "step": 11
    },
    {
      "epoch": 0.014934660858743,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 2.7762,
      "step": 12
    },
    {
      "epoch": 0.016179215930304917,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 2.7799,
      "step": 13
    },
    {
      "epoch": 0.017423771001866834,
      "grad_norm": 0.0,
      "learning_rate": 0.0,
      "loss": 2.7906,
      "step": 14
    },
    {
      "epoch": 0.018668326073428748,
      "grad_norm": 5.01304292678833,
      "learning_rate": 2.0746887966804982e-08,
      "loss": 2.7712,
      "step": 15
    },
    {
      "epoch": 0.019912881144990666,
      "grad_norm": 4.976474761962891,
      "learning_rate": 4.1493775933609963e-08,
      "loss": 2.829,
      "step": 16
    },
    {
      "epoch": 0.021157436216552583,
      "grad_norm": 5.539012908935547,
      "learning_rate": 6.224066390041494e-08,
      "loss": 2.813,
      "step": 17
    },
    {
      "epoch": 0.0224019912881145,
      "grad_norm": 5.4900407791137695,
      "learning_rate": 8.298755186721993e-08,
      "loss": 2.8126,
      "step": 18
    },
    {
      "epoch": 0.023646546359676415,
      "grad_norm": 5.457218647003174,
      "learning_rate": 1.037344398340249e-07,
      "loss": 2.7482,
      "step": 19
    },
    {
      "epoch": 0.024891101431238332,
      "grad_norm": 5.231036186218262,
      "learning_rate": 1.2448132780082988e-07,
      "loss": 2.8323,
      "step": 20
    },
    {
      "epoch": 0.024891101431238332,
      "eval_loss": 2.819715738296509,
      "eval_runtime": 43.9884,
      "eval_samples_per_second": 22.733,
      "eval_steps_per_second": 0.955,
      "step": 20
    },
    {
      "epoch": 0.02613565650280025,
      "grad_norm": 5.337130546569824,
      "learning_rate": 1.4522821576763488e-07,
      "loss": 2.776,
      "step": 21
    },
    {
      "epoch": 0.027380211574362167,
      "grad_norm": 4.973108291625977,
      "learning_rate": 1.6597510373443985e-07,
      "loss": 2.8239,
      "step": 22
    },
    {
      "epoch": 0.02862476664592408,
      "grad_norm": 4.8733906745910645,
      "learning_rate": 1.8672199170124483e-07,
      "loss": 2.8069,
      "step": 23
    },
    {
      "epoch": 0.029869321717486,
      "grad_norm": 5.061434745788574,
      "learning_rate": 2.074688796680498e-07,
      "loss": 2.8063,
      "step": 24
    },
    {
      "epoch": 0.031113876789047916,
      "grad_norm": 4.782018661499023,
      "learning_rate": 2.2821576763485478e-07,
      "loss": 2.7488,
      "step": 25
    },
    {
      "epoch": 0.03235843186060983,
      "grad_norm": 4.394975662231445,
      "learning_rate": 2.4896265560165975e-07,
      "loss": 2.7683,
      "step": 26
    },
    {
      "epoch": 0.03360298693217175,
      "grad_norm": 4.475763320922852,
      "learning_rate": 2.6970954356846476e-07,
      "loss": 2.8143,
      "step": 27
    },
    {
      "epoch": 0.03484754200373367,
      "grad_norm": 4.362309455871582,
      "learning_rate": 2.9045643153526976e-07,
      "loss": 2.7875,
      "step": 28
    },
    {
      "epoch": 0.03609209707529558,
      "grad_norm": 3.9444823265075684,
      "learning_rate": 3.112033195020747e-07,
      "loss": 2.766,
      "step": 29
    },
    {
      "epoch": 0.037336652146857496,
      "grad_norm": 3.4614129066467285,
      "learning_rate": 3.319502074688797e-07,
      "loss": 2.7644,
      "step": 30
    },
    {
      "epoch": 0.037336652146857496,
      "eval_loss": 2.772022008895874,
      "eval_runtime": 45.8513,
      "eval_samples_per_second": 21.81,
      "eval_steps_per_second": 0.916,
      "step": 30
    },
    {
      "epoch": 0.038581207218419414,
      "grad_norm": 3.449410915374756,
      "learning_rate": 3.5269709543568466e-07,
      "loss": 2.7653,
      "step": 31
    },
    {
      "epoch": 0.03982576228998133,
      "grad_norm": 3.3115482330322266,
      "learning_rate": 3.7344398340248966e-07,
      "loss": 2.7032,
      "step": 32
    },
    {
      "epoch": 0.04107031736154325,
      "grad_norm": 3.2057440280914307,
      "learning_rate": 3.941908713692946e-07,
      "loss": 2.7766,
      "step": 33
    },
    {
      "epoch": 0.042314872433105166,
      "grad_norm": 3.4503021240234375,
      "learning_rate": 4.149377593360996e-07,
      "loss": 2.7214,
      "step": 34
    },
    {
      "epoch": 0.043559427504667084,
      "grad_norm": 3.1477363109588623,
      "learning_rate": 4.3568464730290456e-07,
      "loss": 2.7333,
      "step": 35
    },
    {
      "epoch": 0.044803982576229,
      "grad_norm": 2.8025119304656982,
      "learning_rate": 4.5643153526970956e-07,
      "loss": 2.7535,
      "step": 36
    },
    {
      "epoch": 0.04604853764779091,
      "grad_norm": 2.967703342437744,
      "learning_rate": 4.771784232365145e-07,
      "loss": 2.7615,
      "step": 37
    },
    {
      "epoch": 0.04729309271935283,
      "grad_norm": 2.9308114051818848,
      "learning_rate": 4.979253112033195e-07,
      "loss": 2.7115,
      "step": 38
    },
    {
      "epoch": 0.04853764779091475,
      "grad_norm": 2.9203720092773438,
      "learning_rate": 5.186721991701245e-07,
      "loss": 2.701,
      "step": 39
    },
    {
      "epoch": 0.049782202862476664,
      "grad_norm": 2.8226709365844727,
      "learning_rate": 5.394190871369295e-07,
      "loss": 2.6267,
      "step": 40
    },
    {
      "epoch": 0.049782202862476664,
      "eval_loss": 2.6964941024780273,
      "eval_runtime": 46.5181,
      "eval_samples_per_second": 21.497,
      "eval_steps_per_second": 0.903,
      "step": 40
    },
    {
      "epoch": 0.05102675793403858,
      "grad_norm": 2.7212748527526855,
      "learning_rate": 5.601659751037345e-07,
      "loss": 2.6505,
      "step": 41
    },
    {
      "epoch": 0.0522713130056005,
      "grad_norm": 2.658881425857544,
      "learning_rate": 5.809128630705395e-07,
      "loss": 2.6655,
      "step": 42
    },
    {
      "epoch": 0.053515868077162417,
      "grad_norm": 2.4534730911254883,
      "learning_rate": 6.016597510373444e-07,
      "loss": 2.7098,
      "step": 43
    },
    {
      "epoch": 0.054760423148724334,
      "grad_norm": 2.4222075939178467,
      "learning_rate": 6.224066390041494e-07,
      "loss": 2.6793,
      "step": 44
    },
    {
      "epoch": 0.056004978220286245,
      "grad_norm": 2.5824973583221436,
      "learning_rate": 6.431535269709543e-07,
      "loss": 2.6212,
      "step": 45
    },
    {
      "epoch": 0.05724953329184816,
      "grad_norm": 2.5982468128204346,
      "learning_rate": 6.639004149377594e-07,
      "loss": 2.5901,
      "step": 46
    },
    {
      "epoch": 0.05849408836341008,
      "grad_norm": 2.213388442993164,
      "learning_rate": 6.846473029045644e-07,
      "loss": 2.6156,
      "step": 47
    },
    {
      "epoch": 0.059738643434972,
      "grad_norm": 2.097372531890869,
      "learning_rate": 7.053941908713693e-07,
      "loss": 2.6154,
      "step": 48
    },
    {
      "epoch": 0.060983198506533914,
      "grad_norm": 1.9788408279418945,
      "learning_rate": 7.261410788381744e-07,
      "loss": 2.5516,
      "step": 49
    },
    {
      "epoch": 0.06222775357809583,
      "grad_norm": 2.052147150039673,
      "learning_rate": 7.468879668049793e-07,
      "loss": 2.6128,
      "step": 50
    },
    {
      "epoch": 0.06222775357809583,
      "eval_loss": 2.6048905849456787,
      "eval_runtime": 51.2137,
      "eval_samples_per_second": 19.526,
      "eval_steps_per_second": 0.82,
      "step": 50
    },
    {
      "epoch": 0.06347230864965775,
      "grad_norm": 1.8841335773468018,
      "learning_rate": 7.676348547717843e-07,
      "loss": 2.5617,
      "step": 51
    },
    {
      "epoch": 0.06471686372121967,
      "grad_norm": 2.0841684341430664,
      "learning_rate": 7.883817427385892e-07,
      "loss": 2.5627,
      "step": 52
    },
    {
      "epoch": 0.06596141879278158,
      "grad_norm": 1.8119730949401855,
      "learning_rate": 8.091286307053943e-07,
      "loss": 2.5572,
      "step": 53
    },
    {
      "epoch": 0.0672059738643435,
      "grad_norm": 1.7168290615081787,
      "learning_rate": 8.298755186721992e-07,
      "loss": 2.5465,
      "step": 54
    },
    {
      "epoch": 0.06845052893590542,
      "grad_norm": 1.7731754779815674,
      "learning_rate": 8.506224066390042e-07,
      "loss": 2.543,
      "step": 55
    },
    {
      "epoch": 0.06969508400746734,
      "grad_norm": 1.9654561281204224,
      "learning_rate": 8.713692946058091e-07,
      "loss": 2.5547,
      "step": 56
    },
    {
      "epoch": 0.07093963907902924,
      "grad_norm": 1.8032375574111938,
      "learning_rate": 8.921161825726142e-07,
      "loss": 2.5319,
      "step": 57
    },
    {
      "epoch": 0.07218419415059116,
      "grad_norm": 1.8775644302368164,
      "learning_rate": 9.128630705394191e-07,
      "loss": 2.5533,
      "step": 58
    },
    {
      "epoch": 0.07342874922215308,
      "grad_norm": 1.6019998788833618,
      "learning_rate": 9.336099585062241e-07,
      "loss": 2.5402,
      "step": 59
    },
    {
      "epoch": 0.07467330429371499,
      "grad_norm": 1.7785435914993286,
      "learning_rate": 9.54356846473029e-07,
      "loss": 2.4704,
      "step": 60
    },
    {
      "epoch": 0.07467330429371499,
      "eval_loss": 2.541548252105713,
      "eval_runtime": 51.1601,
      "eval_samples_per_second": 19.546,
      "eval_steps_per_second": 0.821,
      "step": 60
    },
    {
      "epoch": 0.07591785936527691,
      "grad_norm": 1.5377200841903687,
      "learning_rate": 9.751037344398341e-07,
      "loss": 2.5028,
      "step": 61
    },
    {
      "epoch": 0.07716241443683883,
      "grad_norm": 1.7212916612625122,
      "learning_rate": 9.95850622406639e-07,
      "loss": 2.4621,
      "step": 62
    },
    {
      "epoch": 0.07840696950840075,
      "grad_norm": 1.632606029510498,
      "learning_rate": 1.0165975103734441e-06,
      "loss": 2.4868,
      "step": 63
    },
    {
      "epoch": 0.07965152457996266,
      "grad_norm": 1.8643872737884521,
      "learning_rate": 1.037344398340249e-06,
      "loss": 2.4582,
      "step": 64
    },
    {
      "epoch": 0.08089607965152458,
      "grad_norm": 1.8329377174377441,
      "learning_rate": 1.058091286307054e-06,
      "loss": 2.464,
      "step": 65
    },
    {
      "epoch": 0.0821406347230865,
      "grad_norm": 1.9488356113433838,
      "learning_rate": 1.078838174273859e-06,
      "loss": 2.4275,
      "step": 66
    },
    {
      "epoch": 0.08338518979464841,
      "grad_norm": 1.7663215398788452,
      "learning_rate": 1.099585062240664e-06,
      "loss": 2.516,
      "step": 67
    },
    {
      "epoch": 0.08462974486621033,
      "grad_norm": 1.7675347328186035,
      "learning_rate": 1.120331950207469e-06,
      "loss": 2.4535,
      "step": 68
    },
    {
      "epoch": 0.08587429993777225,
      "grad_norm": 1.7481404542922974,
      "learning_rate": 1.141078838174274e-06,
      "loss": 2.4089,
      "step": 69
    },
    {
      "epoch": 0.08711885500933417,
      "grad_norm": 1.7963330745697021,
      "learning_rate": 1.161825726141079e-06,
      "loss": 2.4408,
      "step": 70
    },
    {
      "epoch": 0.08711885500933417,
      "eval_loss": 2.466163396835327,
      "eval_runtime": 50.4204,
      "eval_samples_per_second": 19.833,
      "eval_steps_per_second": 0.833,
      "step": 70
    },
    {
      "epoch": 0.08836341008089608,
      "grad_norm": 2.053469657897949,
      "learning_rate": 1.182572614107884e-06,
      "loss": 2.408,
      "step": 71
    },
    {
      "epoch": 0.089607965152458,
      "grad_norm": 1.900294303894043,
      "learning_rate": 1.2033195020746888e-06,
      "loss": 2.4066,
      "step": 72
    },
    {
      "epoch": 0.09085252022401992,
      "grad_norm": 2.38637113571167,
      "learning_rate": 1.224066390041494e-06,
      "loss": 2.3548,
      "step": 73
    },
    {
      "epoch": 0.09209707529558182,
      "grad_norm": 1.8274579048156738,
      "learning_rate": 1.2448132780082988e-06,
      "loss": 2.3844,
      "step": 74
    },
    {
      "epoch": 0.09334163036714374,
      "grad_norm": 1.9040184020996094,
      "learning_rate": 1.2655601659751037e-06,
      "loss": 2.4406,
      "step": 75
    },
    {
      "epoch": 0.09458618543870566,
      "grad_norm": 2.0351619720458984,
      "learning_rate": 1.2863070539419086e-06,
      "loss": 2.3624,
      "step": 76
    },
    {
      "epoch": 0.09583074051026758,
      "grad_norm": 1.989876627922058,
      "learning_rate": 1.307053941908714e-06,
      "loss": 2.4097,
      "step": 77
    },
    {
      "epoch": 0.0970752955818295,
      "grad_norm": 1.8507741689682007,
      "learning_rate": 1.3278008298755188e-06,
      "loss": 2.4482,
      "step": 78
    },
    {
      "epoch": 0.09831985065339141,
      "grad_norm": 2.6118454933166504,
      "learning_rate": 1.3485477178423237e-06,
      "loss": 2.3178,
      "step": 79
    },
    {
      "epoch": 0.09956440572495333,
      "grad_norm": 1.9467326402664185,
      "learning_rate": 1.3692946058091288e-06,
      "loss": 2.3175,
      "step": 80
    },
    {
      "epoch": 0.09956440572495333,
      "eval_loss": 2.3719911575317383,
      "eval_runtime": 52.3293,
      "eval_samples_per_second": 19.11,
      "eval_steps_per_second": 0.803,
      "step": 80
    },
    {
      "epoch": 0.10080896079651525,
      "grad_norm": 2.4913370609283447,
      "learning_rate": 1.3900414937759337e-06,
      "loss": 2.3063,
      "step": 81
    },
    {
      "epoch": 0.10205351586807716,
      "grad_norm": 2.4838919639587402,
      "learning_rate": 1.4107883817427386e-06,
      "loss": 2.3534,
      "step": 82
    },
    {
      "epoch": 0.10329807093963908,
      "grad_norm": 2.798588991165161,
      "learning_rate": 1.4315352697095435e-06,
      "loss": 2.2732,
      "step": 83
    },
    {
      "epoch": 0.104542626011201,
      "grad_norm": 2.688399076461792,
      "learning_rate": 1.4522821576763488e-06,
      "loss": 2.3112,
      "step": 84
    },
    {
      "epoch": 0.10578718108276292,
      "grad_norm": 2.7367143630981445,
      "learning_rate": 1.4730290456431537e-06,
      "loss": 2.3055,
      "step": 85
    },
    {
      "epoch": 0.10703173615432483,
      "grad_norm": 2.6406664848327637,
      "learning_rate": 1.4937759336099586e-06,
      "loss": 2.312,
      "step": 86
    },
    {
      "epoch": 0.10827629122588675,
      "grad_norm": 2.703355312347412,
      "learning_rate": 1.5145228215767635e-06,
      "loss": 2.3305,
      "step": 87
    },
    {
      "epoch": 0.10952084629744867,
      "grad_norm": 2.420084238052368,
      "learning_rate": 1.5352697095435686e-06,
      "loss": 2.2608,
      "step": 88
    },
    {
      "epoch": 0.11076540136901059,
      "grad_norm": 2.999835729598999,
      "learning_rate": 1.5560165975103735e-06,
      "loss": 2.2389,
      "step": 89
    },
    {
      "epoch": 0.11200995644057249,
      "grad_norm": 2.35611629486084,
      "learning_rate": 1.5767634854771784e-06,
      "loss": 2.2671,
      "step": 90
    },
    {
      "epoch": 0.11200995644057249,
      "eval_loss": 2.2900290489196777,
      "eval_runtime": 52.1871,
      "eval_samples_per_second": 19.162,
      "eval_steps_per_second": 0.805,
      "step": 90
    },
    {
      "epoch": 0.1132545115121344,
      "grad_norm": 2.2686431407928467,
      "learning_rate": 1.5975103734439833e-06,
      "loss": 2.2405,
      "step": 91
    },
    {
      "epoch": 0.11449906658369632,
      "grad_norm": 2.883517265319824,
      "learning_rate": 1.6182572614107886e-06,
      "loss": 2.2364,
      "step": 92
    },
    {
      "epoch": 0.11574362165525824,
      "grad_norm": 2.4562909603118896,
      "learning_rate": 1.6390041493775935e-06,
      "loss": 2.2308,
      "step": 93
    },
    {
      "epoch": 0.11698817672682016,
      "grad_norm": 2.7968456745147705,
      "learning_rate": 1.6597510373443984e-06,
      "loss": 2.1959,
      "step": 94
    },
    {
      "epoch": 0.11823273179838208,
      "grad_norm": 2.8692259788513184,
      "learning_rate": 1.6804979253112035e-06,
      "loss": 2.215,
      "step": 95
    },
    {
      "epoch": 0.119477286869944,
      "grad_norm": 2.8436100482940674,
      "learning_rate": 1.7012448132780084e-06,
      "loss": 2.1816,
      "step": 96
    },
    {
      "epoch": 0.12072184194150591,
      "grad_norm": 2.4477386474609375,
      "learning_rate": 1.7219917012448133e-06,
      "loss": 2.1684,
      "step": 97
    },
    {
      "epoch": 0.12196639701306783,
      "grad_norm": 2.610046863555908,
      "learning_rate": 1.7427385892116182e-06,
      "loss": 2.1656,
      "step": 98
    },
    {
      "epoch": 0.12321095208462975,
      "grad_norm": 2.698709487915039,
      "learning_rate": 1.7634854771784235e-06,
      "loss": 2.2115,
      "step": 99
    },
    {
      "epoch": 0.12445550715619166,
      "grad_norm": 2.476600408554077,
      "learning_rate": 1.7842323651452284e-06,
      "loss": 2.1619,
      "step": 100
    },
    {
      "epoch": 0.12445550715619166,
      "eval_loss": 2.221313953399658,
      "eval_runtime": 51.6214,
      "eval_samples_per_second": 19.372,
      "eval_steps_per_second": 0.814,
      "step": 100
    },
    {
      "epoch": 0.12570006222775357,
      "grad_norm": 3.238584518432617,
      "learning_rate": 1.8049792531120333e-06,
      "loss": 2.1543,
      "step": 101
    },
    {
      "epoch": 0.1269446172993155,
      "grad_norm": 2.3609964847564697,
      "learning_rate": 1.8257261410788382e-06,
      "loss": 2.1602,
      "step": 102
    },
    {
      "epoch": 0.1281891723708774,
      "grad_norm": 3.3314859867095947,
      "learning_rate": 1.8464730290456433e-06,
      "loss": 2.1843,
      "step": 103
    },
    {
      "epoch": 0.12943372744243933,
      "grad_norm": 2.42441725730896,
      "learning_rate": 1.8672199170124482e-06,
      "loss": 2.135,
      "step": 104
    },
    {
      "epoch": 0.13067828251400124,
      "grad_norm": 3.2766177654266357,
      "learning_rate": 1.8879668049792531e-06,
      "loss": 2.158,
      "step": 105
    },
    {
      "epoch": 0.13192283758556317,
      "grad_norm": 2.8646531105041504,
      "learning_rate": 1.908713692946058e-06,
      "loss": 2.1214,
      "step": 106
    },
    {
      "epoch": 0.13316739265712507,
      "grad_norm": 2.8570213317871094,
      "learning_rate": 1.929460580912863e-06,
      "loss": 2.1345,
      "step": 107
    },
    {
      "epoch": 0.134411947728687,
      "grad_norm": 2.5014989376068115,
      "learning_rate": 1.9502074688796682e-06,
      "loss": 2.1147,
      "step": 108
    },
    {
      "epoch": 0.1356565028002489,
      "grad_norm": 2.818286895751953,
      "learning_rate": 1.970954356846473e-06,
      "loss": 2.0914,
      "step": 109
    },
    {
      "epoch": 0.13690105787181084,
      "grad_norm": 3.0579800605773926,
      "learning_rate": 1.991701244813278e-06,
      "loss": 2.0879,
      "step": 110
    },
    {
      "epoch": 0.13690105787181084,
      "eval_loss": 2.167081594467163,
      "eval_runtime": 42.3907,
      "eval_samples_per_second": 23.59,
      "eval_steps_per_second": 0.991,
      "step": 110
    },
    {
      "epoch": 0.13814561294337274,
      "grad_norm": 2.6039772033691406,
      "learning_rate": 2.012448132780083e-06,
      "loss": 2.1006,
      "step": 111
    },
    {
      "epoch": 0.13939016801493467,
      "grad_norm": 3.1309447288513184,
      "learning_rate": 2.0331950207468883e-06,
      "loss": 2.0758,
      "step": 112
    },
    {
      "epoch": 0.14063472308649658,
      "grad_norm": 2.6772613525390625,
      "learning_rate": 2.053941908713693e-06,
      "loss": 2.1542,
      "step": 113
    },
    {
      "epoch": 0.14187927815805848,
      "grad_norm": 3.2689080238342285,
      "learning_rate": 2.074688796680498e-06,
      "loss": 2.0873,
      "step": 114
    },
    {
      "epoch": 0.1431238332296204,
      "grad_norm": 2.7289857864379883,
      "learning_rate": 2.095435684647303e-06,
      "loss": 2.0697,
      "step": 115
    },
    {
      "epoch": 0.14436838830118232,
      "grad_norm": 2.8635787963867188,
      "learning_rate": 2.116182572614108e-06,
      "loss": 2.0908,
      "step": 116
    },
    {
      "epoch": 0.14561294337274425,
      "grad_norm": 2.805933713912964,
      "learning_rate": 2.136929460580913e-06,
      "loss": 2.0643,
      "step": 117
    },
    {
      "epoch": 0.14685749844430615,
      "grad_norm": 2.562567710876465,
      "learning_rate": 2.157676348547718e-06,
      "loss": 2.0559,
      "step": 118
    },
    {
      "epoch": 0.14810205351586808,
      "grad_norm": 3.000135898590088,
      "learning_rate": 2.178423236514523e-06,
      "loss": 2.0442,
      "step": 119
    },
    {
      "epoch": 0.14934660858742999,
      "grad_norm": 2.6913654804229736,
      "learning_rate": 2.199170124481328e-06,
      "loss": 2.0551,
      "step": 120
    },
    {
      "epoch": 0.14934660858742999,
      "eval_loss": 2.130500316619873,
      "eval_runtime": 44.3089,
      "eval_samples_per_second": 22.569,
      "eval_steps_per_second": 0.948,
      "step": 120
    },
    {
      "epoch": 0.15059116365899192,
      "grad_norm": 3.38159441947937,
      "learning_rate": 2.219917012448133e-06,
      "loss": 2.0359,
      "step": 121
    },
    {
      "epoch": 0.15183571873055382,
      "grad_norm": 2.4632813930511475,
      "learning_rate": 2.240663900414938e-06,
      "loss": 2.0805,
      "step": 122
    },
    {
      "epoch": 0.15308027380211575,
      "grad_norm": 3.2085115909576416,
      "learning_rate": 2.2614107883817427e-06,
      "loss": 2.0617,
      "step": 123
    },
    {
      "epoch": 0.15432482887367766,
      "grad_norm": 2.7810094356536865,
      "learning_rate": 2.282157676348548e-06,
      "loss": 2.0467,
      "step": 124
    },
    {
      "epoch": 0.1555693839452396,
      "grad_norm": 2.6023035049438477,
      "learning_rate": 2.302904564315353e-06,
      "loss": 2.0578,
      "step": 125
    },
    {
      "epoch": 0.1568139390168015,
      "grad_norm": 3.073814630508423,
      "learning_rate": 2.323651452282158e-06,
      "loss": 2.0742,
      "step": 126
    },
    {
      "epoch": 0.15805849408836342,
      "grad_norm": 2.622281312942505,
      "learning_rate": 2.3443983402489627e-06,
      "loss": 2.1668,
      "step": 127
    },
    {
      "epoch": 0.15930304915992533,
      "grad_norm": 3.3766582012176514,
      "learning_rate": 2.365145228215768e-06,
      "loss": 2.0129,
      "step": 128
    },
    {
      "epoch": 0.16054760423148726,
      "grad_norm": 3.1513423919677734,
      "learning_rate": 2.385892116182573e-06,
      "loss": 2.0215,
      "step": 129
    },
    {
      "epoch": 0.16179215930304916,
      "grad_norm": 3.2852959632873535,
      "learning_rate": 2.4066390041493776e-06,
      "loss": 2.0557,
      "step": 130
    },
    {
      "epoch": 0.16179215930304916,
      "eval_loss": 2.096493721008301,
      "eval_runtime": 46.9447,
      "eval_samples_per_second": 21.302,
      "eval_steps_per_second": 0.895,
      "step": 130
    },
    {
      "epoch": 0.16303671437461106,
      "grad_norm": 2.526399612426758,
      "learning_rate": 2.4273858921161828e-06,
      "loss": 2.065,
      "step": 131
    },
    {
      "epoch": 0.164281269446173,
      "grad_norm": 3.3448667526245117,
      "learning_rate": 2.448132780082988e-06,
      "loss": 2.0472,
      "step": 132
    },
    {
      "epoch": 0.1655258245177349,
      "grad_norm": 2.6260809898376465,
      "learning_rate": 2.468879668049793e-06,
      "loss": 1.9804,
      "step": 133
    },
    {
      "epoch": 0.16677037958929683,
      "grad_norm": 3.6141812801361084,
      "learning_rate": 2.4896265560165977e-06,
      "loss": 2.0197,
      "step": 134
    },
    {
      "epoch": 0.16801493466085873,
      "grad_norm": 2.4911234378814697,
      "learning_rate": 2.5103734439834028e-06,
      "loss": 1.9642,
      "step": 135
    },
    {
      "epoch": 0.16925948973242066,
      "grad_norm": 2.928642749786377,
      "learning_rate": 2.5311203319502074e-06,
      "loss": 1.9489,
      "step": 136
    },
    {
      "epoch": 0.17050404480398257,
      "grad_norm": 3.090965509414673,
      "learning_rate": 2.5518672199170125e-06,
      "loss": 1.994,
      "step": 137
    },
    {
      "epoch": 0.1717485998755445,
      "grad_norm": 3.2897465229034424,
      "learning_rate": 2.5726141078838172e-06,
      "loss": 1.9746,
      "step": 138
    },
    {
      "epoch": 0.1729931549471064,
      "grad_norm": 2.857083320617676,
      "learning_rate": 2.5933609958506228e-06,
      "loss": 1.9774,
      "step": 139
    },
    {
      "epoch": 0.17423771001866833,
      "grad_norm": 3.803220510482788,
      "learning_rate": 2.614107883817428e-06,
      "loss": 1.9731,
      "step": 140
    },
    {
      "epoch": 0.17423771001866833,
      "eval_loss": 2.064887046813965,
      "eval_runtime": 44.0231,
      "eval_samples_per_second": 22.715,
      "eval_steps_per_second": 0.954,
      "step": 140
    },
    {
      "epoch": 0.17548226509023024,
      "grad_norm": 2.9923534393310547,
      "learning_rate": 2.6348547717842326e-06,
      "loss": 1.9854,
      "step": 141
    },
    {
      "epoch": 0.17672682016179217,
      "grad_norm": 3.3368566036224365,
      "learning_rate": 2.6556016597510377e-06,
      "loss": 2.0687,
      "step": 142
    },
    {
      "epoch": 0.17797137523335407,
      "grad_norm": 3.3132379055023193,
      "learning_rate": 2.6763485477178423e-06,
      "loss": 1.9696,
      "step": 143
    },
    {
      "epoch": 0.179215930304916,
      "grad_norm": 3.7914819717407227,
      "learning_rate": 2.6970954356846475e-06,
      "loss": 1.9926,
      "step": 144
    },
    {
      "epoch": 0.1804604853764779,
      "grad_norm": 3.20161509513855,
      "learning_rate": 2.717842323651452e-06,
      "loss": 1.9193,
      "step": 145
    },
    {
      "epoch": 0.18170504044803984,
      "grad_norm": 3.440420150756836,
      "learning_rate": 2.7385892116182577e-06,
      "loss": 1.991,
      "step": 146
    },
    {
      "epoch": 0.18294959551960174,
      "grad_norm": 3.152684450149536,
      "learning_rate": 2.7593360995850628e-06,
      "loss": 1.9695,
      "step": 147
    },
    {
      "epoch": 0.18419415059116365,
      "grad_norm": 3.0402464866638184,
      "learning_rate": 2.7800829875518675e-06,
      "loss": 1.9274,
      "step": 148
    },
    {
      "epoch": 0.18543870566272558,
      "grad_norm": 2.778444290161133,
      "learning_rate": 2.8008298755186726e-06,
      "loss": 1.9198,
      "step": 149
    },
    {
      "epoch": 0.18668326073428748,
      "grad_norm": 2.7946548461914062,
      "learning_rate": 2.8215767634854773e-06,
      "loss": 1.8958,
      "step": 150
    },
    {
      "epoch": 0.18668326073428748,
      "eval_loss": 2.043288230895996,
      "eval_runtime": 44.2268,
      "eval_samples_per_second": 22.611,
      "eval_steps_per_second": 0.95,
      "step": 150
    },
    {
      "epoch": 0.1879278158058494,
      "grad_norm": 3.269259452819824,
      "learning_rate": 2.8423236514522824e-06,
      "loss": 1.9832,
      "step": 151
    },
    {
      "epoch": 0.18917237087741132,
      "grad_norm": 3.2673771381378174,
      "learning_rate": 2.863070539419087e-06,
      "loss": 1.9345,
      "step": 152
    },
    {
      "epoch": 0.19041692594897325,
      "grad_norm": 3.016599655151367,
      "learning_rate": 2.883817427385892e-06,
      "loss": 1.8969,
      "step": 153
    },
    {
      "epoch": 0.19166148102053515,
      "grad_norm": 3.2771544456481934,
      "learning_rate": 2.9045643153526977e-06,
      "loss": 1.921,
      "step": 154
    },
    {
      "epoch": 0.19290603609209708,
      "grad_norm": 3.008080244064331,
      "learning_rate": 2.9253112033195024e-06,
      "loss": 1.9367,
      "step": 155
    },
    {
      "epoch": 0.194150591163659,
      "grad_norm": 3.2312428951263428,
      "learning_rate": 2.9460580912863075e-06,
      "loss": 1.9021,
      "step": 156
    },
    {
      "epoch": 0.19539514623522092,
      "grad_norm": 3.294121503829956,
      "learning_rate": 2.966804979253112e-06,
      "loss": 1.9216,
      "step": 157
    },
    {
      "epoch": 0.19663970130678282,
      "grad_norm": 2.771685838699341,
      "learning_rate": 2.9875518672199173e-06,
      "loss": 1.9435,
      "step": 158
    },
    {
      "epoch": 0.19788425637834475,
      "grad_norm": 2.971971273422241,
      "learning_rate": 3.008298755186722e-06,
      "loss": 1.8851,
      "step": 159
    },
    {
      "epoch": 0.19912881144990666,
      "grad_norm": 3.3144047260284424,
      "learning_rate": 3.029045643153527e-06,
      "loss": 1.853,
      "step": 160
    },
    {
      "epoch": 0.19912881144990666,
      "eval_loss": 2.023491859436035,
      "eval_runtime": 44.9425,
      "eval_samples_per_second": 22.251,
      "eval_steps_per_second": 0.935,
      "step": 160
    },
    {
      "epoch": 0.2003733665214686,
      "grad_norm": 3.3733646869659424,
      "learning_rate": 3.0497925311203326e-06,
      "loss": 1.8836,
      "step": 161
    },
    {
      "epoch": 0.2016179215930305,
      "grad_norm": 3.1801207065582275,
      "learning_rate": 3.0705394190871373e-06,
      "loss": 1.9438,
      "step": 162
    },
    {
      "epoch": 0.2028624766645924,
      "grad_norm": 3.1199593544006348,
      "learning_rate": 3.0912863070539424e-06,
      "loss": 1.9219,
      "step": 163
    },
    {
      "epoch": 0.20410703173615433,
      "grad_norm": 3.344089984893799,
      "learning_rate": 3.112033195020747e-06,
      "loss": 1.9174,
      "step": 164
    },
    {
      "epoch": 0.20535158680771623,
      "grad_norm": 3.269702911376953,
      "learning_rate": 3.132780082987552e-06,
      "loss": 1.894,
      "step": 165
    },
    {
      "epoch": 0.20659614187927816,
      "grad_norm": 3.021744966506958,
      "learning_rate": 3.153526970954357e-06,
      "loss": 1.8799,
      "step": 166
    },
    {
      "epoch": 0.20784069695084006,
      "grad_norm": 3.0104167461395264,
      "learning_rate": 3.174273858921162e-06,
      "loss": 1.9116,
      "step": 167
    },
    {
      "epoch": 0.209085252022402,
      "grad_norm": 3.3018341064453125,
      "learning_rate": 3.1950207468879666e-06,
      "loss": 1.8389,
      "step": 168
    },
    {
      "epoch": 0.2103298070939639,
      "grad_norm": 3.0919857025146484,
      "learning_rate": 3.215767634854772e-06,
      "loss": 1.9522,
      "step": 169
    },
    {
      "epoch": 0.21157436216552583,
      "grad_norm": 3.4702494144439697,
      "learning_rate": 3.2365145228215773e-06,
      "loss": 1.9204,
      "step": 170
    },
    {
      "epoch": 0.21157436216552583,
      "eval_loss": 1.997478723526001,
      "eval_runtime": 43.2143,
      "eval_samples_per_second": 23.14,
      "eval_steps_per_second": 0.972,
      "step": 170
    },
    {
      "epoch": 0.21281891723708773,
      "grad_norm": 3.133046865463257,
      "learning_rate": 3.257261410788382e-06,
      "loss": 1.891,
      "step": 171
    },
    {
      "epoch": 0.21406347230864967,
      "grad_norm": 3.1828863620758057,
      "learning_rate": 3.278008298755187e-06,
      "loss": 1.8816,
      "step": 172
    },
    {
      "epoch": 0.21530802738021157,
      "grad_norm": 3.374898910522461,
      "learning_rate": 3.2987551867219918e-06,
      "loss": 1.8886,
      "step": 173
    },
    {
      "epoch": 0.2165525824517735,
      "grad_norm": 3.6660516262054443,
      "learning_rate": 3.319502074688797e-06,
      "loss": 1.9442,
      "step": 174
    },
    {
      "epoch": 0.2177971375233354,
      "grad_norm": 3.080636978149414,
      "learning_rate": 3.3402489626556016e-06,
      "loss": 1.866,
      "step": 175
    },
    {
      "epoch": 0.21904169259489734,
      "grad_norm": 3.3684473037719727,
      "learning_rate": 3.360995850622407e-06,
      "loss": 1.8699,
      "step": 176
    },
    {
      "epoch": 0.22028624766645924,
      "grad_norm": 3.818382740020752,
      "learning_rate": 3.381742738589212e-06,
      "loss": 1.8541,
      "step": 177
    },
    {
      "epoch": 0.22153080273802117,
      "grad_norm": 3.0995497703552246,
      "learning_rate": 3.402489626556017e-06,
      "loss": 1.8453,
      "step": 178
    },
    {
      "epoch": 0.22277535780958307,
      "grad_norm": 3.5696587562561035,
      "learning_rate": 3.423236514522822e-06,
      "loss": 1.9289,
      "step": 179
    },
    {
      "epoch": 0.22401991288114498,
      "grad_norm": 3.480571985244751,
      "learning_rate": 3.4439834024896267e-06,
      "loss": 1.8175,
      "step": 180
    },
    {
      "epoch": 0.22401991288114498,
      "eval_loss": 1.9842805862426758,
      "eval_runtime": 43.7649,
      "eval_samples_per_second": 22.849,
      "eval_steps_per_second": 0.96,
      "step": 180
    },
    {
      "epoch": 0.2252644679527069,
      "grad_norm": 3.229316473007202,
      "learning_rate": 3.4647302904564318e-06,
      "loss": 1.9048,
      "step": 181
    },
    {
      "epoch": 0.2265090230242688,
      "grad_norm": 3.1970808506011963,
      "learning_rate": 3.4854771784232365e-06,
      "loss": 1.907,
      "step": 182
    },
    {
      "epoch": 0.22775357809583074,
      "grad_norm": 3.5275299549102783,
      "learning_rate": 3.5062240663900416e-06,
      "loss": 1.8147,
      "step": 183
    },
    {
      "epoch": 0.22899813316739265,
      "grad_norm": 3.367077589035034,
      "learning_rate": 3.526970954356847e-06,
      "loss": 1.8189,
      "step": 184
    },
    {
      "epoch": 0.23024268823895458,
      "grad_norm": 3.3052914142608643,
      "learning_rate": 3.5477178423236518e-06,
      "loss": 1.8362,
      "step": 185
    },
    {
      "epoch": 0.23148724331051648,
      "grad_norm": 3.4737954139709473,
      "learning_rate": 3.568464730290457e-06,
      "loss": 1.8747,
      "step": 186
    },
    {
      "epoch": 0.23273179838207841,
      "grad_norm": 2.851705312728882,
      "learning_rate": 3.5892116182572616e-06,
      "loss": 1.7835,
      "step": 187
    },
    {
      "epoch": 0.23397635345364032,
      "grad_norm": 3.615224599838257,
      "learning_rate": 3.6099585062240667e-06,
      "loss": 1.7855,
      "step": 188
    },
    {
      "epoch": 0.23522090852520225,
      "grad_norm": 2.7468442916870117,
      "learning_rate": 3.6307053941908714e-06,
      "loss": 1.843,
      "step": 189
    },
    {
      "epoch": 0.23646546359676415,
      "grad_norm": 3.3228273391723633,
      "learning_rate": 3.6514522821576765e-06,
      "loss": 1.8453,
      "step": 190
    },
    {
      "epoch": 0.23646546359676415,
      "eval_loss": 1.9668089151382446,
      "eval_runtime": 44.9228,
      "eval_samples_per_second": 22.26,
      "eval_steps_per_second": 0.935,
      "step": 190
    },
    {
      "epoch": 0.23771001866832608,
      "grad_norm": 2.7882697582244873,
      "learning_rate": 3.672199170124482e-06,
      "loss": 1.8625,
      "step": 191
    },
    {
      "epoch": 0.238954573739888,
      "grad_norm": 3.244044780731201,
      "learning_rate": 3.6929460580912867e-06,
      "loss": 1.8903,
      "step": 192
    },
    {
      "epoch": 0.24019912881144992,
      "grad_norm": 2.8844549655914307,
      "learning_rate": 3.713692946058092e-06,
      "loss": 1.8929,
      "step": 193
    },
    {
      "epoch": 0.24144368388301182,
      "grad_norm": 2.9207894802093506,
      "learning_rate": 3.7344398340248965e-06,
      "loss": 1.8458,
      "step": 194
    },
    {
      "epoch": 0.24268823895457373,
      "grad_norm": 2.908562183380127,
      "learning_rate": 3.7551867219917016e-06,
      "loss": 1.828,
      "step": 195
    },
    {
      "epoch": 0.24393279402613566,
      "grad_norm": 3.1195480823516846,
      "learning_rate": 3.7759336099585063e-06,
      "loss": 1.855,
      "step": 196
    },
    {
      "epoch": 0.24517734909769756,
      "grad_norm": 3.3910183906555176,
      "learning_rate": 3.7966804979253114e-06,
      "loss": 1.7868,
      "step": 197
    },
    {
      "epoch": 0.2464219041692595,
      "grad_norm": 2.921475887298584,
      "learning_rate": 3.817427385892116e-06,
      "loss": 1.7505,
      "step": 198
    },
    {
      "epoch": 0.2476664592408214,
      "grad_norm": 3.5471713542938232,
      "learning_rate": 3.838174273858922e-06,
      "loss": 1.8445,
      "step": 199
    },
    {
      "epoch": 0.24891101431238333,
      "grad_norm": 3.047698736190796,
      "learning_rate": 3.858921161825726e-06,
      "loss": 1.8037,
      "step": 200
    },
    {
      "epoch": 0.24891101431238333,
      "eval_loss": 1.9400665760040283,
      "eval_runtime": 49.1044,
      "eval_samples_per_second": 20.365,
      "eval_steps_per_second": 0.855,
      "step": 200
    },
    {
      "epoch": 0.25015556938394523,
      "grad_norm": 3.4988913536071777,
      "learning_rate": 3.879668049792531e-06,
      "loss": 1.8061,
      "step": 201
    },
    {
      "epoch": 0.25140012445550713,
      "grad_norm": 3.0373125076293945,
      "learning_rate": 3.9004149377593365e-06,
      "loss": 1.8081,
      "step": 202
    },
    {
      "epoch": 0.2526446795270691,
      "grad_norm": 2.9764745235443115,
      "learning_rate": 3.921161825726142e-06,
      "loss": 1.8244,
      "step": 203
    },
    {
      "epoch": 0.253889234598631,
      "grad_norm": 3.4413554668426514,
      "learning_rate": 3.941908713692946e-06,
      "loss": 1.8492,
      "step": 204
    },
    {
      "epoch": 0.2551337896701929,
      "grad_norm": 2.9980437755584717,
      "learning_rate": 3.962655601659751e-06,
      "loss": 1.805,
      "step": 205
    },
    {
      "epoch": 0.2563783447417548,
      "grad_norm": 3.228750228881836,
      "learning_rate": 3.983402489626556e-06,
      "loss": 1.7683,
      "step": 206
    },
    {
      "epoch": 0.25762289981331676,
      "grad_norm": 3.008496046066284,
      "learning_rate": 4.004149377593361e-06,
      "loss": 1.8602,
      "step": 207
    },
    {
      "epoch": 0.25886745488487867,
      "grad_norm": 3.338935375213623,
      "learning_rate": 4.024896265560166e-06,
      "loss": 1.7978,
      "step": 208
    },
    {
      "epoch": 0.26011200995644057,
      "grad_norm": 3.1450204849243164,
      "learning_rate": 4.045643153526971e-06,
      "loss": 1.8243,
      "step": 209
    },
    {
      "epoch": 0.2613565650280025,
      "grad_norm": 3.5714142322540283,
      "learning_rate": 4.0663900414937765e-06,
      "loss": 1.8386,
      "step": 210
    },
    {
      "epoch": 0.2613565650280025,
      "eval_loss": 1.9310717582702637,
      "eval_runtime": 42.4826,
      "eval_samples_per_second": 23.539,
      "eval_steps_per_second": 0.989,
      "step": 210
    },
    {
      "epoch": 0.26260112009956443,
      "grad_norm": 3.3498189449310303,
      "learning_rate": 4.087136929460581e-06,
      "loss": 1.8055,
      "step": 211
    },
    {
      "epoch": 0.26384567517112634,
      "grad_norm": 3.2423200607299805,
      "learning_rate": 4.107883817427386e-06,
      "loss": 1.8378,
      "step": 212
    },
    {
      "epoch": 0.26509023024268824,
      "grad_norm": 3.0533974170684814,
      "learning_rate": 4.128630705394191e-06,
      "loss": 1.7699,
      "step": 213
    },
    {
      "epoch": 0.26633478531425014,
      "grad_norm": 3.135615587234497,
      "learning_rate": 4.149377593360996e-06,
      "loss": 1.8019,
      "step": 214
    },
    {
      "epoch": 0.26757934038581205,
      "grad_norm": 2.9863040447235107,
      "learning_rate": 4.170124481327801e-06,
      "loss": 1.7817,
      "step": 215
    },
    {
      "epoch": 0.268823895457374,
      "grad_norm": 3.270598888397217,
      "learning_rate": 4.190871369294606e-06,
      "loss": 1.7542,
      "step": 216
    },
    {
      "epoch": 0.2700684505289359,
      "grad_norm": 3.2279393672943115,
      "learning_rate": 4.211618257261411e-06,
      "loss": 1.7964,
      "step": 217
    },
    {
      "epoch": 0.2713130056004978,
      "grad_norm": 3.0626120567321777,
      "learning_rate": 4.232365145228216e-06,
      "loss": 1.7687,
      "step": 218
    },
    {
      "epoch": 0.2725575606720597,
      "grad_norm": 3.070702075958252,
      "learning_rate": 4.253112033195021e-06,
      "loss": 1.8075,
      "step": 219
    },
    {
      "epoch": 0.2738021157436217,
      "grad_norm": 3.3623268604278564,
      "learning_rate": 4.273858921161826e-06,
      "loss": 1.8517,
      "step": 220
    },
    {
      "epoch": 0.2738021157436217,
      "eval_loss": 1.9212934970855713,
      "eval_runtime": 46.8503,
      "eval_samples_per_second": 21.345,
      "eval_steps_per_second": 0.896,
      "step": 220
    },
    {
      "epoch": 0.2750466708151836,
      "grad_norm": 3.005740165710449,
      "learning_rate": 4.294605809128631e-06,
      "loss": 1.816,
      "step": 221
    },
    {
      "epoch": 0.2762912258867455,
      "grad_norm": 3.5257251262664795,
      "learning_rate": 4.315352697095436e-06,
      "loss": 1.8191,
      "step": 222
    },
    {
      "epoch": 0.2775357809583074,
      "grad_norm": 3.361103057861328,
      "learning_rate": 4.336099585062241e-06,
      "loss": 1.8378,
      "step": 223
    },
    {
      "epoch": 0.27878033602986935,
      "grad_norm": 3.787623405456543,
      "learning_rate": 4.356846473029046e-06,
      "loss": 1.7736,
      "step": 224
    },
    {
      "epoch": 0.28002489110143125,
      "grad_norm": 3.45717453956604,
      "learning_rate": 4.3775933609958506e-06,
      "loss": 1.8195,
      "step": 225
    },
    {
      "epoch": 0.28126944617299315,
      "grad_norm": 3.1698601245880127,
      "learning_rate": 4.398340248962656e-06,
      "loss": 1.7756,
      "step": 226
    },
    {
      "epoch": 0.28251400124455506,
      "grad_norm": 3.345258951187134,
      "learning_rate": 4.419087136929461e-06,
      "loss": 1.7272,
      "step": 227
    },
    {
      "epoch": 0.28375855631611696,
      "grad_norm": 2.905773639678955,
      "learning_rate": 4.439834024896266e-06,
      "loss": 1.84,
      "step": 228
    },
    {
      "epoch": 0.2850031113876789,
      "grad_norm": 3.2004425525665283,
      "learning_rate": 4.460580912863071e-06,
      "loss": 1.8164,
      "step": 229
    },
    {
      "epoch": 0.2862476664592408,
      "grad_norm": 2.965275287628174,
      "learning_rate": 4.481327800829876e-06,
      "loss": 1.7718,
      "step": 230
    },
    {
      "epoch": 0.2862476664592408,
      "eval_loss": 1.903228521347046,
      "eval_runtime": 51.1663,
      "eval_samples_per_second": 19.544,
      "eval_steps_per_second": 0.821,
      "step": 230
    },
    {
      "epoch": 0.2874922215308027,
      "grad_norm": 2.8130710124969482,
      "learning_rate": 4.502074688796681e-06,
      "loss": 1.7696,
      "step": 231
    },
    {
      "epoch": 0.28873677660236463,
      "grad_norm": 2.9987664222717285,
      "learning_rate": 4.5228215767634855e-06,
      "loss": 1.775,
      "step": 232
    },
    {
      "epoch": 0.2899813316739266,
      "grad_norm": 3.0936989784240723,
      "learning_rate": 4.543568464730291e-06,
      "loss": 1.7775,
      "step": 233
    },
    {
      "epoch": 0.2912258867454885,
      "grad_norm": 3.0936989784240723,
      "learning_rate": 4.543568464730291e-06,
      "loss": 1.7154,
      "step": 234
    },
    {
      "epoch": 0.2924704418170504,
      "grad_norm": 2.8632843494415283,
      "learning_rate": 4.564315352697096e-06,
      "loss": 1.7771,
      "step": 235
    },
    {
      "epoch": 0.2937149968886123,
      "grad_norm": 3.3314013481140137,
      "learning_rate": 4.585062240663901e-06,
      "loss": 1.7243,
      "step": 236
    },
    {
      "epoch": 0.29495955196017426,
      "grad_norm": 2.8522872924804688,
      "learning_rate": 4.605809128630706e-06,
      "loss": 1.7543,
      "step": 237
    },
    {
      "epoch": 0.29620410703173616,
      "grad_norm": 3.319157600402832,
      "learning_rate": 4.626556016597511e-06,
      "loss": 1.7652,
      "step": 238
    },
    {
      "epoch": 0.29744866210329807,
      "grad_norm": 3.123116970062256,
      "learning_rate": 4.647302904564316e-06,
      "loss": 1.6933,
      "step": 239
    },
    {
      "epoch": 0.29869321717485997,
      "grad_norm": 3.1735854148864746,
      "learning_rate": 4.66804979253112e-06,
      "loss": 1.7437,
      "step": 240
    },
    {
      "epoch": 0.29869321717485997,
      "eval_loss": 1.891274333000183,
      "eval_runtime": 50.0054,
      "eval_samples_per_second": 19.998,
      "eval_steps_per_second": 0.84,
      "step": 240
    },
    {
      "epoch": 0.29993777224642193,
      "grad_norm": 3.0996642112731934,
      "learning_rate": 4.6887966804979255e-06,
      "loss": 1.7594,
      "step": 241
    },
    {
      "epoch": 0.30118232731798383,
      "grad_norm": 3.255063533782959,
      "learning_rate": 4.709543568464731e-06,
      "loss": 1.7645,
      "step": 242
    },
    {
      "epoch": 0.30242688238954574,
      "grad_norm": 2.722285747528076,
      "learning_rate": 4.730290456431536e-06,
      "loss": 1.7549,
      "step": 243
    },
    {
      "epoch": 0.30367143746110764,
      "grad_norm": 3.1903417110443115,
      "learning_rate": 4.751037344398341e-06,
      "loss": 1.7382,
      "step": 244
    },
    {
      "epoch": 0.30491599253266954,
      "grad_norm": 2.8352959156036377,
      "learning_rate": 4.771784232365146e-06,
      "loss": 1.7497,
      "step": 245
    },
    {
      "epoch": 0.3061605476042315,
      "grad_norm": 3.158536434173584,
      "learning_rate": 4.792531120331951e-06,
      "loss": 1.7467,
      "step": 246
    },
    {
      "epoch": 0.3074051026757934,
      "grad_norm": 2.8456921577453613,
      "learning_rate": 4.813278008298755e-06,
      "loss": 1.7472,
      "step": 247
    },
    {
      "epoch": 0.3086496577473553,
      "grad_norm": 3.1215128898620605,
      "learning_rate": 4.83402489626556e-06,
      "loss": 1.7577,
      "step": 248
    },
    {
      "epoch": 0.3098942128189172,
      "grad_norm": 2.928015947341919,
      "learning_rate": 4.8547717842323655e-06,
      "loss": 1.6932,
      "step": 249
    },
    {
      "epoch": 0.3111387678904792,
      "grad_norm": 3.001044511795044,
      "learning_rate": 4.875518672199171e-06,
      "loss": 1.7276,
      "step": 250
    },
    {
      "epoch": 0.3111387678904792,
      "eval_loss": 1.8855507373809814,
      "eval_runtime": 46.0143,
      "eval_samples_per_second": 21.732,
      "eval_steps_per_second": 0.913,
      "step": 250
    },
    {
      "epoch": 0.3123833229620411,
      "grad_norm": 2.988994598388672,
      "learning_rate": 4.896265560165976e-06,
      "loss": 1.739,
      "step": 251
    },
    {
      "epoch": 0.313627878033603,
      "grad_norm": 3.0091233253479004,
      "learning_rate": 4.91701244813278e-06,
      "loss": 1.7118,
      "step": 252
    },
    {
      "epoch": 0.3148724331051649,
      "grad_norm": 3.2735323905944824,
      "learning_rate": 4.937759336099586e-06,
      "loss": 1.7832,
      "step": 253
    },
    {
      "epoch": 0.31611698817672684,
      "grad_norm": 3.055468797683716,
      "learning_rate": 4.95850622406639e-06,
      "loss": 1.7151,
      "step": 254
    },
    {
      "epoch": 0.31736154324828875,
      "grad_norm": 3.0889883041381836,
      "learning_rate": 4.979253112033195e-06,
      "loss": 1.7411,
      "step": 255
    },
    {
      "epoch": 0.31860609831985065,
      "grad_norm": 2.8418147563934326,
      "learning_rate": 5e-06,
      "loss": 1.7575,
      "step": 256
    },
    {
      "epoch": 0.31985065339141255,
      "grad_norm": 3.09739351272583,
      "learning_rate": 4.99769372693727e-06,
      "loss": 1.7988,
      "step": 257
    },
    {
      "epoch": 0.3210952084629745,
      "grad_norm": 3.245515823364258,
      "learning_rate": 4.995387453874539e-06,
      "loss": 1.7706,
      "step": 258
    },
    {
      "epoch": 0.3223397635345364,
      "grad_norm": 3.250432014465332,
      "learning_rate": 4.993081180811809e-06,
      "loss": 1.7839,
      "step": 259
    },
    {
      "epoch": 0.3235843186060983,
      "grad_norm": 3.1325368881225586,
      "learning_rate": 4.990774907749078e-06,
      "loss": 1.754,
      "step": 260
    },
    {
      "epoch": 0.3235843186060983,
      "eval_loss": 1.869175672531128,
      "eval_runtime": 49.0199,
      "eval_samples_per_second": 20.4,
      "eval_steps_per_second": 0.857,
      "step": 260
    },
    {
      "epoch": 0.3248288736776602,
      "grad_norm": 3.1627941131591797,
      "learning_rate": 4.988468634686347e-06,
      "loss": 1.7123,
      "step": 261
    },
    {
      "epoch": 0.3260734287492221,
      "grad_norm": 2.848202705383301,
      "learning_rate": 4.986162361623617e-06,
      "loss": 1.7231,
      "step": 262
    },
    {
      "epoch": 0.3273179838207841,
      "grad_norm": 2.9585494995117188,
      "learning_rate": 4.983856088560886e-06,
      "loss": 1.7301,
      "step": 263
    },
    {
      "epoch": 0.328562538892346,
      "grad_norm": 3.160170316696167,
      "learning_rate": 4.981549815498156e-06,
      "loss": 1.7283,
      "step": 264
    },
    {
      "epoch": 0.3298070939639079,
      "grad_norm": 2.6879520416259766,
      "learning_rate": 4.979243542435424e-06,
      "loss": 1.7494,
      "step": 265
    },
    {
      "epoch": 0.3310516490354698,
      "grad_norm": 3.1313576698303223,
      "learning_rate": 4.976937269372694e-06,
      "loss": 1.7224,
      "step": 266
    },
    {
      "epoch": 0.33229620410703176,
      "grad_norm": 3.0477867126464844,
      "learning_rate": 4.974630996309964e-06,
      "loss": 1.6951,
      "step": 267
    },
    {
      "epoch": 0.33354075917859366,
      "grad_norm": 3.133974313735962,
      "learning_rate": 4.972324723247233e-06,
      "loss": 1.7362,
      "step": 268
    },
    {
      "epoch": 0.33478531425015556,
      "grad_norm": 3.4925222396850586,
      "learning_rate": 4.970018450184502e-06,
      "loss": 1.7053,
      "step": 269
    },
    {
      "epoch": 0.33602986932171747,
      "grad_norm": 2.848466634750366,
      "learning_rate": 4.9677121771217715e-06,
      "loss": 1.7336,
      "step": 270
    },
    {
      "epoch": 0.33602986932171747,
      "eval_loss": 1.8723454475402832,
      "eval_runtime": 47.9758,
      "eval_samples_per_second": 20.844,
      "eval_steps_per_second": 0.875,
      "step": 270
    },
    {
      "epoch": 0.3372744243932794,
      "grad_norm": 3.405003309249878,
      "learning_rate": 4.965405904059041e-06,
      "loss": 1.7178,
      "step": 271
    },
    {
      "epoch": 0.33851897946484133,
      "grad_norm": 2.982916831970215,
      "learning_rate": 4.96309963099631e-06,
      "loss": 1.7255,
      "step": 272
    },
    {
      "epoch": 0.33976353453640323,
      "grad_norm": 3.1712088584899902,
      "learning_rate": 4.96079335793358e-06,
      "loss": 1.7328,
      "step": 273
    },
    {
      "epoch": 0.34100808960796514,
      "grad_norm": 3.325481414794922,
      "learning_rate": 4.958487084870849e-06,
      "loss": 1.7268,
      "step": 274
    },
    {
      "epoch": 0.3422526446795271,
      "grad_norm": 3.0751800537109375,
      "learning_rate": 4.956180811808119e-06,
      "loss": 1.7916,
      "step": 275
    },
    {
      "epoch": 0.343497199751089,
      "grad_norm": 2.9451465606689453,
      "learning_rate": 4.953874538745388e-06,
      "loss": 1.7588,
      "step": 276
    },
    {
      "epoch": 0.3447417548226509,
      "grad_norm": 3.1492040157318115,
      "learning_rate": 4.9515682656826574e-06,
      "loss": 1.7312,
      "step": 277
    },
    {
      "epoch": 0.3459863098942128,
      "grad_norm": 3.165748119354248,
      "learning_rate": 4.949261992619927e-06,
      "loss": 1.7645,
      "step": 278
    },
    {
      "epoch": 0.3472308649657747,
      "grad_norm": 3.059307336807251,
      "learning_rate": 4.946955719557196e-06,
      "loss": 1.6783,
      "step": 279
    },
    {
      "epoch": 0.34847542003733667,
      "grad_norm": 2.940147638320923,
      "learning_rate": 4.944649446494466e-06,
      "loss": 1.7079,
      "step": 280
    },
    {
      "epoch": 0.34847542003733667,
      "eval_loss": 1.8577951192855835,
      "eval_runtime": 50.1741,
      "eval_samples_per_second": 19.931,
      "eval_steps_per_second": 0.837,
      "step": 280
    },
    {
      "epoch": 0.3497199751088986,
      "grad_norm": 2.6892037391662598,
      "learning_rate": 4.942343173431734e-06,
      "loss": 1.7104,
      "step": 281
    },
    {
      "epoch": 0.3509645301804605,
      "grad_norm": 3.136791467666626,
      "learning_rate": 4.940036900369004e-06,
      "loss": 1.763,
      "step": 282
    },
    {
      "epoch": 0.3522090852520224,
      "grad_norm": 2.6873879432678223,
      "learning_rate": 4.937730627306274e-06,
      "loss": 1.722,
      "step": 283
    },
    {
      "epoch": 0.35345364032358434,
      "grad_norm": 3.1375961303710938,
      "learning_rate": 4.9354243542435426e-06,
      "loss": 1.7048,
      "step": 284
    },
    {
      "epoch": 0.35469819539514624,
      "grad_norm": 2.920725107192993,
      "learning_rate": 4.933118081180812e-06,
      "loss": 1.7014,
      "step": 285
    },
    {
      "epoch": 0.35594275046670815,
      "grad_norm": 2.750018835067749,
      "learning_rate": 4.930811808118081e-06,
      "loss": 1.6283,
      "step": 286
    },
    {
      "epoch": 0.35718730553827005,
      "grad_norm": 3.399036169052124,
      "learning_rate": 4.928505535055351e-06,
      "loss": 1.7265,
      "step": 287
    },
    {
      "epoch": 0.358431860609832,
      "grad_norm": 3.1734066009521484,
      "learning_rate": 4.92619926199262e-06,
      "loss": 1.7007,
      "step": 288
    },
    {
      "epoch": 0.3596764156813939,
      "grad_norm": 3.038865089416504,
      "learning_rate": 4.92389298892989e-06,
      "loss": 1.7357,
      "step": 289
    },
    {
      "epoch": 0.3609209707529558,
      "grad_norm": 3.0309486389160156,
      "learning_rate": 4.921586715867159e-06,
      "loss": 1.7487,
      "step": 290
    },
    {
      "epoch": 0.3609209707529558,
      "eval_loss": 1.8484960794448853,
      "eval_runtime": 50.374,
      "eval_samples_per_second": 19.852,
      "eval_steps_per_second": 0.834,
      "step": 290
    },
    {
      "epoch": 0.3621655258245177,
      "grad_norm": 2.867432117462158,
      "learning_rate": 4.9192804428044285e-06,
      "loss": 1.6988,
      "step": 291
    },
    {
      "epoch": 0.3634100808960797,
      "grad_norm": 3.31636905670166,
      "learning_rate": 4.916974169741698e-06,
      "loss": 1.7003,
      "step": 292
    },
    {
      "epoch": 0.3646546359676416,
      "grad_norm": 3.1181511878967285,
      "learning_rate": 4.914667896678967e-06,
      "loss": 1.6762,
      "step": 293
    },
    {
      "epoch": 0.3658991910392035,
      "grad_norm": 2.978194236755371,
      "learning_rate": 4.912361623616237e-06,
      "loss": 1.6435,
      "step": 294
    },
    {
      "epoch": 0.3671437461107654,
      "grad_norm": 3.0623116493225098,
      "learning_rate": 4.910055350553506e-06,
      "loss": 1.7038,
      "step": 295
    },
    {
      "epoch": 0.3683883011823273,
      "grad_norm": 2.83353853225708,
      "learning_rate": 4.907749077490776e-06,
      "loss": 1.7237,
      "step": 296
    },
    {
      "epoch": 0.36963285625388925,
      "grad_norm": 2.8587100505828857,
      "learning_rate": 4.905442804428044e-06,
      "loss": 1.6927,
      "step": 297
    },
    {
      "epoch": 0.37087741132545116,
      "grad_norm": 2.7930493354797363,
      "learning_rate": 4.903136531365314e-06,
      "loss": 1.7381,
      "step": 298
    },
    {
      "epoch": 0.37212196639701306,
      "grad_norm": 2.8992788791656494,
      "learning_rate": 4.900830258302584e-06,
      "loss": 1.6485,
      "step": 299
    },
    {
      "epoch": 0.37336652146857496,
      "grad_norm": 2.8896944522857666,
      "learning_rate": 4.898523985239853e-06,
      "loss": 1.7143,
      "step": 300
    },
    {
      "epoch": 0.37336652146857496,
      "eval_loss": 1.8331259489059448,
      "eval_runtime": 51.0214,
      "eval_samples_per_second": 19.6,
      "eval_steps_per_second": 0.823,
      "step": 300
    },
    {
      "epoch": 0.3746110765401369,
      "grad_norm": 2.885444164276123,
      "learning_rate": 4.896217712177122e-06,
      "loss": 1.6914,
      "step": 301
    },
    {
      "epoch": 0.3758556316116988,
      "grad_norm": 3.2996826171875,
      "learning_rate": 4.893911439114391e-06,
      "loss": 1.7353,
      "step": 302
    },
    {
      "epoch": 0.37710018668326073,
      "grad_norm": 2.7231531143188477,
      "learning_rate": 4.891605166051661e-06,
      "loss": 1.6837,
      "step": 303
    },
    {
      "epoch": 0.37834474175482263,
      "grad_norm": 2.9413955211639404,
      "learning_rate": 4.88929889298893e-06,
      "loss": 1.6448,
      "step": 304
    },
    {
      "epoch": 0.3795892968263846,
      "grad_norm": 2.6486589908599854,
      "learning_rate": 4.8869926199262e-06,
      "loss": 1.6826,
      "step": 305
    },
    {
      "epoch": 0.3808338518979465,
      "grad_norm": 2.977836847305298,
      "learning_rate": 4.884686346863469e-06,
      "loss": 1.6737,
      "step": 306
    },
    {
      "epoch": 0.3820784069695084,
      "grad_norm": 2.635324716567993,
      "learning_rate": 4.8823800738007384e-06,
      "loss": 1.7112,
      "step": 307
    },
    {
      "epoch": 0.3833229620410703,
      "grad_norm": 3.0650839805603027,
      "learning_rate": 4.880073800738008e-06,
      "loss": 1.7123,
      "step": 308
    },
    {
      "epoch": 0.3845675171126322,
      "grad_norm": 3.059629440307617,
      "learning_rate": 4.877767527675277e-06,
      "loss": 1.662,
      "step": 309
    },
    {
      "epoch": 0.38581207218419417,
      "grad_norm": 2.7559597492218018,
      "learning_rate": 4.875461254612546e-06,
      "loss": 1.7418,
      "step": 310
    },
    {
      "epoch": 0.38581207218419417,
      "eval_loss": 1.8277366161346436,
      "eval_runtime": 42.6671,
      "eval_samples_per_second": 23.437,
      "eval_steps_per_second": 0.984,
      "step": 310
    },
    {
      "epoch": 0.38705662725575607,
      "grad_norm": 2.7987024784088135,
      "learning_rate": 4.873154981549816e-06,
      "loss": 1.7326,
      "step": 311
    },
    {
      "epoch": 0.388301182327318,
      "grad_norm": 2.999824047088623,
      "learning_rate": 4.8708487084870856e-06,
      "loss": 1.7344,
      "step": 312
    },
    {
      "epoch": 0.3895457373988799,
      "grad_norm": 2.942995071411133,
      "learning_rate": 4.868542435424355e-06,
      "loss": 1.7055,
      "step": 313
    },
    {
      "epoch": 0.39079029247044184,
      "grad_norm": 2.823906183242798,
      "learning_rate": 4.8662361623616235e-06,
      "loss": 1.6819,
      "step": 314
    },
    {
      "epoch": 0.39203484754200374,
      "grad_norm": 2.7798759937286377,
      "learning_rate": 4.863929889298894e-06,
      "loss": 1.6694,
      "step": 315
    },
    {
      "epoch": 0.39327940261356564,
      "grad_norm": 2.599005699157715,
      "learning_rate": 4.861623616236163e-06,
      "loss": 1.6924,
      "step": 316
    },
    {
      "epoch": 0.39452395768512755,
      "grad_norm": 2.687539577484131,
      "learning_rate": 4.859317343173432e-06,
      "loss": 1.685,
      "step": 317
    },
    {
      "epoch": 0.3957685127566895,
      "grad_norm": 2.9679677486419678,
      "learning_rate": 4.857011070110701e-06,
      "loss": 1.718,
      "step": 318
    },
    {
      "epoch": 0.3970130678282514,
      "grad_norm": 2.8611817359924316,
      "learning_rate": 4.854704797047971e-06,
      "loss": 1.7247,
      "step": 319
    },
    {
      "epoch": 0.3982576228998133,
      "grad_norm": 2.7461249828338623,
      "learning_rate": 4.85239852398524e-06,
      "loss": 1.6434,
      "step": 320
    },
    {
      "epoch": 0.3982576228998133,
      "eval_loss": 1.8234667778015137,
      "eval_runtime": 44.3433,
      "eval_samples_per_second": 22.551,
      "eval_steps_per_second": 0.947,
      "step": 320
    },
    {
      "epoch": 0.3995021779713752,
      "grad_norm": 2.7142515182495117,
      "learning_rate": 4.8500922509225095e-06,
      "loss": 1.6855,
      "step": 321
    },
    {
      "epoch": 0.4007467330429372,
      "grad_norm": 2.87727689743042,
      "learning_rate": 4.847785977859779e-06,
      "loss": 1.705,
      "step": 322
    },
    {
      "epoch": 0.4019912881144991,
      "grad_norm": 2.9598982334136963,
      "learning_rate": 4.845479704797048e-06,
      "loss": 1.6482,
      "step": 323
    },
    {
      "epoch": 0.403235843186061,
      "grad_norm": 2.8654463291168213,
      "learning_rate": 4.843173431734318e-06,
      "loss": 1.5816,
      "step": 324
    },
    {
      "epoch": 0.4044803982576229,
      "grad_norm": 2.983837842941284,
      "learning_rate": 4.840867158671587e-06,
      "loss": 1.6962,
      "step": 325
    },
    {
      "epoch": 0.4057249533291848,
      "grad_norm": 2.730914354324341,
      "learning_rate": 4.838560885608857e-06,
      "loss": 1.6991,
      "step": 326
    },
    {
      "epoch": 0.40696950840074675,
      "grad_norm": 2.6901161670684814,
      "learning_rate": 4.836254612546126e-06,
      "loss": 1.6839,
      "step": 327
    },
    {
      "epoch": 0.40821406347230865,
      "grad_norm": 2.6661486625671387,
      "learning_rate": 4.8339483394833955e-06,
      "loss": 1.6894,
      "step": 328
    },
    {
      "epoch": 0.40945861854387056,
      "grad_norm": 2.64243221282959,
      "learning_rate": 4.831642066420665e-06,
      "loss": 1.6533,
      "step": 329
    },
    {
      "epoch": 0.41070317361543246,
      "grad_norm": 2.586725950241089,
      "learning_rate": 4.8293357933579335e-06,
      "loss": 1.6486,
      "step": 330
    },
    {
      "epoch": 0.41070317361543246,
      "eval_loss": 1.82283616065979,
      "eval_runtime": 46.5937,
      "eval_samples_per_second": 21.462,
      "eval_steps_per_second": 0.901,
      "step": 330
    },
    {
      "epoch": 0.4119477286869944,
      "grad_norm": 2.899968147277832,
      "learning_rate": 4.827029520295204e-06,
      "loss": 1.7136,
      "step": 331
    },
    {
      "epoch": 0.4131922837585563,
      "grad_norm": 3.067671537399292,
      "learning_rate": 4.824723247232473e-06,
      "loss": 1.672,
      "step": 332
    },
    {
      "epoch": 0.4144368388301182,
      "grad_norm": 2.8144404888153076,
      "learning_rate": 4.822416974169742e-06,
      "loss": 1.7173,
      "step": 333
    },
    {
      "epoch": 0.41568139390168013,
      "grad_norm": 3.0551373958587646,
      "learning_rate": 4.820110701107011e-06,
      "loss": 1.6843,
      "step": 334
    },
    {
      "epoch": 0.4169259489732421,
      "grad_norm": 2.7585153579711914,
      "learning_rate": 4.817804428044281e-06,
      "loss": 1.6834,
      "step": 335
    },
    {
      "epoch": 0.418170504044804,
      "grad_norm": 2.8691210746765137,
      "learning_rate": 4.81549815498155e-06,
      "loss": 1.6814,
      "step": 336
    },
    {
      "epoch": 0.4194150591163659,
      "grad_norm": 2.871523380279541,
      "learning_rate": 4.8131918819188194e-06,
      "loss": 1.6222,
      "step": 337
    },
    {
      "epoch": 0.4206596141879278,
      "grad_norm": 2.8928768634796143,
      "learning_rate": 4.810885608856089e-06,
      "loss": 1.6984,
      "step": 338
    },
    {
      "epoch": 0.42190416925948976,
      "grad_norm": 3.285256862640381,
      "learning_rate": 4.808579335793358e-06,
      "loss": 1.7029,
      "step": 339
    },
    {
      "epoch": 0.42314872433105166,
      "grad_norm": 2.785504102706909,
      "learning_rate": 4.806273062730628e-06,
      "loss": 1.6857,
      "step": 340
    },
    {
      "epoch": 0.42314872433105166,
      "eval_loss": 1.8053412437438965,
      "eval_runtime": 48.0656,
      "eval_samples_per_second": 20.805,
      "eval_steps_per_second": 0.874,
      "step": 340
    },
    {
      "epoch": 0.42439327940261357,
      "grad_norm": 3.0227086544036865,
      "learning_rate": 4.803966789667897e-06,
      "loss": 1.6903,
      "step": 341
    },
    {
      "epoch": 0.42563783447417547,
      "grad_norm": 2.616769790649414,
      "learning_rate": 4.8016605166051665e-06,
      "loss": 1.6144,
      "step": 342
    },
    {
      "epoch": 0.4268823895457374,
      "grad_norm": 2.9519753456115723,
      "learning_rate": 4.799354243542436e-06,
      "loss": 1.6157,
      "step": 343
    },
    {
      "epoch": 0.42812694461729933,
      "grad_norm": 2.764512062072754,
      "learning_rate": 4.797047970479705e-06,
      "loss": 1.6248,
      "step": 344
    },
    {
      "epoch": 0.42937149968886124,
      "grad_norm": 2.999178886413574,
      "learning_rate": 4.794741697416975e-06,
      "loss": 1.6764,
      "step": 345
    },
    {
      "epoch": 0.43061605476042314,
      "grad_norm": 3.0472185611724854,
      "learning_rate": 4.792435424354243e-06,
      "loss": 1.6421,
      "step": 346
    },
    {
      "epoch": 0.43186060983198504,
      "grad_norm": 2.6638400554656982,
      "learning_rate": 4.790129151291514e-06,
      "loss": 1.6304,
      "step": 347
    },
    {
      "epoch": 0.433105164903547,
      "grad_norm": 2.945295810699463,
      "learning_rate": 4.787822878228783e-06,
      "loss": 1.6181,
      "step": 348
    },
    {
      "epoch": 0.4343497199751089,
      "grad_norm": 2.915125846862793,
      "learning_rate": 4.7855166051660525e-06,
      "loss": 1.6789,
      "step": 349
    },
    {
      "epoch": 0.4355942750466708,
      "grad_norm": 2.5674026012420654,
      "learning_rate": 4.783210332103321e-06,
      "loss": 1.6723,
      "step": 350
    },
    {
      "epoch": 0.4355942750466708,
      "eval_loss": 1.8109745979309082,
      "eval_runtime": 43.9943,
      "eval_samples_per_second": 22.73,
      "eval_steps_per_second": 0.955,
      "step": 350
    },
    {
      "epoch": 0.4368388301182327,
      "grad_norm": 2.9533393383026123,
      "learning_rate": 4.7809040590405905e-06,
      "loss": 1.6271,
      "step": 351
    },
    {
      "epoch": 0.43808338518979467,
      "grad_norm": 2.7677664756774902,
      "learning_rate": 4.778597785977861e-06,
      "loss": 1.6633,
      "step": 352
    },
    {
      "epoch": 0.4393279402613566,
      "grad_norm": 2.527456521987915,
      "learning_rate": 4.776291512915129e-06,
      "loss": 1.5644,
      "step": 353
    },
    {
      "epoch": 0.4405724953329185,
      "grad_norm": 2.9778528213500977,
      "learning_rate": 4.773985239852399e-06,
      "loss": 1.6166,
      "step": 354
    },
    {
      "epoch": 0.4418170504044804,
      "grad_norm": 2.5554800033569336,
      "learning_rate": 4.771678966789668e-06,
      "loss": 1.5573,
      "step": 355
    },
    {
      "epoch": 0.44306160547604234,
      "grad_norm": 2.7415409088134766,
      "learning_rate": 4.769372693726938e-06,
      "loss": 1.6564,
      "step": 356
    },
    {
      "epoch": 0.44430616054760425,
      "grad_norm": 3.0275440216064453,
      "learning_rate": 4.767066420664207e-06,
      "loss": 1.67,
      "step": 357
    },
    {
      "epoch": 0.44555071561916615,
      "grad_norm": 2.720694065093994,
      "learning_rate": 4.7647601476014765e-06,
      "loss": 1.6374,
      "step": 358
    },
    {
      "epoch": 0.44679527069072805,
      "grad_norm": 2.97589111328125,
      "learning_rate": 4.762453874538746e-06,
      "loss": 1.6596,
      "step": 359
    },
    {
      "epoch": 0.44803982576228996,
      "grad_norm": 2.8005781173706055,
      "learning_rate": 4.760147601476015e-06,
      "loss": 1.6281,
      "step": 360
    },
    {
      "epoch": 0.44803982576228996,
      "eval_loss": 1.8139524459838867,
      "eval_runtime": 44.5817,
      "eval_samples_per_second": 22.431,
      "eval_steps_per_second": 0.942,
      "step": 360
    },
    {
      "epoch": 0.4492843808338519,
      "grad_norm": 3.3947463035583496,
      "learning_rate": 4.757841328413285e-06,
      "loss": 1.7149,
      "step": 361
    },
    {
      "epoch": 0.4505289359054138,
      "grad_norm": 2.991457939147949,
      "learning_rate": 4.755535055350554e-06,
      "loss": 1.5971,
      "step": 362
    },
    {
      "epoch": 0.4517734909769757,
      "grad_norm": 2.653313398361206,
      "learning_rate": 4.753228782287823e-06,
      "loss": 1.6465,
      "step": 363
    },
    {
      "epoch": 0.4530180460485376,
      "grad_norm": 2.716360569000244,
      "learning_rate": 4.750922509225093e-06,
      "loss": 1.6775,
      "step": 364
    },
    {
      "epoch": 0.4542626011200996,
      "grad_norm": 2.733598470687866,
      "learning_rate": 4.748616236162362e-06,
      "loss": 1.6087,
      "step": 365
    },
    {
      "epoch": 0.4555071561916615,
      "grad_norm": 2.581547737121582,
      "learning_rate": 4.746309963099631e-06,
      "loss": 1.6339,
      "step": 366
    },
    {
      "epoch": 0.4567517112632234,
      "grad_norm": 2.6841800212860107,
      "learning_rate": 4.7440036900369e-06,
      "loss": 1.6193,
      "step": 367
    },
    {
      "epoch": 0.4579962663347853,
      "grad_norm": 2.658369302749634,
      "learning_rate": 4.741697416974171e-06,
      "loss": 1.6287,
      "step": 368
    },
    {
      "epoch": 0.45924082140634725,
      "grad_norm": 2.7204179763793945,
      "learning_rate": 4.739391143911439e-06,
      "loss": 1.6755,
      "step": 369
    },
    {
      "epoch": 0.46048537647790916,
      "grad_norm": 2.6627941131591797,
      "learning_rate": 4.737084870848709e-06,
      "loss": 1.6142,
      "step": 370
    },
    {
      "epoch": 0.46048537647790916,
      "eval_loss": 1.8019787073135376,
      "eval_runtime": 42.7804,
      "eval_samples_per_second": 23.375,
      "eval_steps_per_second": 0.982,
      "step": 370
    },
    {
      "epoch": 0.46172993154947106,
      "grad_norm": 2.650705337524414,
      "learning_rate": 4.734778597785978e-06,
      "loss": 1.641,
      "step": 371
    },
    {
      "epoch": 0.46297448662103297,
      "grad_norm": 2.7031619548797607,
      "learning_rate": 4.7324723247232475e-06,
      "loss": 1.6596,
      "step": 372
    },
    {
      "epoch": 0.4642190416925949,
      "grad_norm": 3.3050339221954346,
      "learning_rate": 4.730166051660517e-06,
      "loss": 1.6664,
      "step": 373
    },
    {
      "epoch": 0.46546359676415683,
      "grad_norm": 2.841404676437378,
      "learning_rate": 4.727859778597786e-06,
      "loss": 1.6474,
      "step": 374
    },
    {
      "epoch": 0.46670815183571873,
      "grad_norm": 3.0990259647369385,
      "learning_rate": 4.725553505535056e-06,
      "loss": 1.6008,
      "step": 375
    },
    {
      "epoch": 0.46795270690728064,
      "grad_norm": 2.987772226333618,
      "learning_rate": 4.723247232472325e-06,
      "loss": 1.6374,
      "step": 376
    },
    {
      "epoch": 0.46919726197884254,
      "grad_norm": 3.3345255851745605,
      "learning_rate": 4.720940959409595e-06,
      "loss": 1.6405,
      "step": 377
    },
    {
      "epoch": 0.4704418170504045,
      "grad_norm": 3.0810067653656006,
      "learning_rate": 4.718634686346864e-06,
      "loss": 1.6229,
      "step": 378
    },
    {
      "epoch": 0.4716863721219664,
      "grad_norm": 3.246685266494751,
      "learning_rate": 4.716328413284133e-06,
      "loss": 1.6262,
      "step": 379
    },
    {
      "epoch": 0.4729309271935283,
      "grad_norm": 2.6052889823913574,
      "learning_rate": 4.714022140221403e-06,
      "loss": 1.6241,
      "step": 380
    },
    {
      "epoch": 0.4729309271935283,
      "eval_loss": 1.7957485914230347,
      "eval_runtime": 43.6626,
      "eval_samples_per_second": 22.903,
      "eval_steps_per_second": 0.962,
      "step": 380
    },
    {
      "epoch": 0.4741754822650902,
      "grad_norm": 3.747142791748047,
      "learning_rate": 4.711715867158672e-06,
      "loss": 1.6764,
      "step": 381
    },
    {
      "epoch": 0.47542003733665217,
      "grad_norm": 2.8508121967315674,
      "learning_rate": 4.709409594095941e-06,
      "loss": 1.6597,
      "step": 382
    },
    {
      "epoch": 0.47666459240821407,
      "grad_norm": 3.5615711212158203,
      "learning_rate": 4.70710332103321e-06,
      "loss": 1.6089,
      "step": 383
    },
    {
      "epoch": 0.477909147479776,
      "grad_norm": 2.5886309146881104,
      "learning_rate": 4.704797047970481e-06,
      "loss": 1.5996,
      "step": 384
    },
    {
      "epoch": 0.4791537025513379,
      "grad_norm": 3.3900864124298096,
      "learning_rate": 4.702490774907749e-06,
      "loss": 1.7137,
      "step": 385
    },
    {
      "epoch": 0.48039825762289984,
      "grad_norm": 2.913641929626465,
      "learning_rate": 4.700184501845019e-06,
      "loss": 1.6512,
      "step": 386
    },
    {
      "epoch": 0.48164281269446174,
      "grad_norm": 2.810722827911377,
      "learning_rate": 4.697878228782288e-06,
      "loss": 1.6939,
      "step": 387
    },
    {
      "epoch": 0.48288736776602365,
      "grad_norm": 3.2701401710510254,
      "learning_rate": 4.6955719557195575e-06,
      "loss": 1.6455,
      "step": 388
    },
    {
      "epoch": 0.48413192283758555,
      "grad_norm": 2.972931385040283,
      "learning_rate": 4.693265682656827e-06,
      "loss": 1.6272,
      "step": 389
    },
    {
      "epoch": 0.48537647790914745,
      "grad_norm": 3.2157890796661377,
      "learning_rate": 4.690959409594096e-06,
      "loss": 1.6409,
      "step": 390
    },
    {
      "epoch": 0.48537647790914745,
      "eval_loss": 1.7859243154525757,
      "eval_runtime": 45.2093,
      "eval_samples_per_second": 22.119,
      "eval_steps_per_second": 0.929,
      "step": 390
    },
    {
      "epoch": 0.4866210329807094,
      "grad_norm": 2.783360719680786,
      "learning_rate": 4.688653136531366e-06,
      "loss": 1.6303,
      "step": 391
    },
    {
      "epoch": 0.4878655880522713,
      "grad_norm": 2.6783013343811035,
      "learning_rate": 4.686346863468635e-06,
      "loss": 1.7259,
      "step": 392
    },
    {
      "epoch": 0.4891101431238332,
      "grad_norm": 2.899916172027588,
      "learning_rate": 4.6840405904059046e-06,
      "loss": 1.6434,
      "step": 393
    },
    {
      "epoch": 0.4903546981953951,
      "grad_norm": 2.604377269744873,
      "learning_rate": 4.681734317343174e-06,
      "loss": 1.6838,
      "step": 394
    },
    {
      "epoch": 0.4915992532669571,
      "grad_norm": 3.0830864906311035,
      "learning_rate": 4.6794280442804426e-06,
      "loss": 1.6053,
      "step": 395
    },
    {
      "epoch": 0.492843808338519,
      "grad_norm": 3.1381475925445557,
      "learning_rate": 4.677121771217713e-06,
      "loss": 1.6132,
      "step": 396
    },
    {
      "epoch": 0.4940883634100809,
      "grad_norm": 3.097426652908325,
      "learning_rate": 4.674815498154982e-06,
      "loss": 1.6237,
      "step": 397
    },
    {
      "epoch": 0.4953329184816428,
      "grad_norm": 2.9942305088043213,
      "learning_rate": 4.672509225092252e-06,
      "loss": 1.6554,
      "step": 398
    },
    {
      "epoch": 0.49657747355320475,
      "grad_norm": 2.834199905395508,
      "learning_rate": 4.67020295202952e-06,
      "loss": 1.6389,
      "step": 399
    },
    {
      "epoch": 0.49782202862476665,
      "grad_norm": 3.183389902114868,
      "learning_rate": 4.66789667896679e-06,
      "loss": 1.6518,
      "step": 400
    },
    {
      "epoch": 0.49782202862476665,
      "eval_loss": 1.781000018119812,
      "eval_runtime": 45.4127,
      "eval_samples_per_second": 22.02,
      "eval_steps_per_second": 0.925,
      "step": 400
    },
    {
      "epoch": 0.49906658369632856,
      "grad_norm": 2.8186683654785156,
      "learning_rate": 4.66559040590406e-06,
      "loss": 1.669,
      "step": 401
    },
    {
      "epoch": 0.5003111387678905,
      "grad_norm": 2.618595600128174,
      "learning_rate": 4.6632841328413285e-06,
      "loss": 1.6767,
      "step": 402
    },
    {
      "epoch": 0.5015556938394524,
      "grad_norm": 3.080432653427124,
      "learning_rate": 4.660977859778598e-06,
      "loss": 1.7096,
      "step": 403
    },
    {
      "epoch": 0.5028002489110143,
      "grad_norm": 2.6780874729156494,
      "learning_rate": 4.658671586715867e-06,
      "loss": 1.6576,
      "step": 404
    },
    {
      "epoch": 0.5040448039825762,
      "grad_norm": 2.8194704055786133,
      "learning_rate": 4.656365313653137e-06,
      "loss": 1.6405,
      "step": 405
    },
    {
      "epoch": 0.5052893590541382,
      "grad_norm": 3.2195825576782227,
      "learning_rate": 4.654059040590406e-06,
      "loss": 1.6631,
      "step": 406
    },
    {
      "epoch": 0.5065339141257,
      "grad_norm": 2.7719528675079346,
      "learning_rate": 4.651752767527676e-06,
      "loss": 1.6036,
      "step": 407
    },
    {
      "epoch": 0.507778469197262,
      "grad_norm": 3.186394453048706,
      "learning_rate": 4.649446494464945e-06,
      "loss": 1.6508,
      "step": 408
    },
    {
      "epoch": 0.5090230242688238,
      "grad_norm": 2.674922227859497,
      "learning_rate": 4.6471402214022145e-06,
      "loss": 1.6033,
      "step": 409
    },
    {
      "epoch": 0.5102675793403858,
      "grad_norm": 3.2772562503814697,
      "learning_rate": 4.644833948339484e-06,
      "loss": 1.5916,
      "step": 410
    },
    {
      "epoch": 0.5102675793403858,
      "eval_loss": 1.784346342086792,
      "eval_runtime": 42.2178,
      "eval_samples_per_second": 23.687,
      "eval_steps_per_second": 0.995,
      "step": 410
    },
    {
      "epoch": 0.5115121344119478,
      "grad_norm": 2.9295618534088135,
      "learning_rate": 4.642527675276753e-06,
      "loss": 1.6693,
      "step": 411
    },
    {
      "epoch": 0.5127566894835096,
      "grad_norm": 2.8492650985717773,
      "learning_rate": 4.640221402214023e-06,
      "loss": 1.658,
      "step": 412
    },
    {
      "epoch": 0.5140012445550716,
      "grad_norm": 2.921211004257202,
      "learning_rate": 4.637915129151292e-06,
      "loss": 1.6559,
      "step": 413
    },
    {
      "epoch": 0.5152457996266335,
      "grad_norm": 2.583395004272461,
      "learning_rate": 4.635608856088562e-06,
      "loss": 1.6869,
      "step": 414
    },
    {
      "epoch": 0.5164903546981954,
      "grad_norm": 2.79984712600708,
      "learning_rate": 4.63330258302583e-06,
      "loss": 1.6007,
      "step": 415
    },
    {
      "epoch": 0.5177349097697573,
      "grad_norm": 2.872856616973877,
      "learning_rate": 4.6309963099631e-06,
      "loss": 1.6132,
      "step": 416
    },
    {
      "epoch": 0.5189794648413192,
      "grad_norm": 2.7482125759124756,
      "learning_rate": 4.62869003690037e-06,
      "loss": 1.5805,
      "step": 417
    },
    {
      "epoch": 0.5202240199128811,
      "grad_norm": 2.9227206707000732,
      "learning_rate": 4.6263837638376384e-06,
      "loss": 1.6362,
      "step": 418
    },
    {
      "epoch": 0.5214685749844431,
      "grad_norm": 2.858302354812622,
      "learning_rate": 4.624077490774908e-06,
      "loss": 1.5934,
      "step": 419
    },
    {
      "epoch": 0.522713130056005,
      "grad_norm": 3.2696762084960938,
      "learning_rate": 4.621771217712177e-06,
      "loss": 1.6215,
      "step": 420
    },
    {
      "epoch": 0.522713130056005,
      "eval_loss": 1.7770200967788696,
      "eval_runtime": 46.0922,
      "eval_samples_per_second": 21.696,
      "eval_steps_per_second": 0.911,
      "step": 420
    },
    {
      "epoch": 0.5239576851275669,
      "grad_norm": 2.525740623474121,
      "learning_rate": 4.619464944649447e-06,
      "loss": 1.6193,
      "step": 421
    },
    {
      "epoch": 0.5252022401991289,
      "grad_norm": 2.9050614833831787,
      "learning_rate": 4.617158671586716e-06,
      "loss": 1.69,
      "step": 422
    },
    {
      "epoch": 0.5264467952706907,
      "grad_norm": 2.959547519683838,
      "learning_rate": 4.6148523985239856e-06,
      "loss": 1.6874,
      "step": 423
    },
    {
      "epoch": 0.5276913503422527,
      "grad_norm": 2.6331565380096436,
      "learning_rate": 4.612546125461255e-06,
      "loss": 1.5795,
      "step": 424
    },
    {
      "epoch": 0.5289359054138145,
      "grad_norm": 2.7323524951934814,
      "learning_rate": 4.610239852398524e-06,
      "loss": 1.5882,
      "step": 425
    },
    {
      "epoch": 0.5301804604853765,
      "grad_norm": 2.8021790981292725,
      "learning_rate": 4.607933579335794e-06,
      "loss": 1.5852,
      "step": 426
    },
    {
      "epoch": 0.5314250155569384,
      "grad_norm": 2.984299659729004,
      "learning_rate": 4.605627306273063e-06,
      "loss": 1.6168,
      "step": 427
    },
    {
      "epoch": 0.5326695706285003,
      "grad_norm": 2.92749285697937,
      "learning_rate": 4.603321033210333e-06,
      "loss": 1.5768,
      "step": 428
    },
    {
      "epoch": 0.5339141257000622,
      "grad_norm": 2.7681751251220703,
      "learning_rate": 4.601014760147602e-06,
      "loss": 1.5831,
      "step": 429
    },
    {
      "epoch": 0.5351586807716241,
      "grad_norm": 2.965458631515503,
      "learning_rate": 4.5987084870848715e-06,
      "loss": 1.6412,
      "step": 430
    },
    {
      "epoch": 0.5351586807716241,
      "eval_loss": 1.766748070716858,
      "eval_runtime": 51.1098,
      "eval_samples_per_second": 19.566,
      "eval_steps_per_second": 0.822,
      "step": 430
    },
    {
      "epoch": 0.536403235843186,
      "grad_norm": 2.681853771209717,
      "learning_rate": 4.59640221402214e-06,
      "loss": 1.6302,
      "step": 431
    },
    {
      "epoch": 0.537647790914748,
      "grad_norm": 2.698493480682373,
      "learning_rate": 4.5940959409594095e-06,
      "loss": 1.6774,
      "step": 432
    },
    {
      "epoch": 0.5388923459863099,
      "grad_norm": 3.1715168952941895,
      "learning_rate": 4.59178966789668e-06,
      "loss": 1.5617,
      "step": 433
    },
    {
      "epoch": 0.5401369010578718,
      "grad_norm": 2.98128080368042,
      "learning_rate": 4.589483394833948e-06,
      "loss": 1.5901,
      "step": 434
    },
    {
      "epoch": 0.5413814561294338,
      "grad_norm": 3.3887572288513184,
      "learning_rate": 4.587177121771218e-06,
      "loss": 1.6448,
      "step": 435
    },
    {
      "epoch": 0.5426260112009956,
      "grad_norm": 2.647975206375122,
      "learning_rate": 4.584870848708487e-06,
      "loss": 1.6209,
      "step": 436
    },
    {
      "epoch": 0.5438705662725576,
      "grad_norm": 2.833264112472534,
      "learning_rate": 4.5825645756457575e-06,
      "loss": 1.6875,
      "step": 437
    },
    {
      "epoch": 0.5451151213441194,
      "grad_norm": 3.0966858863830566,
      "learning_rate": 4.580258302583026e-06,
      "loss": 1.6291,
      "step": 438
    },
    {
      "epoch": 0.5463596764156814,
      "grad_norm": 2.768793821334839,
      "learning_rate": 4.5779520295202955e-06,
      "loss": 1.6183,
      "step": 439
    },
    {
      "epoch": 0.5476042314872434,
      "grad_norm": 3.4009342193603516,
      "learning_rate": 4.575645756457565e-06,
      "loss": 1.6602,
      "step": 440
    },
    {
      "epoch": 0.5476042314872434,
      "eval_loss": 1.76938796043396,
      "eval_runtime": 46.2093,
      "eval_samples_per_second": 21.641,
      "eval_steps_per_second": 0.909,
      "step": 440
    },
    {
      "epoch": 0.5488487865588052,
      "grad_norm": 2.769113063812256,
      "learning_rate": 4.573339483394834e-06,
      "loss": 1.6453,
      "step": 441
    },
    {
      "epoch": 0.5500933416303672,
      "grad_norm": 3.3592801094055176,
      "learning_rate": 4.571033210332104e-06,
      "loss": 1.6602,
      "step": 442
    },
    {
      "epoch": 0.551337896701929,
      "grad_norm": 2.6817564964294434,
      "learning_rate": 4.568726937269373e-06,
      "loss": 1.5714,
      "step": 443
    },
    {
      "epoch": 0.552582451773491,
      "grad_norm": 3.093724250793457,
      "learning_rate": 4.566420664206643e-06,
      "loss": 1.6924,
      "step": 444
    },
    {
      "epoch": 0.5538270068450529,
      "grad_norm": 2.766580104827881,
      "learning_rate": 4.564114391143912e-06,
      "loss": 1.5714,
      "step": 445
    },
    {
      "epoch": 0.5550715619166148,
      "grad_norm": 2.868870973587036,
      "learning_rate": 4.5618081180811814e-06,
      "loss": 1.5823,
      "step": 446
    },
    {
      "epoch": 0.5563161169881767,
      "grad_norm": 2.8286170959472656,
      "learning_rate": 4.559501845018451e-06,
      "loss": 1.6123,
      "step": 447
    },
    {
      "epoch": 0.5575606720597387,
      "grad_norm": 2.66096568107605,
      "learning_rate": 4.5571955719557194e-06,
      "loss": 1.6044,
      "step": 448
    },
    {
      "epoch": 0.5588052271313005,
      "grad_norm": 2.7411820888519287,
      "learning_rate": 4.55488929889299e-06,
      "loss": 1.6361,
      "step": 449
    },
    {
      "epoch": 0.5600497822028625,
      "grad_norm": 2.7591240406036377,
      "learning_rate": 4.552583025830259e-06,
      "loss": 1.6257,
      "step": 450
    },
    {
      "epoch": 0.5600497822028625,
      "eval_loss": 1.758321762084961,
      "eval_runtime": 52.9135,
      "eval_samples_per_second": 18.899,
      "eval_steps_per_second": 0.794,
      "step": 450
    },
    {
      "epoch": 0.5612943372744243,
      "grad_norm": 2.733456611633301,
      "learning_rate": 4.550276752767528e-06,
      "loss": 1.7159,
      "step": 451
    },
    {
      "epoch": 0.5625388923459863,
      "grad_norm": 2.7318639755249023,
      "learning_rate": 4.547970479704797e-06,
      "loss": 1.6062,
      "step": 452
    },
    {
      "epoch": 0.5637834474175483,
      "grad_norm": 2.8235232830047607,
      "learning_rate": 4.5456642066420666e-06,
      "loss": 1.6616,
      "step": 453
    },
    {
      "epoch": 0.5650280024891101,
      "grad_norm": 2.499028444290161,
      "learning_rate": 4.543357933579336e-06,
      "loss": 1.5795,
      "step": 454
    },
    {
      "epoch": 0.5662725575606721,
      "grad_norm": 2.7547316551208496,
      "learning_rate": 4.541051660516605e-06,
      "loss": 1.6204,
      "step": 455
    },
    {
      "epoch": 0.5675171126322339,
      "grad_norm": 2.7086215019226074,
      "learning_rate": 4.538745387453875e-06,
      "loss": 1.5611,
      "step": 456
    },
    {
      "epoch": 0.5687616677037959,
      "grad_norm": 2.911731719970703,
      "learning_rate": 4.536439114391144e-06,
      "loss": 1.645,
      "step": 457
    },
    {
      "epoch": 0.5700062227753578,
      "grad_norm": 2.5553250312805176,
      "learning_rate": 4.534132841328414e-06,
      "loss": 1.6093,
      "step": 458
    },
    {
      "epoch": 0.5712507778469197,
      "grad_norm": 2.795919895172119,
      "learning_rate": 4.531826568265683e-06,
      "loss": 1.6,
      "step": 459
    },
    {
      "epoch": 0.5724953329184816,
      "grad_norm": 2.92104434967041,
      "learning_rate": 4.5295202952029525e-06,
      "loss": 1.6444,
      "step": 460
    },
    {
      "epoch": 0.5724953329184816,
      "eval_loss": 1.760049819946289,
      "eval_runtime": 48.811,
      "eval_samples_per_second": 20.487,
      "eval_steps_per_second": 0.86,
      "step": 460
    },
    {
      "epoch": 0.5737398879900436,
      "grad_norm": 2.6339287757873535,
      "learning_rate": 4.527214022140222e-06,
      "loss": 1.6133,
      "step": 461
    },
    {
      "epoch": 0.5749844430616055,
      "grad_norm": 2.531398057937622,
      "learning_rate": 4.524907749077491e-06,
      "loss": 1.5931,
      "step": 462
    },
    {
      "epoch": 0.5762289981331674,
      "grad_norm": 2.8858656883239746,
      "learning_rate": 4.522601476014761e-06,
      "loss": 1.6287,
      "step": 463
    },
    {
      "epoch": 0.5774735532047293,
      "grad_norm": 2.7309634685516357,
      "learning_rate": 4.520295202952029e-06,
      "loss": 1.6102,
      "step": 464
    },
    {
      "epoch": 0.5787181082762912,
      "grad_norm": 2.7758548259735107,
      "learning_rate": 4.5179889298893e-06,
      "loss": 1.6292,
      "step": 465
    },
    {
      "epoch": 0.5799626633478532,
      "grad_norm": 2.6181721687316895,
      "learning_rate": 4.515682656826569e-06,
      "loss": 1.6113,
      "step": 466
    },
    {
      "epoch": 0.581207218419415,
      "grad_norm": 2.6215691566467285,
      "learning_rate": 4.513376383763838e-06,
      "loss": 1.5158,
      "step": 467
    },
    {
      "epoch": 0.582451773490977,
      "grad_norm": 2.727487087249756,
      "learning_rate": 4.511070110701107e-06,
      "loss": 1.6212,
      "step": 468
    },
    {
      "epoch": 0.583696328562539,
      "grad_norm": 2.6248037815093994,
      "learning_rate": 4.5087638376383765e-06,
      "loss": 1.6117,
      "step": 469
    },
    {
      "epoch": 0.5849408836341008,
      "grad_norm": 2.714291572570801,
      "learning_rate": 4.506457564575646e-06,
      "loss": 1.612,
      "step": 470
    },
    {
      "epoch": 0.5849408836341008,
      "eval_loss": 1.7569347620010376,
      "eval_runtime": 51.8475,
      "eval_samples_per_second": 19.287,
      "eval_steps_per_second": 0.81,
      "step": 470
    },
    {
      "epoch": 0.5861854387056628,
      "grad_norm": 2.752577066421509,
      "learning_rate": 4.504151291512915e-06,
      "loss": 1.5969,
      "step": 471
    },
    {
      "epoch": 0.5874299937772246,
      "grad_norm": 2.688563346862793,
      "learning_rate": 4.501845018450185e-06,
      "loss": 1.6085,
      "step": 472
    },
    {
      "epoch": 0.5886745488487866,
      "grad_norm": 2.8581559658050537,
      "learning_rate": 4.499538745387454e-06,
      "loss": 1.6688,
      "step": 473
    },
    {
      "epoch": 0.5899191039203485,
      "grad_norm": 2.8520970344543457,
      "learning_rate": 4.497232472324724e-06,
      "loss": 1.6034,
      "step": 474
    },
    {
      "epoch": 0.5911636589919104,
      "grad_norm": 2.8052237033843994,
      "learning_rate": 4.494926199261993e-06,
      "loss": 1.5656,
      "step": 475
    },
    {
      "epoch": 0.5924082140634723,
      "grad_norm": 2.619663953781128,
      "learning_rate": 4.4926199261992624e-06,
      "loss": 1.5543,
      "step": 476
    },
    {
      "epoch": 0.5936527691350342,
      "grad_norm": 2.604964017868042,
      "learning_rate": 4.490313653136532e-06,
      "loss": 1.5671,
      "step": 477
    },
    {
      "epoch": 0.5948973242065961,
      "grad_norm": 2.985004186630249,
      "learning_rate": 4.488007380073801e-06,
      "loss": 1.5681,
      "step": 478
    },
    {
      "epoch": 0.5961418792781581,
      "grad_norm": 2.8348851203918457,
      "learning_rate": 4.485701107011071e-06,
      "loss": 1.6214,
      "step": 479
    },
    {
      "epoch": 0.5973864343497199,
      "grad_norm": 2.580967664718628,
      "learning_rate": 4.483394833948339e-06,
      "loss": 1.5753,
      "step": 480
    },
    {
      "epoch": 0.5973864343497199,
      "eval_loss": 1.7542102336883545,
      "eval_runtime": 49.5092,
      "eval_samples_per_second": 20.198,
      "eval_steps_per_second": 0.848,
      "step": 480
    },
    {
      "epoch": 0.5986309894212819,
      "grad_norm": 3.0104517936706543,
      "learning_rate": 4.4810885608856096e-06,
      "loss": 1.6282,
      "step": 481
    },
    {
      "epoch": 0.5998755444928439,
      "grad_norm": 2.9261820316314697,
      "learning_rate": 4.478782287822879e-06,
      "loss": 1.5811,
      "step": 482
    },
    {
      "epoch": 0.6011200995644057,
      "grad_norm": 2.9920477867126465,
      "learning_rate": 4.4764760147601476e-06,
      "loss": 1.5905,
      "step": 483
    },
    {
      "epoch": 0.6023646546359677,
      "grad_norm": 3.046887159347534,
      "learning_rate": 4.474169741697417e-06,
      "loss": 1.6587,
      "step": 484
    },
    {
      "epoch": 0.6036092097075295,
      "grad_norm": 2.9660050868988037,
      "learning_rate": 4.471863468634686e-06,
      "loss": 1.5888,
      "step": 485
    },
    {
      "epoch": 0.6048537647790915,
      "grad_norm": 2.770362377166748,
      "learning_rate": 4.469557195571957e-06,
      "loss": 1.5925,
      "step": 486
    },
    {
      "epoch": 0.6060983198506534,
      "grad_norm": 2.621699094772339,
      "learning_rate": 4.467250922509225e-06,
      "loss": 1.5671,
      "step": 487
    },
    {
      "epoch": 0.6073428749222153,
      "grad_norm": 2.749340295791626,
      "learning_rate": 4.464944649446495e-06,
      "loss": 1.5672,
      "step": 488
    },
    {
      "epoch": 0.6085874299937772,
      "grad_norm": 3.091601848602295,
      "learning_rate": 4.462638376383764e-06,
      "loss": 1.5965,
      "step": 489
    },
    {
      "epoch": 0.6098319850653391,
      "grad_norm": 2.6989245414733887,
      "learning_rate": 4.4603321033210335e-06,
      "loss": 1.5845,
      "step": 490
    },
    {
      "epoch": 0.6098319850653391,
      "eval_loss": 1.7461379766464233,
      "eval_runtime": 52.879,
      "eval_samples_per_second": 18.911,
      "eval_steps_per_second": 0.794,
      "step": 490
    },
    {
      "epoch": 0.611076540136901,
      "grad_norm": 2.8998405933380127,
      "learning_rate": 4.458025830258303e-06,
      "loss": 1.5764,
      "step": 491
    },
    {
      "epoch": 0.612321095208463,
      "grad_norm": 2.6035947799682617,
      "learning_rate": 4.455719557195572e-06,
      "loss": 1.5988,
      "step": 492
    },
    {
      "epoch": 0.6135656502800249,
      "grad_norm": 2.712857484817505,
      "learning_rate": 4.453413284132842e-06,
      "loss": 1.6276,
      "step": 493
    },
    {
      "epoch": 0.6148102053515868,
      "grad_norm": 2.73052978515625,
      "learning_rate": 4.451107011070111e-06,
      "loss": 1.5907,
      "step": 494
    },
    {
      "epoch": 0.6160547604231488,
      "grad_norm": 2.8072566986083984,
      "learning_rate": 4.448800738007381e-06,
      "loss": 1.5771,
      "step": 495
    },
    {
      "epoch": 0.6172993154947106,
      "grad_norm": 2.6016132831573486,
      "learning_rate": 4.446494464944649e-06,
      "loss": 1.5983,
      "step": 496
    },
    {
      "epoch": 0.6185438705662726,
      "grad_norm": 2.9697721004486084,
      "learning_rate": 4.4441881918819195e-06,
      "loss": 1.5974,
      "step": 497
    },
    {
      "epoch": 0.6197884256378344,
      "grad_norm": 2.6966018676757812,
      "learning_rate": 4.441881918819189e-06,
      "loss": 1.6042,
      "step": 498
    },
    {
      "epoch": 0.6210329807093964,
      "grad_norm": 2.871964931488037,
      "learning_rate": 4.439575645756458e-06,
      "loss": 1.5668,
      "step": 499
    },
    {
      "epoch": 0.6222775357809583,
      "grad_norm": 2.791602611541748,
      "learning_rate": 4.437269372693727e-06,
      "loss": 1.5222,
      "step": 500
    },
    {
      "epoch": 0.6222775357809583,
      "eval_loss": 1.756259799003601,
      "eval_runtime": 53.0959,
      "eval_samples_per_second": 18.834,
      "eval_steps_per_second": 0.791,
      "step": 500
    },
    {
      "epoch": 0.6235220908525202,
      "grad_norm": 2.6534979343414307,
      "learning_rate": 4.434963099630996e-06,
      "loss": 1.5618,
      "step": 501
    },
    {
      "epoch": 0.6247666459240822,
      "grad_norm": 3.0166008472442627,
      "learning_rate": 4.432656826568267e-06,
      "loss": 1.639,
      "step": 502
    },
    {
      "epoch": 0.6260112009956441,
      "grad_norm": 2.7879786491394043,
      "learning_rate": 4.430350553505535e-06,
      "loss": 1.5521,
      "step": 503
    },
    {
      "epoch": 0.627255756067206,
      "grad_norm": 2.695650339126587,
      "learning_rate": 4.428044280442805e-06,
      "loss": 1.6086,
      "step": 504
    },
    {
      "epoch": 0.6285003111387679,
      "grad_norm": 2.8340883255004883,
      "learning_rate": 4.425738007380074e-06,
      "loss": 1.583,
      "step": 505
    },
    {
      "epoch": 0.6297448662103298,
      "grad_norm": 2.872870683670044,
      "learning_rate": 4.4234317343173434e-06,
      "loss": 1.5633,
      "step": 506
    },
    {
      "epoch": 0.6309894212818917,
      "grad_norm": 2.840087652206421,
      "learning_rate": 4.421125461254613e-06,
      "loss": 1.603,
      "step": 507
    },
    {
      "epoch": 0.6322339763534537,
      "grad_norm": 2.86578631401062,
      "learning_rate": 4.418819188191882e-06,
      "loss": 1.6031,
      "step": 508
    },
    {
      "epoch": 0.6334785314250155,
      "grad_norm": 2.419501543045044,
      "learning_rate": 4.416512915129152e-06,
      "loss": 1.5975,
      "step": 509
    },
    {
      "epoch": 0.6347230864965775,
      "grad_norm": 2.6204769611358643,
      "learning_rate": 4.414206642066421e-06,
      "loss": 1.608,
      "step": 510
    },
    {
      "epoch": 0.6347230864965775,
      "eval_loss": 1.7449734210968018,
      "eval_runtime": 44.3185,
      "eval_samples_per_second": 22.564,
      "eval_steps_per_second": 0.948,
      "step": 510
    },
    {
      "epoch": 0.6359676415681393,
      "grad_norm": 2.837980270385742,
      "learning_rate": 4.4119003690036905e-06,
      "loss": 1.625,
      "step": 511
    },
    {
      "epoch": 0.6372121966397013,
      "grad_norm": 2.589782238006592,
      "learning_rate": 4.40959409594096e-06,
      "loss": 1.5976,
      "step": 512
    },
    {
      "epoch": 0.6384567517112633,
      "grad_norm": 2.596064567565918,
      "learning_rate": 4.407287822878229e-06,
      "loss": 1.5568,
      "step": 513
    },
    {
      "epoch": 0.6397013067828251,
      "grad_norm": 3.031325578689575,
      "learning_rate": 4.404981549815499e-06,
      "loss": 1.5744,
      "step": 514
    },
    {
      "epoch": 0.6409458618543871,
      "grad_norm": 2.581284761428833,
      "learning_rate": 4.402675276752768e-06,
      "loss": 1.5822,
      "step": 515
    },
    {
      "epoch": 0.642190416925949,
      "grad_norm": 2.7712416648864746,
      "learning_rate": 4.400369003690037e-06,
      "loss": 1.606,
      "step": 516
    },
    {
      "epoch": 0.6434349719975109,
      "grad_norm": 2.5830018520355225,
      "learning_rate": 4.398062730627306e-06,
      "loss": 1.5845,
      "step": 517
    },
    {
      "epoch": 0.6446795270690728,
      "grad_norm": 2.6646499633789062,
      "learning_rate": 4.3957564575645765e-06,
      "loss": 1.6323,
      "step": 518
    },
    {
      "epoch": 0.6459240821406347,
      "grad_norm": 2.614750862121582,
      "learning_rate": 4.393450184501845e-06,
      "loss": 1.5447,
      "step": 519
    },
    {
      "epoch": 0.6471686372121966,
      "grad_norm": 2.6826741695404053,
      "learning_rate": 4.3911439114391145e-06,
      "loss": 1.6081,
      "step": 520
    },
    {
      "epoch": 0.6471686372121966,
      "eval_loss": 1.7463606595993042,
      "eval_runtime": 45.7383,
      "eval_samples_per_second": 21.864,
      "eval_steps_per_second": 0.918,
      "step": 520
    },
    {
      "epoch": 0.6484131922837586,
      "grad_norm": 2.9165496826171875,
      "learning_rate": 4.388837638376384e-06,
      "loss": 1.5853,
      "step": 521
    },
    {
      "epoch": 0.6496577473553204,
      "grad_norm": 2.789862871170044,
      "learning_rate": 4.386531365313653e-06,
      "loss": 1.6115,
      "step": 522
    },
    {
      "epoch": 0.6509023024268824,
      "grad_norm": 2.690093994140625,
      "learning_rate": 4.384225092250923e-06,
      "loss": 1.5625,
      "step": 523
    },
    {
      "epoch": 0.6521468574984443,
      "grad_norm": 2.660036087036133,
      "learning_rate": 4.381918819188192e-06,
      "loss": 1.6065,
      "step": 524
    },
    {
      "epoch": 0.6533914125700062,
      "grad_norm": 2.5939362049102783,
      "learning_rate": 4.379612546125462e-06,
      "loss": 1.5593,
      "step": 525
    },
    {
      "epoch": 0.6546359676415682,
      "grad_norm": 2.6998841762542725,
      "learning_rate": 4.377306273062731e-06,
      "loss": 1.6168,
      "step": 526
    },
    {
      "epoch": 0.65588052271313,
      "grad_norm": 2.5756912231445312,
      "learning_rate": 4.3750000000000005e-06,
      "loss": 1.528,
      "step": 527
    },
    {
      "epoch": 0.657125077784692,
      "grad_norm": 2.6959640979766846,
      "learning_rate": 4.37269372693727e-06,
      "loss": 1.5987,
      "step": 528
    },
    {
      "epoch": 0.6583696328562539,
      "grad_norm": 2.773494005203247,
      "learning_rate": 4.370387453874539e-06,
      "loss": 1.6018,
      "step": 529
    },
    {
      "epoch": 0.6596141879278158,
      "grad_norm": 2.8194501399993896,
      "learning_rate": 4.368081180811809e-06,
      "loss": 1.6082,
      "step": 530
    },
    {
      "epoch": 0.6596141879278158,
      "eval_loss": 1.7327930927276611,
      "eval_runtime": 46.7893,
      "eval_samples_per_second": 21.372,
      "eval_steps_per_second": 0.898,
      "step": 530
    },
    {
      "epoch": 0.6608587429993777,
      "grad_norm": 2.721564292907715,
      "learning_rate": 4.365774907749078e-06,
      "loss": 1.5549,
      "step": 531
    },
    {
      "epoch": 0.6621032980709396,
      "grad_norm": 2.6635236740112305,
      "learning_rate": 4.363468634686347e-06,
      "loss": 1.6016,
      "step": 532
    },
    {
      "epoch": 0.6633478531425016,
      "grad_norm": 2.6541085243225098,
      "learning_rate": 4.361162361623616e-06,
      "loss": 1.5574,
      "step": 533
    },
    {
      "epoch": 0.6645924082140635,
      "grad_norm": 2.9270925521850586,
      "learning_rate": 4.3588560885608864e-06,
      "loss": 1.5722,
      "step": 534
    },
    {
      "epoch": 0.6658369632856254,
      "grad_norm": 2.6041259765625,
      "learning_rate": 4.356549815498156e-06,
      "loss": 1.6181,
      "step": 535
    },
    {
      "epoch": 0.6670815183571873,
      "grad_norm": 2.7029967308044434,
      "learning_rate": 4.354243542435424e-06,
      "loss": 1.5576,
      "step": 536
    },
    {
      "epoch": 0.6683260734287492,
      "grad_norm": 2.805567979812622,
      "learning_rate": 4.351937269372694e-06,
      "loss": 1.6279,
      "step": 537
    },
    {
      "epoch": 0.6695706285003111,
      "grad_norm": 2.6513259410858154,
      "learning_rate": 4.349630996309963e-06,
      "loss": 1.5626,
      "step": 538
    },
    {
      "epoch": 0.6708151835718731,
      "grad_norm": 2.7320876121520996,
      "learning_rate": 4.347324723247233e-06,
      "loss": 1.6,
      "step": 539
    },
    {
      "epoch": 0.6720597386434349,
      "grad_norm": 2.9504754543304443,
      "learning_rate": 4.345018450184502e-06,
      "loss": 1.6755,
      "step": 540
    },
    {
      "epoch": 0.6720597386434349,
      "eval_loss": 1.733923077583313,
      "eval_runtime": 45.4809,
      "eval_samples_per_second": 21.987,
      "eval_steps_per_second": 0.923,
      "step": 540
    },
    {
      "epoch": 0.6733042937149969,
      "grad_norm": 2.7561354637145996,
      "learning_rate": 4.3427121771217715e-06,
      "loss": 1.5624,
      "step": 541
    },
    {
      "epoch": 0.6745488487865589,
      "grad_norm": 2.612337827682495,
      "learning_rate": 4.340405904059041e-06,
      "loss": 1.5806,
      "step": 542
    },
    {
      "epoch": 0.6757934038581207,
      "grad_norm": 3.2763278484344482,
      "learning_rate": 4.33809963099631e-06,
      "loss": 1.5939,
      "step": 543
    },
    {
      "epoch": 0.6770379589296827,
      "grad_norm": 2.632627248764038,
      "learning_rate": 4.33579335793358e-06,
      "loss": 1.564,
      "step": 544
    },
    {
      "epoch": 0.6782825140012445,
      "grad_norm": 2.9776413440704346,
      "learning_rate": 4.333487084870848e-06,
      "loss": 1.5624,
      "step": 545
    },
    {
      "epoch": 0.6795270690728065,
      "grad_norm": 3.083935260772705,
      "learning_rate": 4.331180811808119e-06,
      "loss": 1.5957,
      "step": 546
    },
    {
      "epoch": 0.6807716241443684,
      "grad_norm": 2.8425586223602295,
      "learning_rate": 4.328874538745388e-06,
      "loss": 1.5559,
      "step": 547
    },
    {
      "epoch": 0.6820161792159303,
      "grad_norm": 3.0011188983917236,
      "learning_rate": 4.3265682656826575e-06,
      "loss": 1.5308,
      "step": 548
    },
    {
      "epoch": 0.6832607342874922,
      "grad_norm": 2.7645163536071777,
      "learning_rate": 4.324261992619926e-06,
      "loss": 1.6114,
      "step": 549
    },
    {
      "epoch": 0.6845052893590542,
      "grad_norm": 2.8784844875335693,
      "learning_rate": 4.321955719557196e-06,
      "loss": 1.5591,
      "step": 550
    },
    {
      "epoch": 0.6845052893590542,
      "eval_loss": 1.7328442335128784,
      "eval_runtime": 45.5776,
      "eval_samples_per_second": 21.941,
      "eval_steps_per_second": 0.922,
      "step": 550
    },
    {
      "epoch": 0.685749844430616,
      "grad_norm": 3.1323604583740234,
      "learning_rate": 4.319649446494466e-06,
      "loss": 1.5626,
      "step": 551
    },
    {
      "epoch": 0.686994399502178,
      "grad_norm": 2.7367660999298096,
      "learning_rate": 4.317343173431734e-06,
      "loss": 1.5914,
      "step": 552
    },
    {
      "epoch": 0.6882389545737398,
      "grad_norm": 3.056155204772949,
      "learning_rate": 4.315036900369004e-06,
      "loss": 1.6014,
      "step": 553
    },
    {
      "epoch": 0.6894835096453018,
      "grad_norm": 2.7824926376342773,
      "learning_rate": 4.312730627306273e-06,
      "loss": 1.5508,
      "step": 554
    },
    {
      "epoch": 0.6907280647168638,
      "grad_norm": 2.629610538482666,
      "learning_rate": 4.310424354243543e-06,
      "loss": 1.5385,
      "step": 555
    },
    {
      "epoch": 0.6919726197884256,
      "grad_norm": 2.753021240234375,
      "learning_rate": 4.308118081180812e-06,
      "loss": 1.6211,
      "step": 556
    },
    {
      "epoch": 0.6932171748599876,
      "grad_norm": 2.8113718032836914,
      "learning_rate": 4.3058118081180815e-06,
      "loss": 1.5904,
      "step": 557
    },
    {
      "epoch": 0.6944617299315494,
      "grad_norm": 2.849471092224121,
      "learning_rate": 4.303505535055351e-06,
      "loss": 1.6126,
      "step": 558
    },
    {
      "epoch": 0.6957062850031114,
      "grad_norm": 2.6744682788848877,
      "learning_rate": 4.30119926199262e-06,
      "loss": 1.5743,
      "step": 559
    },
    {
      "epoch": 0.6969508400746733,
      "grad_norm": 2.7900655269622803,
      "learning_rate": 4.29889298892989e-06,
      "loss": 1.5865,
      "step": 560
    },
    {
      "epoch": 0.6969508400746733,
      "eval_loss": 1.724618673324585,
      "eval_runtime": 44.2181,
      "eval_samples_per_second": 22.615,
      "eval_steps_per_second": 0.95,
      "step": 560
    },
    {
      "epoch": 0.6981953951462352,
      "grad_norm": 2.7363393306732178,
      "learning_rate": 4.296586715867159e-06,
      "loss": 1.5347,
      "step": 561
    },
    {
      "epoch": 0.6994399502177971,
      "grad_norm": 2.696748733520508,
      "learning_rate": 4.2942804428044286e-06,
      "loss": 1.5463,
      "step": 562
    },
    {
      "epoch": 0.7006845052893591,
      "grad_norm": 2.5793895721435547,
      "learning_rate": 4.291974169741698e-06,
      "loss": 1.6231,
      "step": 563
    },
    {
      "epoch": 0.701929060360921,
      "grad_norm": 2.576230764389038,
      "learning_rate": 4.289667896678967e-06,
      "loss": 1.5663,
      "step": 564
    },
    {
      "epoch": 0.7031736154324829,
      "grad_norm": 2.718759298324585,
      "learning_rate": 4.287361623616236e-06,
      "loss": 1.5316,
      "step": 565
    },
    {
      "epoch": 0.7044181705040448,
      "grad_norm": 2.651751756668091,
      "learning_rate": 4.285055350553506e-06,
      "loss": 1.5777,
      "step": 566
    },
    {
      "epoch": 0.7056627255756067,
      "grad_norm": 2.931264638900757,
      "learning_rate": 4.282749077490776e-06,
      "loss": 1.5647,
      "step": 567
    },
    {
      "epoch": 0.7069072806471687,
      "grad_norm": 2.8062546253204346,
      "learning_rate": 4.280442804428044e-06,
      "loss": 1.5685,
      "step": 568
    },
    {
      "epoch": 0.7081518357187305,
      "grad_norm": 2.7551019191741943,
      "learning_rate": 4.278136531365314e-06,
      "loss": 1.5659,
      "step": 569
    },
    {
      "epoch": 0.7093963907902925,
      "grad_norm": 2.799724578857422,
      "learning_rate": 4.275830258302583e-06,
      "loss": 1.5917,
      "step": 570
    },
    {
      "epoch": 0.7093963907902925,
      "eval_loss": 1.7217484712600708,
      "eval_runtime": 44.5312,
      "eval_samples_per_second": 22.456,
      "eval_steps_per_second": 0.943,
      "step": 570
    },
    {
      "epoch": 0.7106409458618543,
      "grad_norm": 2.8496406078338623,
      "learning_rate": 4.273523985239853e-06,
      "loss": 1.5607,
      "step": 571
    },
    {
      "epoch": 0.7118855009334163,
      "grad_norm": 2.5612425804138184,
      "learning_rate": 4.271217712177122e-06,
      "loss": 1.5496,
      "step": 572
    },
    {
      "epoch": 0.7131300560049783,
      "grad_norm": 2.6605045795440674,
      "learning_rate": 4.268911439114391e-06,
      "loss": 1.5371,
      "step": 573
    },
    {
      "epoch": 0.7143746110765401,
      "grad_norm": 2.823802947998047,
      "learning_rate": 4.266605166051661e-06,
      "loss": 1.5699,
      "step": 574
    },
    {
      "epoch": 0.7156191661481021,
      "grad_norm": 2.8212928771972656,
      "learning_rate": 4.26429889298893e-06,
      "loss": 1.5529,
      "step": 575
    },
    {
      "epoch": 0.716863721219664,
      "grad_norm": 2.7193386554718018,
      "learning_rate": 4.2619926199262e-06,
      "loss": 1.5723,
      "step": 576
    },
    {
      "epoch": 0.7181082762912259,
      "grad_norm": 2.8348941802978516,
      "learning_rate": 4.259686346863469e-06,
      "loss": 1.5212,
      "step": 577
    },
    {
      "epoch": 0.7193528313627878,
      "grad_norm": 2.7177276611328125,
      "learning_rate": 4.2573800738007385e-06,
      "loss": 1.5688,
      "step": 578
    },
    {
      "epoch": 0.7205973864343497,
      "grad_norm": 2.675734043121338,
      "learning_rate": 4.255073800738008e-06,
      "loss": 1.5345,
      "step": 579
    },
    {
      "epoch": 0.7218419415059116,
      "grad_norm": 2.640942335128784,
      "learning_rate": 4.252767527675277e-06,
      "loss": 1.5874,
      "step": 580
    },
    {
      "epoch": 0.7218419415059116,
      "eval_loss": 1.72821044921875,
      "eval_runtime": 45.9409,
      "eval_samples_per_second": 21.767,
      "eval_steps_per_second": 0.914,
      "step": 580
    },
    {
      "epoch": 0.7230864965774736,
      "grad_norm": 2.785066604614258,
      "learning_rate": 4.250461254612546e-06,
      "loss": 1.5685,
      "step": 581
    },
    {
      "epoch": 0.7243310516490354,
      "grad_norm": 2.8149311542510986,
      "learning_rate": 4.248154981549816e-06,
      "loss": 1.5106,
      "step": 582
    },
    {
      "epoch": 0.7255756067205974,
      "grad_norm": 2.5781238079071045,
      "learning_rate": 4.245848708487086e-06,
      "loss": 1.5895,
      "step": 583
    },
    {
      "epoch": 0.7268201617921594,
      "grad_norm": 2.6275103092193604,
      "learning_rate": 4.243542435424355e-06,
      "loss": 1.601,
      "step": 584
    },
    {
      "epoch": 0.7280647168637212,
      "grad_norm": 2.5282299518585205,
      "learning_rate": 4.241236162361624e-06,
      "loss": 1.5376,
      "step": 585
    },
    {
      "epoch": 0.7293092719352832,
      "grad_norm": 2.7331771850585938,
      "learning_rate": 4.238929889298893e-06,
      "loss": 1.5855,
      "step": 586
    },
    {
      "epoch": 0.730553827006845,
      "grad_norm": 2.6941936016082764,
      "learning_rate": 4.236623616236163e-06,
      "loss": 1.5937,
      "step": 587
    },
    {
      "epoch": 0.731798382078407,
      "grad_norm": 2.924062728881836,
      "learning_rate": 4.234317343173432e-06,
      "loss": 1.5914,
      "step": 588
    },
    {
      "epoch": 0.7330429371499689,
      "grad_norm": 2.687530279159546,
      "learning_rate": 4.232011070110701e-06,
      "loss": 1.5598,
      "step": 589
    },
    {
      "epoch": 0.7342874922215308,
      "grad_norm": 2.7230913639068604,
      "learning_rate": 4.229704797047971e-06,
      "loss": 1.6048,
      "step": 590
    },
    {
      "epoch": 0.7342874922215308,
      "eval_loss": 1.7240839004516602,
      "eval_runtime": 46.2617,
      "eval_samples_per_second": 21.616,
      "eval_steps_per_second": 0.908,
      "step": 590
    },
    {
      "epoch": 0.7355320472930927,
      "grad_norm": 2.833481788635254,
      "learning_rate": 4.22739852398524e-06,
      "loss": 1.545,
      "step": 591
    },
    {
      "epoch": 0.7367766023646546,
      "grad_norm": 2.7287282943725586,
      "learning_rate": 4.2250922509225096e-06,
      "loss": 1.5795,
      "step": 592
    },
    {
      "epoch": 0.7380211574362165,
      "grad_norm": 2.847174882888794,
      "learning_rate": 4.222785977859779e-06,
      "loss": 1.5469,
      "step": 593
    },
    {
      "epoch": 0.7392657125077785,
      "grad_norm": 2.9214589595794678,
      "learning_rate": 4.220479704797048e-06,
      "loss": 1.5791,
      "step": 594
    },
    {
      "epoch": 0.7405102675793404,
      "grad_norm": 2.7506580352783203,
      "learning_rate": 4.218173431734318e-06,
      "loss": 1.6578,
      "step": 595
    },
    {
      "epoch": 0.7417548226509023,
      "grad_norm": 2.7006874084472656,
      "learning_rate": 4.215867158671587e-06,
      "loss": 1.5598,
      "step": 596
    },
    {
      "epoch": 0.7429993777224643,
      "grad_norm": 2.795417547225952,
      "learning_rate": 4.213560885608857e-06,
      "loss": 1.5862,
      "step": 597
    },
    {
      "epoch": 0.7442439327940261,
      "grad_norm": 2.7418057918548584,
      "learning_rate": 4.211254612546125e-06,
      "loss": 1.5717,
      "step": 598
    },
    {
      "epoch": 0.7454884878655881,
      "grad_norm": 2.884915590286255,
      "learning_rate": 4.2089483394833955e-06,
      "loss": 1.5828,
      "step": 599
    },
    {
      "epoch": 0.7467330429371499,
      "grad_norm": 2.752328395843506,
      "learning_rate": 4.206642066420665e-06,
      "loss": 1.576,
      "step": 600
    },
    {
      "epoch": 0.7467330429371499,
      "eval_loss": 1.7281302213668823,
      "eval_runtime": 45.1978,
      "eval_samples_per_second": 22.125,
      "eval_steps_per_second": 0.929,
      "step": 600
    },
    {
      "epoch": 0.7479775980087119,
      "grad_norm": 2.9103147983551025,
      "learning_rate": 4.2043357933579335e-06,
      "loss": 1.4959,
      "step": 601
    },
    {
      "epoch": 0.7492221530802738,
      "grad_norm": 2.5317649841308594,
      "learning_rate": 4.202029520295203e-06,
      "loss": 1.5092,
      "step": 602
    },
    {
      "epoch": 0.7504667081518357,
      "grad_norm": 2.7383415699005127,
      "learning_rate": 4.199723247232473e-06,
      "loss": 1.4696,
      "step": 603
    },
    {
      "epoch": 0.7517112632233977,
      "grad_norm": 2.890305519104004,
      "learning_rate": 4.197416974169742e-06,
      "loss": 1.511,
      "step": 604
    },
    {
      "epoch": 0.7529558182949595,
      "grad_norm": 2.7425734996795654,
      "learning_rate": 4.195110701107011e-06,
      "loss": 1.5658,
      "step": 605
    },
    {
      "epoch": 0.7542003733665215,
      "grad_norm": 2.657726764678955,
      "learning_rate": 4.192804428044281e-06,
      "loss": 1.5648,
      "step": 606
    },
    {
      "epoch": 0.7554449284380834,
      "grad_norm": 3.181201696395874,
      "learning_rate": 4.19049815498155e-06,
      "loss": 1.583,
      "step": 607
    },
    {
      "epoch": 0.7566894835096453,
      "grad_norm": 2.665555715560913,
      "learning_rate": 4.1881918819188195e-06,
      "loss": 1.5439,
      "step": 608
    },
    {
      "epoch": 0.7579340385812072,
      "grad_norm": 2.7328009605407715,
      "learning_rate": 4.185885608856089e-06,
      "loss": 1.5264,
      "step": 609
    },
    {
      "epoch": 0.7591785936527692,
      "grad_norm": 2.7027499675750732,
      "learning_rate": 4.183579335793358e-06,
      "loss": 1.5327,
      "step": 610
    },
    {
      "epoch": 0.7591785936527692,
      "eval_loss": 1.7161823511123657,
      "eval_runtime": 42.4811,
      "eval_samples_per_second": 23.54,
      "eval_steps_per_second": 0.989,
      "step": 610
    },
    {
      "epoch": 0.760423148724331,
      "grad_norm": 2.6982595920562744,
      "learning_rate": 4.181273062730628e-06,
      "loss": 1.585,
      "step": 611
    },
    {
      "epoch": 0.761667703795893,
      "grad_norm": 2.780768394470215,
      "learning_rate": 4.178966789667897e-06,
      "loss": 1.5754,
      "step": 612
    },
    {
      "epoch": 0.7629122588674548,
      "grad_norm": 2.7140064239501953,
      "learning_rate": 4.176660516605167e-06,
      "loss": 1.5649,
      "step": 613
    },
    {
      "epoch": 0.7641568139390168,
      "grad_norm": 2.60182523727417,
      "learning_rate": 4.174354243542435e-06,
      "loss": 1.5879,
      "step": 614
    },
    {
      "epoch": 0.7654013690105788,
      "grad_norm": 2.6529550552368164,
      "learning_rate": 4.1720479704797054e-06,
      "loss": 1.5418,
      "step": 615
    },
    {
      "epoch": 0.7666459240821406,
      "grad_norm": 2.7936999797821045,
      "learning_rate": 4.169741697416975e-06,
      "loss": 1.5993,
      "step": 616
    },
    {
      "epoch": 0.7678904791537026,
      "grad_norm": 2.7894914150238037,
      "learning_rate": 4.1674354243542434e-06,
      "loss": 1.5962,
      "step": 617
    },
    {
      "epoch": 0.7691350342252644,
      "grad_norm": 2.8094332218170166,
      "learning_rate": 4.165129151291513e-06,
      "loss": 1.5898,
      "step": 618
    },
    {
      "epoch": 0.7703795892968264,
      "grad_norm": 2.9678053855895996,
      "learning_rate": 4.162822878228783e-06,
      "loss": 1.5602,
      "step": 619
    },
    {
      "epoch": 0.7716241443683883,
      "grad_norm": 2.7238247394561768,
      "learning_rate": 4.1605166051660526e-06,
      "loss": 1.554,
      "step": 620
    },
    {
      "epoch": 0.7716241443683883,
      "eval_loss": 1.720330834388733,
      "eval_runtime": 47.0576,
      "eval_samples_per_second": 21.251,
      "eval_steps_per_second": 0.893,
      "step": 620
    },
    {
      "epoch": 0.7728686994399502,
      "grad_norm": 2.6484885215759277,
      "learning_rate": 4.158210332103321e-06,
      "loss": 1.5464,
      "step": 621
    },
    {
      "epoch": 0.7741132545115121,
      "grad_norm": 2.722929000854492,
      "learning_rate": 4.1559040590405906e-06,
      "loss": 1.5471,
      "step": 622
    },
    {
      "epoch": 0.7753578095830741,
      "grad_norm": 2.842886209487915,
      "learning_rate": 4.15359778597786e-06,
      "loss": 1.5512,
      "step": 623
    },
    {
      "epoch": 0.776602364654636,
      "grad_norm": 2.647134780883789,
      "learning_rate": 4.151291512915129e-06,
      "loss": 1.5264,
      "step": 624
    },
    {
      "epoch": 0.7778469197261979,
      "grad_norm": 2.763277053833008,
      "learning_rate": 4.148985239852399e-06,
      "loss": 1.5102,
      "step": 625
    },
    {
      "epoch": 0.7790914747977598,
      "grad_norm": 2.9078922271728516,
      "learning_rate": 4.146678966789668e-06,
      "loss": 1.5538,
      "step": 626
    },
    {
      "epoch": 0.7803360298693217,
      "grad_norm": 2.60658860206604,
      "learning_rate": 4.144372693726938e-06,
      "loss": 1.486,
      "step": 627
    },
    {
      "epoch": 0.7815805849408837,
      "grad_norm": 2.5513789653778076,
      "learning_rate": 4.142066420664207e-06,
      "loss": 1.5323,
      "step": 628
    },
    {
      "epoch": 0.7828251400124455,
      "grad_norm": 3.0412704944610596,
      "learning_rate": 4.1397601476014765e-06,
      "loss": 1.5835,
      "step": 629
    },
    {
      "epoch": 0.7840696950840075,
      "grad_norm": 2.9483704566955566,
      "learning_rate": 4.137453874538745e-06,
      "loss": 1.5849,
      "step": 630
    },
    {
      "epoch": 0.7840696950840075,
      "eval_loss": 1.7138603925704956,
      "eval_runtime": 47.7618,
      "eval_samples_per_second": 20.937,
      "eval_steps_per_second": 0.879,
      "step": 630
    },
    {
      "epoch": 0.7853142501555694,
      "grad_norm": 2.784747362136841,
      "learning_rate": 4.135147601476015e-06,
      "loss": 1.5625,
      "step": 631
    },
    {
      "epoch": 0.7865588052271313,
      "grad_norm": 2.775791883468628,
      "learning_rate": 4.132841328413285e-06,
      "loss": 1.5827,
      "step": 632
    },
    {
      "epoch": 0.7878033602986932,
      "grad_norm": 2.867126226425171,
      "learning_rate": 4.130535055350554e-06,
      "loss": 1.5174,
      "step": 633
    },
    {
      "epoch": 0.7890479153702551,
      "grad_norm": 2.7829818725585938,
      "learning_rate": 4.128228782287823e-06,
      "loss": 1.5853,
      "step": 634
    },
    {
      "epoch": 0.790292470441817,
      "grad_norm": 2.630842924118042,
      "learning_rate": 4.125922509225092e-06,
      "loss": 1.5853,
      "step": 635
    },
    {
      "epoch": 0.791537025513379,
      "grad_norm": 2.8100547790527344,
      "learning_rate": 4.1236162361623625e-06,
      "loss": 1.5162,
      "step": 636
    },
    {
      "epoch": 0.7927815805849409,
      "grad_norm": 2.5983335971832275,
      "learning_rate": 4.121309963099631e-06,
      "loss": 1.5926,
      "step": 637
    },
    {
      "epoch": 0.7940261356565028,
      "grad_norm": 2.826624870300293,
      "learning_rate": 4.1190036900369005e-06,
      "loss": 1.5527,
      "step": 638
    },
    {
      "epoch": 0.7952706907280647,
      "grad_norm": 2.863215684890747,
      "learning_rate": 4.11669741697417e-06,
      "loss": 1.5514,
      "step": 639
    },
    {
      "epoch": 0.7965152457996266,
      "grad_norm": 2.7492613792419434,
      "learning_rate": 4.114391143911439e-06,
      "loss": 1.5101,
      "step": 640
    },
    {
      "epoch": 0.7965152457996266,
      "eval_loss": 1.7091425657272339,
      "eval_runtime": 46.5279,
      "eval_samples_per_second": 21.492,
      "eval_steps_per_second": 0.903,
      "step": 640
    },
    {
      "epoch": 0.7977598008711886,
      "grad_norm": 2.9577860832214355,
      "learning_rate": 4.112084870848709e-06,
      "loss": 1.5496,
      "step": 641
    },
    {
      "epoch": 0.7990043559427504,
      "grad_norm": 2.612668991088867,
      "learning_rate": 4.109778597785978e-06,
      "loss": 1.5204,
      "step": 642
    },
    {
      "epoch": 0.8002489110143124,
      "grad_norm": 2.5427377223968506,
      "learning_rate": 4.107472324723248e-06,
      "loss": 1.5038,
      "step": 643
    },
    {
      "epoch": 0.8014934660858744,
      "grad_norm": 2.7057406902313232,
      "learning_rate": 4.105166051660517e-06,
      "loss": 1.5831,
      "step": 644
    },
    {
      "epoch": 0.8027380211574362,
      "grad_norm": 2.664353609085083,
      "learning_rate": 4.1028597785977864e-06,
      "loss": 1.555,
      "step": 645
    },
    {
      "epoch": 0.8039825762289982,
      "grad_norm": 2.5452728271484375,
      "learning_rate": 4.100553505535056e-06,
      "loss": 1.5847,
      "step": 646
    },
    {
      "epoch": 0.80522713130056,
      "grad_norm": 2.5370709896087646,
      "learning_rate": 4.098247232472325e-06,
      "loss": 1.5396,
      "step": 647
    },
    {
      "epoch": 0.806471686372122,
      "grad_norm": 2.730109930038452,
      "learning_rate": 4.095940959409595e-06,
      "loss": 1.5234,
      "step": 648
    },
    {
      "epoch": 0.8077162414436839,
      "grad_norm": 2.586108446121216,
      "learning_rate": 4.093634686346864e-06,
      "loss": 1.5885,
      "step": 649
    },
    {
      "epoch": 0.8089607965152458,
      "grad_norm": 2.557053804397583,
      "learning_rate": 4.091328413284133e-06,
      "loss": 1.483,
      "step": 650
    },
    {
      "epoch": 0.8089607965152458,
      "eval_loss": 1.7204149961471558,
      "eval_runtime": 48.0041,
      "eval_samples_per_second": 20.832,
      "eval_steps_per_second": 0.875,
      "step": 650
    },
    {
      "epoch": 0.8102053515868077,
      "grad_norm": 2.673672676086426,
      "learning_rate": 4.089022140221402e-06,
      "loss": 1.4837,
      "step": 651
    },
    {
      "epoch": 0.8114499066583696,
      "grad_norm": 2.6465933322906494,
      "learning_rate": 4.086715867158672e-06,
      "loss": 1.5219,
      "step": 652
    },
    {
      "epoch": 0.8126944617299315,
      "grad_norm": 2.6897783279418945,
      "learning_rate": 4.084409594095941e-06,
      "loss": 1.5116,
      "step": 653
    },
    {
      "epoch": 0.8139390168014935,
      "grad_norm": 2.644369602203369,
      "learning_rate": 4.08210332103321e-06,
      "loss": 1.5141,
      "step": 654
    },
    {
      "epoch": 0.8151835718730553,
      "grad_norm": 2.7475595474243164,
      "learning_rate": 4.07979704797048e-06,
      "loss": 1.5509,
      "step": 655
    },
    {
      "epoch": 0.8164281269446173,
      "grad_norm": 2.72043514251709,
      "learning_rate": 4.077490774907749e-06,
      "loss": 1.6098,
      "step": 656
    },
    {
      "epoch": 0.8176726820161793,
      "grad_norm": 2.648069381713867,
      "learning_rate": 4.075184501845019e-06,
      "loss": 1.5368,
      "step": 657
    },
    {
      "epoch": 0.8189172370877411,
      "grad_norm": 2.644893169403076,
      "learning_rate": 4.072878228782288e-06,
      "loss": 1.492,
      "step": 658
    },
    {
      "epoch": 0.8201617921593031,
      "grad_norm": 2.9441068172454834,
      "learning_rate": 4.0705719557195575e-06,
      "loss": 1.5904,
      "step": 659
    },
    {
      "epoch": 0.8214063472308649,
      "grad_norm": 2.752913475036621,
      "learning_rate": 4.068265682656827e-06,
      "loss": 1.5544,
      "step": 660
    },
    {
      "epoch": 0.8214063472308649,
      "eval_loss": 1.709177017211914,
      "eval_runtime": 48.0401,
      "eval_samples_per_second": 20.816,
      "eval_steps_per_second": 0.874,
      "step": 660
    },
    {
      "epoch": 0.8226509023024269,
      "grad_norm": 2.6959197521209717,
      "learning_rate": 4.065959409594096e-06,
      "loss": 1.5553,
      "step": 661
    },
    {
      "epoch": 0.8238954573739888,
      "grad_norm": 2.8757998943328857,
      "learning_rate": 4.063653136531366e-06,
      "loss": 1.554,
      "step": 662
    },
    {
      "epoch": 0.8251400124455507,
      "grad_norm": 2.7723867893218994,
      "learning_rate": 4.061346863468635e-06,
      "loss": 1.5499,
      "step": 663
    },
    {
      "epoch": 0.8263845675171126,
      "grad_norm": 2.7620463371276855,
      "learning_rate": 4.059040590405905e-06,
      "loss": 1.5962,
      "step": 664
    },
    {
      "epoch": 0.8276291225886746,
      "grad_norm": 2.8601174354553223,
      "learning_rate": 4.056734317343174e-06,
      "loss": 1.5218,
      "step": 665
    },
    {
      "epoch": 0.8288736776602365,
      "grad_norm": 2.81356143951416,
      "learning_rate": 4.054428044280443e-06,
      "loss": 1.5382,
      "step": 666
    },
    {
      "epoch": 0.8301182327317984,
      "grad_norm": 2.904679298400879,
      "learning_rate": 4.052121771217712e-06,
      "loss": 1.5314,
      "step": 667
    },
    {
      "epoch": 0.8313627878033603,
      "grad_norm": 2.798236846923828,
      "learning_rate": 4.049815498154982e-06,
      "loss": 1.5989,
      "step": 668
    },
    {
      "epoch": 0.8326073428749222,
      "grad_norm": 2.8923728466033936,
      "learning_rate": 4.047509225092252e-06,
      "loss": 1.5475,
      "step": 669
    },
    {
      "epoch": 0.8338518979464842,
      "grad_norm": 2.6134703159332275,
      "learning_rate": 4.04520295202952e-06,
      "loss": 1.5232,
      "step": 670
    },
    {
      "epoch": 0.8338518979464842,
      "eval_loss": 1.7080007791519165,
      "eval_runtime": 52.6531,
      "eval_samples_per_second": 18.992,
      "eval_steps_per_second": 0.798,
      "step": 670
    },
    {
      "epoch": 0.835096453018046,
      "grad_norm": 2.6788907051086426,
      "learning_rate": 4.04289667896679e-06,
      "loss": 1.5437,
      "step": 671
    },
    {
      "epoch": 0.836341008089608,
      "grad_norm": 2.872429847717285,
      "learning_rate": 4.04059040590406e-06,
      "loss": 1.5368,
      "step": 672
    },
    {
      "epoch": 0.8375855631611698,
      "grad_norm": 2.714857578277588,
      "learning_rate": 4.038284132841329e-06,
      "loss": 1.5404,
      "step": 673
    },
    {
      "epoch": 0.8388301182327318,
      "grad_norm": 2.6955931186676025,
      "learning_rate": 4.035977859778598e-06,
      "loss": 1.5501,
      "step": 674
    },
    {
      "epoch": 0.8400746733042938,
      "grad_norm": 2.7090582847595215,
      "learning_rate": 4.0336715867158674e-06,
      "loss": 1.5068,
      "step": 675
    },
    {
      "epoch": 0.8413192283758556,
      "grad_norm": 2.8406383991241455,
      "learning_rate": 4.031365313653137e-06,
      "loss": 1.5529,
      "step": 676
    },
    {
      "epoch": 0.8425637834474176,
      "grad_norm": 2.965557098388672,
      "learning_rate": 4.029059040590406e-06,
      "loss": 1.5847,
      "step": 677
    },
    {
      "epoch": 0.8438083385189795,
      "grad_norm": 2.664083242416382,
      "learning_rate": 4.026752767527676e-06,
      "loss": 1.5701,
      "step": 678
    },
    {
      "epoch": 0.8450528935905414,
      "grad_norm": 2.7362401485443115,
      "learning_rate": 4.024446494464945e-06,
      "loss": 1.5496,
      "step": 679
    },
    {
      "epoch": 0.8462974486621033,
      "grad_norm": 2.695342540740967,
      "learning_rate": 4.0221402214022145e-06,
      "loss": 1.5009,
      "step": 680
    },
    {
      "epoch": 0.8462974486621033,
      "eval_loss": 1.7063637971878052,
      "eval_runtime": 52.7749,
      "eval_samples_per_second": 18.948,
      "eval_steps_per_second": 0.796,
      "step": 680
    },
    {
      "epoch": 0.8475420037336652,
      "grad_norm": 2.615994453430176,
      "learning_rate": 4.019833948339484e-06,
      "loss": 1.5281,
      "step": 681
    },
    {
      "epoch": 0.8487865588052271,
      "grad_norm": 3.065732002258301,
      "learning_rate": 4.017527675276753e-06,
      "loss": 1.5616,
      "step": 682
    },
    {
      "epoch": 0.8500311138767891,
      "grad_norm": 2.812166690826416,
      "learning_rate": 4.015221402214022e-06,
      "loss": 1.5573,
      "step": 683
    },
    {
      "epoch": 0.8512756689483509,
      "grad_norm": 2.745252847671509,
      "learning_rate": 4.012915129151292e-06,
      "loss": 1.5426,
      "step": 684
    },
    {
      "epoch": 0.8525202240199129,
      "grad_norm": 2.862264633178711,
      "learning_rate": 4.010608856088562e-06,
      "loss": 1.4973,
      "step": 685
    },
    {
      "epoch": 0.8537647790914747,
      "grad_norm": 2.752469062805176,
      "learning_rate": 4.00830258302583e-06,
      "loss": 1.5416,
      "step": 686
    },
    {
      "epoch": 0.8550093341630367,
      "grad_norm": 2.5692927837371826,
      "learning_rate": 4.0059963099631e-06,
      "loss": 1.5021,
      "step": 687
    },
    {
      "epoch": 0.8562538892345987,
      "grad_norm": 2.896444320678711,
      "learning_rate": 4.003690036900369e-06,
      "loss": 1.5794,
      "step": 688
    },
    {
      "epoch": 0.8574984443061605,
      "grad_norm": 2.7448434829711914,
      "learning_rate": 4.0013837638376385e-06,
      "loss": 1.5138,
      "step": 689
    },
    {
      "epoch": 0.8587429993777225,
      "grad_norm": 2.8280975818634033,
      "learning_rate": 3.999077490774908e-06,
      "loss": 1.5071,
      "step": 690
    },
    {
      "epoch": 0.8587429993777225,
      "eval_loss": 1.7048547267913818,
      "eval_runtime": 52.8015,
      "eval_samples_per_second": 18.939,
      "eval_steps_per_second": 0.795,
      "step": 690
    },
    {
      "epoch": 0.8599875544492844,
      "grad_norm": 2.701293706893921,
      "learning_rate": 3.996771217712177e-06,
      "loss": 1.5457,
      "step": 691
    },
    {
      "epoch": 0.8612321095208463,
      "grad_norm": 2.673452138900757,
      "learning_rate": 3.994464944649447e-06,
      "loss": 1.5355,
      "step": 692
    },
    {
      "epoch": 0.8624766645924082,
      "grad_norm": 2.7728021144866943,
      "learning_rate": 3.992158671586716e-06,
      "loss": 1.5486,
      "step": 693
    },
    {
      "epoch": 0.8637212196639701,
      "grad_norm": 2.747614860534668,
      "learning_rate": 3.989852398523986e-06,
      "loss": 1.5771,
      "step": 694
    },
    {
      "epoch": 0.864965774735532,
      "grad_norm": 2.6240482330322266,
      "learning_rate": 3.987546125461255e-06,
      "loss": 1.5338,
      "step": 695
    },
    {
      "epoch": 0.866210329807094,
      "grad_norm": 2.8253114223480225,
      "learning_rate": 3.9852398523985245e-06,
      "loss": 1.5029,
      "step": 696
    },
    {
      "epoch": 0.8674548848786559,
      "grad_norm": 2.7980496883392334,
      "learning_rate": 3.982933579335794e-06,
      "loss": 1.5239,
      "step": 697
    },
    {
      "epoch": 0.8686994399502178,
      "grad_norm": 2.937443971633911,
      "learning_rate": 3.980627306273063e-06,
      "loss": 1.5406,
      "step": 698
    },
    {
      "epoch": 0.8699439950217797,
      "grad_norm": 2.714353322982788,
      "learning_rate": 3.978321033210332e-06,
      "loss": 1.5724,
      "step": 699
    },
    {
      "epoch": 0.8711885500933416,
      "grad_norm": 2.7986929416656494,
      "learning_rate": 3.976014760147602e-06,
      "loss": 1.5687,
      "step": 700
    },
    {
      "epoch": 0.8711885500933416,
      "eval_loss": 1.6998120546340942,
      "eval_runtime": 46.4448,
      "eval_samples_per_second": 21.531,
      "eval_steps_per_second": 0.904,
      "step": 700
    },
    {
      "epoch": 0.8724331051649036,
      "grad_norm": 2.755740165710449,
      "learning_rate": 3.973708487084872e-06,
      "loss": 1.5047,
      "step": 701
    },
    {
      "epoch": 0.8736776602364654,
      "grad_norm": 2.760399341583252,
      "learning_rate": 3.97140221402214e-06,
      "loss": 1.5406,
      "step": 702
    },
    {
      "epoch": 0.8749222153080274,
      "grad_norm": 2.780214786529541,
      "learning_rate": 3.96909594095941e-06,
      "loss": 1.551,
      "step": 703
    },
    {
      "epoch": 0.8761667703795893,
      "grad_norm": 2.669801950454712,
      "learning_rate": 3.966789667896679e-06,
      "loss": 1.5578,
      "step": 704
    },
    {
      "epoch": 0.8774113254511512,
      "grad_norm": 2.729586362838745,
      "learning_rate": 3.964483394833948e-06,
      "loss": 1.4808,
      "step": 705
    },
    {
      "epoch": 0.8786558805227132,
      "grad_norm": 2.608551263809204,
      "learning_rate": 3.962177121771218e-06,
      "loss": 1.4649,
      "step": 706
    },
    {
      "epoch": 0.879900435594275,
      "grad_norm": 3.0885748863220215,
      "learning_rate": 3.959870848708487e-06,
      "loss": 1.5602,
      "step": 707
    },
    {
      "epoch": 0.881144990665837,
      "grad_norm": 2.659917116165161,
      "learning_rate": 3.957564575645757e-06,
      "loss": 1.5341,
      "step": 708
    },
    {
      "epoch": 0.8823895457373989,
      "grad_norm": 2.718696355819702,
      "learning_rate": 3.955258302583026e-06,
      "loss": 1.5232,
      "step": 709
    },
    {
      "epoch": 0.8836341008089608,
      "grad_norm": 2.902811288833618,
      "learning_rate": 3.9529520295202955e-06,
      "loss": 1.517,
      "step": 710
    },
    {
      "epoch": 0.8836341008089608,
      "eval_loss": 1.7022311687469482,
      "eval_runtime": 43.1291,
      "eval_samples_per_second": 23.186,
      "eval_steps_per_second": 0.974,
      "step": 710
    },
    {
      "epoch": 0.8848786558805227,
      "grad_norm": 2.816802978515625,
      "learning_rate": 3.950645756457565e-06,
      "loss": 1.5097,
      "step": 711
    },
    {
      "epoch": 0.8861232109520847,
      "grad_norm": 2.8189518451690674,
      "learning_rate": 3.948339483394834e-06,
      "loss": 1.5842,
      "step": 712
    },
    {
      "epoch": 0.8873677660236465,
      "grad_norm": 2.879767417907715,
      "learning_rate": 3.946033210332104e-06,
      "loss": 1.5484,
      "step": 713
    },
    {
      "epoch": 0.8886123210952085,
      "grad_norm": 2.8263540267944336,
      "learning_rate": 3.943726937269373e-06,
      "loss": 1.5259,
      "step": 714
    },
    {
      "epoch": 0.8898568761667703,
      "grad_norm": 2.655139923095703,
      "learning_rate": 3.941420664206642e-06,
      "loss": 1.482,
      "step": 715
    },
    {
      "epoch": 0.8911014312383323,
      "grad_norm": 2.63429856300354,
      "learning_rate": 3.939114391143912e-06,
      "loss": 1.512,
      "step": 716
    },
    {
      "epoch": 0.8923459863098943,
      "grad_norm": 2.9241650104522705,
      "learning_rate": 3.9368081180811815e-06,
      "loss": 1.5347,
      "step": 717
    },
    {
      "epoch": 0.8935905413814561,
      "grad_norm": 2.8512847423553467,
      "learning_rate": 3.934501845018451e-06,
      "loss": 1.5791,
      "step": 718
    },
    {
      "epoch": 0.8948350964530181,
      "grad_norm": 2.5280821323394775,
      "learning_rate": 3.9321955719557195e-06,
      "loss": 1.4823,
      "step": 719
    },
    {
      "epoch": 0.8960796515245799,
      "grad_norm": 2.6618733406066895,
      "learning_rate": 3.929889298892989e-06,
      "loss": 1.4708,
      "step": 720
    },
    {
      "epoch": 0.8960796515245799,
      "eval_loss": 1.6904946565628052,
      "eval_runtime": 49.3149,
      "eval_samples_per_second": 20.278,
      "eval_steps_per_second": 0.852,
      "step": 720
    },
    {
      "epoch": 0.8973242065961419,
      "grad_norm": 2.8406248092651367,
      "learning_rate": 3.927583025830259e-06,
      "loss": 1.5425,
      "step": 721
    },
    {
      "epoch": 0.8985687616677038,
      "grad_norm": 2.807971715927124,
      "learning_rate": 3.925276752767528e-06,
      "loss": 1.5617,
      "step": 722
    },
    {
      "epoch": 0.8998133167392657,
      "grad_norm": 2.6703925132751465,
      "learning_rate": 3.922970479704797e-06,
      "loss": 1.5544,
      "step": 723
    },
    {
      "epoch": 0.9010578718108276,
      "grad_norm": 2.785083770751953,
      "learning_rate": 3.920664206642067e-06,
      "loss": 1.5696,
      "step": 724
    },
    {
      "epoch": 0.9023024268823896,
      "grad_norm": 2.7778568267822266,
      "learning_rate": 3.918357933579336e-06,
      "loss": 1.5262,
      "step": 725
    },
    {
      "epoch": 0.9035469819539514,
      "grad_norm": 2.6343681812286377,
      "learning_rate": 3.9160516605166055e-06,
      "loss": 1.5509,
      "step": 726
    },
    {
      "epoch": 0.9047915370255134,
      "grad_norm": 2.6923940181732178,
      "learning_rate": 3.913745387453875e-06,
      "loss": 1.5425,
      "step": 727
    },
    {
      "epoch": 0.9060360920970753,
      "grad_norm": 2.732797861099243,
      "learning_rate": 3.911439114391144e-06,
      "loss": 1.4524,
      "step": 728
    },
    {
      "epoch": 0.9072806471686372,
      "grad_norm": 2.6365227699279785,
      "learning_rate": 3.909132841328414e-06,
      "loss": 1.5617,
      "step": 729
    },
    {
      "epoch": 0.9085252022401992,
      "grad_norm": 2.646639823913574,
      "learning_rate": 3.906826568265683e-06,
      "loss": 1.5122,
      "step": 730
    },
    {
      "epoch": 0.9085252022401992,
      "eval_loss": 1.6920651197433472,
      "eval_runtime": 48.5914,
      "eval_samples_per_second": 20.58,
      "eval_steps_per_second": 0.864,
      "step": 730
    },
    {
      "epoch": 0.909769757311761,
      "grad_norm": 2.5834903717041016,
      "learning_rate": 3.9045202952029526e-06,
      "loss": 1.5095,
      "step": 731
    },
    {
      "epoch": 0.911014312383323,
      "grad_norm": 2.4998092651367188,
      "learning_rate": 3.902214022140222e-06,
      "loss": 1.4376,
      "step": 732
    },
    {
      "epoch": 0.9122588674548848,
      "grad_norm": 2.5143818855285645,
      "learning_rate": 3.899907749077491e-06,
      "loss": 1.5222,
      "step": 733
    },
    {
      "epoch": 0.9135034225264468,
      "grad_norm": 2.7795114517211914,
      "learning_rate": 3.897601476014761e-06,
      "loss": 1.5498,
      "step": 734
    },
    {
      "epoch": 0.9147479775980087,
      "grad_norm": 2.819859027862549,
      "learning_rate": 3.895295202952029e-06,
      "loss": 1.5288,
      "step": 735
    },
    {
      "epoch": 0.9159925326695706,
      "grad_norm": 2.663306951522827,
      "learning_rate": 3.892988929889299e-06,
      "loss": 1.4517,
      "step": 736
    },
    {
      "epoch": 0.9172370877411326,
      "grad_norm": 2.775521993637085,
      "learning_rate": 3.890682656826569e-06,
      "loss": 1.529,
      "step": 737
    },
    {
      "epoch": 0.9184816428126945,
      "grad_norm": 2.761237859725952,
      "learning_rate": 3.888376383763838e-06,
      "loss": 1.4781,
      "step": 738
    },
    {
      "epoch": 0.9197261978842564,
      "grad_norm": 2.5410993099212646,
      "learning_rate": 3.886070110701107e-06,
      "loss": 1.4205,
      "step": 739
    },
    {
      "epoch": 0.9209707529558183,
      "grad_norm": 2.6289987564086914,
      "learning_rate": 3.8837638376383765e-06,
      "loss": 1.5253,
      "step": 740
    },
    {
      "epoch": 0.9209707529558183,
      "eval_loss": 1.700616717338562,
      "eval_runtime": 46.1038,
      "eval_samples_per_second": 21.69,
      "eval_steps_per_second": 0.911,
      "step": 740
    },
    {
      "epoch": 0.9222153080273802,
      "grad_norm": 2.694406270980835,
      "learning_rate": 3.881457564575646e-06,
      "loss": 1.5335,
      "step": 741
    },
    {
      "epoch": 0.9234598630989421,
      "grad_norm": 2.597160577774048,
      "learning_rate": 3.879151291512915e-06,
      "loss": 1.5026,
      "step": 742
    },
    {
      "epoch": 0.9247044181705041,
      "grad_norm": 2.625049114227295,
      "learning_rate": 3.876845018450185e-06,
      "loss": 1.4885,
      "step": 743
    },
    {
      "epoch": 0.9259489732420659,
      "grad_norm": 2.5983152389526367,
      "learning_rate": 3.874538745387454e-06,
      "loss": 1.5015,
      "step": 744
    },
    {
      "epoch": 0.9271935283136279,
      "grad_norm": 2.6912894248962402,
      "learning_rate": 3.872232472324724e-06,
      "loss": 1.5179,
      "step": 745
    },
    {
      "epoch": 0.9284380833851898,
      "grad_norm": 2.699328899383545,
      "learning_rate": 3.869926199261993e-06,
      "loss": 1.5571,
      "step": 746
    },
    {
      "epoch": 0.9296826384567517,
      "grad_norm": 2.7117769718170166,
      "learning_rate": 3.8676199261992625e-06,
      "loss": 1.5225,
      "step": 747
    },
    {
      "epoch": 0.9309271935283137,
      "grad_norm": 2.785900354385376,
      "learning_rate": 3.865313653136532e-06,
      "loss": 1.5571,
      "step": 748
    },
    {
      "epoch": 0.9321717485998755,
      "grad_norm": 2.6901021003723145,
      "learning_rate": 3.863007380073801e-06,
      "loss": 1.5108,
      "step": 749
    },
    {
      "epoch": 0.9334163036714375,
      "grad_norm": 2.55047607421875,
      "learning_rate": 3.860701107011071e-06,
      "loss": 1.4934,
      "step": 750
    },
    {
      "epoch": 0.9334163036714375,
      "eval_loss": 1.693824291229248,
      "eval_runtime": 45.7403,
      "eval_samples_per_second": 21.863,
      "eval_steps_per_second": 0.918,
      "step": 750
    },
    {
      "epoch": 0.9346608587429994,
      "grad_norm": 2.7620902061462402,
      "learning_rate": 3.858394833948339e-06,
      "loss": 1.5129,
      "step": 751
    },
    {
      "epoch": 0.9359054138145613,
      "grad_norm": 2.798583984375,
      "learning_rate": 3.856088560885609e-06,
      "loss": 1.4986,
      "step": 752
    },
    {
      "epoch": 0.9371499688861232,
      "grad_norm": 2.5885400772094727,
      "learning_rate": 3.853782287822879e-06,
      "loss": 1.4779,
      "step": 753
    },
    {
      "epoch": 0.9383945239576851,
      "grad_norm": 2.8280391693115234,
      "learning_rate": 3.851476014760148e-06,
      "loss": 1.6156,
      "step": 754
    },
    {
      "epoch": 0.939639079029247,
      "grad_norm": 2.7616288661956787,
      "learning_rate": 3.849169741697417e-06,
      "loss": 1.5051,
      "step": 755
    },
    {
      "epoch": 0.940883634100809,
      "grad_norm": 2.598370313644409,
      "learning_rate": 3.8468634686346865e-06,
      "loss": 1.5569,
      "step": 756
    },
    {
      "epoch": 0.9421281891723708,
      "grad_norm": 2.649433135986328,
      "learning_rate": 3.844557195571956e-06,
      "loss": 1.515,
      "step": 757
    },
    {
      "epoch": 0.9433727442439328,
      "grad_norm": 2.568361759185791,
      "learning_rate": 3.842250922509225e-06,
      "loss": 1.4926,
      "step": 758
    },
    {
      "epoch": 0.9446172993154948,
      "grad_norm": 2.6539297103881836,
      "learning_rate": 3.839944649446495e-06,
      "loss": 1.4813,
      "step": 759
    },
    {
      "epoch": 0.9458618543870566,
      "grad_norm": 2.6790902614593506,
      "learning_rate": 3.837638376383764e-06,
      "loss": 1.5056,
      "step": 760
    },
    {
      "epoch": 0.9458618543870566,
      "eval_loss": 1.6913493871688843,
      "eval_runtime": 44.5061,
      "eval_samples_per_second": 22.469,
      "eval_steps_per_second": 0.944,
      "step": 760
    },
    {
      "epoch": 0.9471064094586186,
      "grad_norm": 2.6265010833740234,
      "learning_rate": 3.8353321033210336e-06,
      "loss": 1.5421,
      "step": 761
    },
    {
      "epoch": 0.9483509645301804,
      "grad_norm": 2.5852062702178955,
      "learning_rate": 3.833025830258303e-06,
      "loss": 1.5145,
      "step": 762
    },
    {
      "epoch": 0.9495955196017424,
      "grad_norm": 2.6579225063323975,
      "learning_rate": 3.830719557195572e-06,
      "loss": 1.4988,
      "step": 763
    },
    {
      "epoch": 0.9508400746733043,
      "grad_norm": 2.7944133281707764,
      "learning_rate": 3.828413284132842e-06,
      "loss": 1.5199,
      "step": 764
    },
    {
      "epoch": 0.9520846297448662,
      "grad_norm": 2.5813052654266357,
      "learning_rate": 3.826107011070111e-06,
      "loss": 1.5072,
      "step": 765
    },
    {
      "epoch": 0.9533291848164281,
      "grad_norm": 2.551501512527466,
      "learning_rate": 3.823800738007381e-06,
      "loss": 1.5348,
      "step": 766
    },
    {
      "epoch": 0.95457373988799,
      "grad_norm": 2.755141496658325,
      "learning_rate": 3.821494464944649e-06,
      "loss": 1.4787,
      "step": 767
    },
    {
      "epoch": 0.955818294959552,
      "grad_norm": 2.521984338760376,
      "learning_rate": 3.819188191881919e-06,
      "loss": 1.4778,
      "step": 768
    },
    {
      "epoch": 0.9570628500311139,
      "grad_norm": 2.6301770210266113,
      "learning_rate": 3.816881918819189e-06,
      "loss": 1.4703,
      "step": 769
    },
    {
      "epoch": 0.9583074051026758,
      "grad_norm": 2.972749710083008,
      "learning_rate": 3.814575645756458e-06,
      "loss": 1.5516,
      "step": 770
    },
    {
      "epoch": 0.9583074051026758,
      "eval_loss": 1.689251184463501,
      "eval_runtime": 44.3352,
      "eval_samples_per_second": 22.555,
      "eval_steps_per_second": 0.947,
      "step": 770
    },
    {
      "epoch": 0.9595519601742377,
      "grad_norm": 2.7247636318206787,
      "learning_rate": 3.812269372693727e-06,
      "loss": 1.5659,
      "step": 771
    },
    {
      "epoch": 0.9607965152457997,
      "grad_norm": 2.6853489875793457,
      "learning_rate": 3.809963099630997e-06,
      "loss": 1.5149,
      "step": 772
    },
    {
      "epoch": 0.9620410703173615,
      "grad_norm": 2.6767218112945557,
      "learning_rate": 3.8076568265682662e-06,
      "loss": 1.5573,
      "step": 773
    },
    {
      "epoch": 0.9632856253889235,
      "grad_norm": 2.5949296951293945,
      "learning_rate": 3.8053505535055352e-06,
      "loss": 1.517,
      "step": 774
    },
    {
      "epoch": 0.9645301804604853,
      "grad_norm": 2.841171979904175,
      "learning_rate": 3.8030442804428046e-06,
      "loss": 1.515,
      "step": 775
    },
    {
      "epoch": 0.9657747355320473,
      "grad_norm": 2.7451512813568115,
      "learning_rate": 3.800738007380074e-06,
      "loss": 1.516,
      "step": 776
    },
    {
      "epoch": 0.9670192906036092,
      "grad_norm": 2.6852564811706543,
      "learning_rate": 3.798431734317343e-06,
      "loss": 1.5187,
      "step": 777
    },
    {
      "epoch": 0.9682638456751711,
      "grad_norm": 2.72956919670105,
      "learning_rate": 3.796125461254613e-06,
      "loss": 1.5344,
      "step": 778
    },
    {
      "epoch": 0.9695084007467331,
      "grad_norm": 2.672933578491211,
      "learning_rate": 3.7938191881918823e-06,
      "loss": 1.5143,
      "step": 779
    },
    {
      "epoch": 0.9707529558182949,
      "grad_norm": 2.6565003395080566,
      "learning_rate": 3.7915129151291518e-06,
      "loss": 1.4759,
      "step": 780
    },
    {
      "epoch": 0.9707529558182949,
      "eval_loss": 1.6836074590682983,
      "eval_runtime": 44.0331,
      "eval_samples_per_second": 22.71,
      "eval_steps_per_second": 0.954,
      "step": 780
    },
    {
      "epoch": 0.9719975108898569,
      "grad_norm": 2.665921449661255,
      "learning_rate": 3.7892066420664208e-06,
      "loss": 1.4985,
      "step": 781
    },
    {
      "epoch": 0.9732420659614188,
      "grad_norm": 2.671757221221924,
      "learning_rate": 3.7869003690036906e-06,
      "loss": 1.5014,
      "step": 782
    },
    {
      "epoch": 0.9744866210329807,
      "grad_norm": 2.7320780754089355,
      "learning_rate": 3.78459409594096e-06,
      "loss": 1.5052,
      "step": 783
    },
    {
      "epoch": 0.9757311761045426,
      "grad_norm": 3.0772581100463867,
      "learning_rate": 3.782287822878229e-06,
      "loss": 1.5666,
      "step": 784
    },
    {
      "epoch": 0.9769757311761046,
      "grad_norm": 2.651129722595215,
      "learning_rate": 3.7799815498154984e-06,
      "loss": 1.4954,
      "step": 785
    },
    {
      "epoch": 0.9782202862476664,
      "grad_norm": 2.6942505836486816,
      "learning_rate": 3.777675276752768e-06,
      "loss": 1.5186,
      "step": 786
    },
    {
      "epoch": 0.9794648413192284,
      "grad_norm": 2.767474412918091,
      "learning_rate": 3.775369003690037e-06,
      "loss": 1.4699,
      "step": 787
    },
    {
      "epoch": 0.9807093963907902,
      "grad_norm": 2.728269100189209,
      "learning_rate": 3.7730627306273067e-06,
      "loss": 1.5176,
      "step": 788
    },
    {
      "epoch": 0.9819539514623522,
      "grad_norm": 2.697984218597412,
      "learning_rate": 3.770756457564576e-06,
      "loss": 1.5327,
      "step": 789
    },
    {
      "epoch": 0.9831985065339142,
      "grad_norm": 2.8172037601470947,
      "learning_rate": 3.768450184501845e-06,
      "loss": 1.4996,
      "step": 790
    },
    {
      "epoch": 0.9831985065339142,
      "eval_loss": 1.7013198137283325,
      "eval_runtime": 44.5233,
      "eval_samples_per_second": 22.46,
      "eval_steps_per_second": 0.943,
      "step": 790
    },
    {
      "epoch": 0.984443061605476,
      "grad_norm": 2.8202197551727295,
      "learning_rate": 3.7661439114391146e-06,
      "loss": 1.5193,
      "step": 791
    },
    {
      "epoch": 0.985687616677038,
      "grad_norm": 2.764653444290161,
      "learning_rate": 3.763837638376384e-06,
      "loss": 1.5544,
      "step": 792
    },
    {
      "epoch": 0.9869321717485999,
      "grad_norm": 2.6528818607330322,
      "learning_rate": 3.761531365313654e-06,
      "loss": 1.5328,
      "step": 793
    },
    {
      "epoch": 0.9881767268201618,
      "grad_norm": 2.643666982650757,
      "learning_rate": 3.759225092250923e-06,
      "loss": 1.5279,
      "step": 794
    },
    {
      "epoch": 0.9894212818917237,
      "grad_norm": 2.631007432937622,
      "learning_rate": 3.7569188191881922e-06,
      "loss": 1.5135,
      "step": 795
    },
    {
      "epoch": 0.9906658369632856,
      "grad_norm": 2.6622400283813477,
      "learning_rate": 3.7546125461254617e-06,
      "loss": 1.5099,
      "step": 796
    },
    {
      "epoch": 0.9919103920348475,
      "grad_norm": 2.5758702754974365,
      "learning_rate": 3.7523062730627307e-06,
      "loss": 1.4691,
      "step": 797
    },
    {
      "epoch": 0.9931549471064095,
      "grad_norm": 2.822147846221924,
      "learning_rate": 3.7500000000000005e-06,
      "loss": 1.4846,
      "step": 798
    },
    {
      "epoch": 0.9943995021779714,
      "grad_norm": 3.166783571243286,
      "learning_rate": 3.74769372693727e-06,
      "loss": 1.5586,
      "step": 799
    },
    {
      "epoch": 0.9956440572495333,
      "grad_norm": 2.931654691696167,
      "learning_rate": 3.745387453874539e-06,
      "loss": 1.4773,
      "step": 800
    },
    {
      "epoch": 0.9956440572495333,
      "eval_loss": 1.6904431581497192,
      "eval_runtime": 43.927,
      "eval_samples_per_second": 22.765,
      "eval_steps_per_second": 0.956,
      "step": 800
    },
    {
      "epoch": 0.9968886123210952,
      "grad_norm": 2.97959303855896,
      "learning_rate": 3.7430811808118084e-06,
      "loss": 1.4555,
      "step": 801
    },
    {
      "epoch": 0.9981331673926571,
      "grad_norm": 2.796096086502075,
      "learning_rate": 3.7407749077490778e-06,
      "loss": 1.5243,
      "step": 802
    },
    {
      "epoch": 0.9993777224642191,
      "grad_norm": 2.8529677391052246,
      "learning_rate": 3.7384686346863468e-06,
      "loss": 1.4919,
      "step": 803
    },
    {
      "epoch": 1.000622277535781,
      "grad_norm": 2.9269766807556152,
      "learning_rate": 3.7361623616236166e-06,
      "loss": 1.5188,
      "step": 804
    },
    {
      "epoch": 1.0018668326073428,
      "grad_norm": 2.8332459926605225,
      "learning_rate": 3.733856088560886e-06,
      "loss": 1.499,
      "step": 805
    },
    {
      "epoch": 1.0031113876789048,
      "grad_norm": 2.783031702041626,
      "learning_rate": 3.7315498154981555e-06,
      "loss": 1.4383,
      "step": 806
    },
    {
      "epoch": 1.0043559427504667,
      "grad_norm": 3.097334623336792,
      "learning_rate": 3.7292435424354245e-06,
      "loss": 1.5,
      "step": 807
    },
    {
      "epoch": 1.0056004978220285,
      "grad_norm": 3.1136531829833984,
      "learning_rate": 3.726937269372694e-06,
      "loss": 1.5161,
      "step": 808
    },
    {
      "epoch": 1.0068450528935906,
      "grad_norm": 2.8723881244659424,
      "learning_rate": 3.7246309963099637e-06,
      "loss": 1.4716,
      "step": 809
    },
    {
      "epoch": 1.0080896079651525,
      "grad_norm": 2.798966407775879,
      "learning_rate": 3.7223247232472327e-06,
      "loss": 1.4874,
      "step": 810
    },
    {
      "epoch": 1.0080896079651525,
      "eval_loss": 1.6889101266860962,
      "eval_runtime": 42.0407,
      "eval_samples_per_second": 23.786,
      "eval_steps_per_second": 0.999,
      "step": 810
    },
    {
      "epoch": 1.0093341630367143,
      "grad_norm": 2.731884241104126,
      "learning_rate": 3.720018450184502e-06,
      "loss": 1.443,
      "step": 811
    },
    {
      "epoch": 1.0105787181082764,
      "grad_norm": 2.651594877243042,
      "learning_rate": 3.7177121771217716e-06,
      "loss": 1.4537,
      "step": 812
    },
    {
      "epoch": 1.0118232731798382,
      "grad_norm": 2.8403851985931396,
      "learning_rate": 3.7154059040590406e-06,
      "loss": 1.4801,
      "step": 813
    },
    {
      "epoch": 1.0130678282514,
      "grad_norm": 2.8779821395874023,
      "learning_rate": 3.71309963099631e-06,
      "loss": 1.4716,
      "step": 814
    },
    {
      "epoch": 1.0143123833229621,
      "grad_norm": 2.8755605220794678,
      "learning_rate": 3.71079335793358e-06,
      "loss": 1.471,
      "step": 815
    },
    {
      "epoch": 1.015556938394524,
      "grad_norm": 2.9748878479003906,
      "learning_rate": 3.708487084870849e-06,
      "loss": 1.4988,
      "step": 816
    },
    {
      "epoch": 1.0168014934660858,
      "grad_norm": 2.9609413146972656,
      "learning_rate": 3.7061808118081183e-06,
      "loss": 1.5184,
      "step": 817
    },
    {
      "epoch": 1.0180460485376477,
      "grad_norm": 2.8808226585388184,
      "learning_rate": 3.7038745387453877e-06,
      "loss": 1.5161,
      "step": 818
    },
    {
      "epoch": 1.0192906036092098,
      "grad_norm": 3.0764009952545166,
      "learning_rate": 3.7015682656826576e-06,
      "loss": 1.4991,
      "step": 819
    },
    {
      "epoch": 1.0205351586807716,
      "grad_norm": 2.802974224090576,
      "learning_rate": 3.6992619926199266e-06,
      "loss": 1.4967,
      "step": 820
    },
    {
      "epoch": 1.0205351586807716,
      "eval_loss": 1.6831101179122925,
      "eval_runtime": 44.5974,
      "eval_samples_per_second": 22.423,
      "eval_steps_per_second": 0.942,
      "step": 820
    },
    {
      "epoch": 1.0217797137523335,
      "grad_norm": 2.8415472507476807,
      "learning_rate": 3.696955719557196e-06,
      "loss": 1.4636,
      "step": 821
    },
    {
      "epoch": 1.0230242688238955,
      "grad_norm": 2.830817461013794,
      "learning_rate": 3.6946494464944654e-06,
      "loss": 1.4648,
      "step": 822
    },
    {
      "epoch": 1.0242688238954574,
      "grad_norm": 2.993065357208252,
      "learning_rate": 3.6923431734317344e-06,
      "loss": 1.4923,
      "step": 823
    },
    {
      "epoch": 1.0255133789670192,
      "grad_norm": 2.913902521133423,
      "learning_rate": 3.690036900369004e-06,
      "loss": 1.4825,
      "step": 824
    },
    {
      "epoch": 1.0267579340385813,
      "grad_norm": 2.984265089035034,
      "learning_rate": 3.6877306273062737e-06,
      "loss": 1.457,
      "step": 825
    },
    {
      "epoch": 1.0280024891101431,
      "grad_norm": 3.063187599182129,
      "learning_rate": 3.6854243542435427e-06,
      "loss": 1.5537,
      "step": 826
    },
    {
      "epoch": 1.029247044181705,
      "grad_norm": 2.876314878463745,
      "learning_rate": 3.683118081180812e-06,
      "loss": 1.4555,
      "step": 827
    },
    {
      "epoch": 1.030491599253267,
      "grad_norm": 3.039470911026001,
      "learning_rate": 3.6808118081180815e-06,
      "loss": 1.4862,
      "step": 828
    },
    {
      "epoch": 1.031736154324829,
      "grad_norm": 2.8783342838287354,
      "learning_rate": 3.678505535055351e-06,
      "loss": 1.5264,
      "step": 829
    },
    {
      "epoch": 1.0329807093963908,
      "grad_norm": 2.829578161239624,
      "learning_rate": 3.67619926199262e-06,
      "loss": 1.5054,
      "step": 830
    },
    {
      "epoch": 1.0329807093963908,
      "eval_loss": 1.6823173761367798,
      "eval_runtime": 45.0076,
      "eval_samples_per_second": 22.218,
      "eval_steps_per_second": 0.933,
      "step": 830
    },
    {
      "epoch": 1.0342252644679526,
      "grad_norm": 2.9113621711730957,
      "learning_rate": 3.6738929889298898e-06,
      "loss": 1.4211,
      "step": 831
    },
    {
      "epoch": 1.0354698195395147,
      "grad_norm": 2.901256561279297,
      "learning_rate": 3.671586715867159e-06,
      "loss": 1.5566,
      "step": 832
    },
    {
      "epoch": 1.0367143746110765,
      "grad_norm": 2.689814805984497,
      "learning_rate": 3.669280442804428e-06,
      "loss": 1.4062,
      "step": 833
    },
    {
      "epoch": 1.0379589296826384,
      "grad_norm": 2.848036766052246,
      "learning_rate": 3.6669741697416976e-06,
      "loss": 1.524,
      "step": 834
    },
    {
      "epoch": 1.0392034847542004,
      "grad_norm": 2.7704179286956787,
      "learning_rate": 3.6646678966789675e-06,
      "loss": 1.5119,
      "step": 835
    },
    {
      "epoch": 1.0404480398257623,
      "grad_norm": 2.9936625957489014,
      "learning_rate": 3.6623616236162365e-06,
      "loss": 1.5188,
      "step": 836
    },
    {
      "epoch": 1.0416925948973241,
      "grad_norm": 2.752682685852051,
      "learning_rate": 3.660055350553506e-06,
      "loss": 1.4848,
      "step": 837
    },
    {
      "epoch": 1.0429371499688862,
      "grad_norm": 2.7587647438049316,
      "learning_rate": 3.6577490774907753e-06,
      "loss": 1.4784,
      "step": 838
    },
    {
      "epoch": 1.044181705040448,
      "grad_norm": 2.6882312297821045,
      "learning_rate": 3.6554428044280443e-06,
      "loss": 1.4574,
      "step": 839
    },
    {
      "epoch": 1.04542626011201,
      "grad_norm": 2.847754955291748,
      "learning_rate": 3.6531365313653137e-06,
      "loss": 1.5025,
      "step": 840
    },
    {
      "epoch": 1.04542626011201,
      "eval_loss": 1.6905474662780762,
      "eval_runtime": 47.5605,
      "eval_samples_per_second": 21.026,
      "eval_steps_per_second": 0.883,
      "step": 840
    },
    {
      "epoch": 1.046670815183572,
      "grad_norm": 2.7890496253967285,
      "learning_rate": 3.6508302583025836e-06,
      "loss": 1.5182,
      "step": 841
    },
    {
      "epoch": 1.0479153702551338,
      "grad_norm": 2.761654853820801,
      "learning_rate": 3.648523985239853e-06,
      "loss": 1.4691,
      "step": 842
    },
    {
      "epoch": 1.0491599253266957,
      "grad_norm": 2.8627774715423584,
      "learning_rate": 3.646217712177122e-06,
      "loss": 1.4785,
      "step": 843
    },
    {
      "epoch": 1.0504044803982575,
      "grad_norm": 2.8348138332366943,
      "learning_rate": 3.6439114391143914e-06,
      "loss": 1.4418,
      "step": 844
    },
    {
      "epoch": 1.0516490354698196,
      "grad_norm": 2.774764060974121,
      "learning_rate": 3.641605166051661e-06,
      "loss": 1.4872,
      "step": 845
    },
    {
      "epoch": 1.0528935905413814,
      "grad_norm": 2.9172654151916504,
      "learning_rate": 3.63929889298893e-06,
      "loss": 1.473,
      "step": 846
    },
    {
      "epoch": 1.0541381456129433,
      "grad_norm": 2.7001171112060547,
      "learning_rate": 3.6369926199261997e-06,
      "loss": 1.422,
      "step": 847
    },
    {
      "epoch": 1.0553827006845053,
      "grad_norm": 2.8742928504943848,
      "learning_rate": 3.634686346863469e-06,
      "loss": 1.4152,
      "step": 848
    },
    {
      "epoch": 1.0566272557560672,
      "grad_norm": 2.9559481143951416,
      "learning_rate": 3.632380073800738e-06,
      "loss": 1.5244,
      "step": 849
    },
    {
      "epoch": 1.057871810827629,
      "grad_norm": 2.7938740253448486,
      "learning_rate": 3.6300738007380075e-06,
      "loss": 1.4555,
      "step": 850
    },
    {
      "epoch": 1.057871810827629,
      "eval_loss": 1.6855521202087402,
      "eval_runtime": 47.3414,
      "eval_samples_per_second": 21.123,
      "eval_steps_per_second": 0.887,
      "step": 850
    },
    {
      "epoch": 1.0591163658991911,
      "grad_norm": 3.011599540710449,
      "learning_rate": 3.627767527675277e-06,
      "loss": 1.5024,
      "step": 851
    },
    {
      "epoch": 1.060360920970753,
      "grad_norm": 2.720555305480957,
      "learning_rate": 3.625461254612546e-06,
      "loss": 1.5183,
      "step": 852
    },
    {
      "epoch": 1.0616054760423148,
      "grad_norm": 2.9110827445983887,
      "learning_rate": 3.623154981549816e-06,
      "loss": 1.4608,
      "step": 853
    },
    {
      "epoch": 1.0628500311138769,
      "grad_norm": 3.001163959503174,
      "learning_rate": 3.6208487084870852e-06,
      "loss": 1.4797,
      "step": 854
    },
    {
      "epoch": 1.0640945861854387,
      "grad_norm": 2.862942934036255,
      "learning_rate": 3.6185424354243547e-06,
      "loss": 1.4596,
      "step": 855
    },
    {
      "epoch": 1.0653391412570006,
      "grad_norm": 2.907052755355835,
      "learning_rate": 3.6162361623616237e-06,
      "loss": 1.4452,
      "step": 856
    },
    {
      "epoch": 1.0665836963285624,
      "grad_norm": 3.063230276107788,
      "learning_rate": 3.6139298892988935e-06,
      "loss": 1.4864,
      "step": 857
    },
    {
      "epoch": 1.0678282514001245,
      "grad_norm": 3.0228397846221924,
      "learning_rate": 3.611623616236163e-06,
      "loss": 1.4971,
      "step": 858
    },
    {
      "epoch": 1.0690728064716863,
      "grad_norm": 3.149158477783203,
      "learning_rate": 3.609317343173432e-06,
      "loss": 1.4578,
      "step": 859
    },
    {
      "epoch": 1.0703173615432482,
      "grad_norm": 2.7961742877960205,
      "learning_rate": 3.6070110701107014e-06,
      "loss": 1.4246,
      "step": 860
    },
    {
      "epoch": 1.0703173615432482,
      "eval_loss": 1.6836330890655518,
      "eval_runtime": 48.4106,
      "eval_samples_per_second": 20.657,
      "eval_steps_per_second": 0.868,
      "step": 860
    },
    {
      "epoch": 1.0715619166148103,
      "grad_norm": 3.029447078704834,
      "learning_rate": 3.6047047970479708e-06,
      "loss": 1.4864,
      "step": 861
    },
    {
      "epoch": 1.072806471686372,
      "grad_norm": 2.9505438804626465,
      "learning_rate": 3.6023985239852398e-06,
      "loss": 1.46,
      "step": 862
    },
    {
      "epoch": 1.074051026757934,
      "grad_norm": 2.899418592453003,
      "learning_rate": 3.6000922509225096e-06,
      "loss": 1.4937,
      "step": 863
    },
    {
      "epoch": 1.075295581829496,
      "grad_norm": 2.889408588409424,
      "learning_rate": 3.597785977859779e-06,
      "loss": 1.4639,
      "step": 864
    },
    {
      "epoch": 1.0765401369010579,
      "grad_norm": 3.0129477977752686,
      "learning_rate": 3.595479704797048e-06,
      "loss": 1.515,
      "step": 865
    },
    {
      "epoch": 1.0777846919726197,
      "grad_norm": 3.0222294330596924,
      "learning_rate": 3.5931734317343175e-06,
      "loss": 1.49,
      "step": 866
    },
    {
      "epoch": 1.0790292470441818,
      "grad_norm": 3.229149580001831,
      "learning_rate": 3.590867158671587e-06,
      "loss": 1.5106,
      "step": 867
    },
    {
      "epoch": 1.0802738021157436,
      "grad_norm": 3.0588462352752686,
      "learning_rate": 3.5885608856088567e-06,
      "loss": 1.5002,
      "step": 868
    },
    {
      "epoch": 1.0815183571873055,
      "grad_norm": 3.0674571990966797,
      "learning_rate": 3.5862546125461257e-06,
      "loss": 1.4592,
      "step": 869
    },
    {
      "epoch": 1.0827629122588673,
      "grad_norm": 2.9173855781555176,
      "learning_rate": 3.583948339483395e-06,
      "loss": 1.4813,
      "step": 870
    },
    {
      "epoch": 1.0827629122588673,
      "eval_loss": 1.677014946937561,
      "eval_runtime": 49.4597,
      "eval_samples_per_second": 20.218,
      "eval_steps_per_second": 0.849,
      "step": 870
    },
    {
      "epoch": 1.0840074673304294,
      "grad_norm": 2.8278772830963135,
      "learning_rate": 3.5816420664206646e-06,
      "loss": 1.4805,
      "step": 871
    },
    {
      "epoch": 1.0852520224019913,
      "grad_norm": 2.8501267433166504,
      "learning_rate": 3.5793357933579336e-06,
      "loss": 1.471,
      "step": 872
    },
    {
      "epoch": 1.086496577473553,
      "grad_norm": 2.922009229660034,
      "learning_rate": 3.5770295202952034e-06,
      "loss": 1.4801,
      "step": 873
    },
    {
      "epoch": 1.0877411325451152,
      "grad_norm": 2.9171924591064453,
      "learning_rate": 3.574723247232473e-06,
      "loss": 1.473,
      "step": 874
    },
    {
      "epoch": 1.088985687616677,
      "grad_norm": 2.992713212966919,
      "learning_rate": 3.572416974169742e-06,
      "loss": 1.4611,
      "step": 875
    },
    {
      "epoch": 1.0902302426882389,
      "grad_norm": 2.7770748138427734,
      "learning_rate": 3.5701107011070113e-06,
      "loss": 1.474,
      "step": 876
    },
    {
      "epoch": 1.091474797759801,
      "grad_norm": 2.6337552070617676,
      "learning_rate": 3.5678044280442807e-06,
      "loss": 1.4982,
      "step": 877
    },
    {
      "epoch": 1.0927193528313628,
      "grad_norm": 2.8013854026794434,
      "learning_rate": 3.5654981549815497e-06,
      "loss": 1.5154,
      "step": 878
    },
    {
      "epoch": 1.0939639079029246,
      "grad_norm": 2.731400728225708,
      "learning_rate": 3.5631918819188195e-06,
      "loss": 1.5056,
      "step": 879
    },
    {
      "epoch": 1.0952084629744867,
      "grad_norm": 2.8960869312286377,
      "learning_rate": 3.560885608856089e-06,
      "loss": 1.4578,
      "step": 880
    },
    {
      "epoch": 1.0952084629744867,
      "eval_loss": 1.6756274700164795,
      "eval_runtime": 50.0986,
      "eval_samples_per_second": 19.961,
      "eval_steps_per_second": 0.838,
      "step": 880
    },
    {
      "epoch": 1.0964530180460486,
      "grad_norm": 2.736156940460205,
      "learning_rate": 3.5585793357933584e-06,
      "loss": 1.4014,
      "step": 881
    },
    {
      "epoch": 1.0976975731176104,
      "grad_norm": 2.8206899166107178,
      "learning_rate": 3.5562730627306274e-06,
      "loss": 1.519,
      "step": 882
    },
    {
      "epoch": 1.0989421281891725,
      "grad_norm": 2.8125741481781006,
      "learning_rate": 3.553966789667897e-06,
      "loss": 1.493,
      "step": 883
    },
    {
      "epoch": 1.1001866832607343,
      "grad_norm": 2.7721848487854004,
      "learning_rate": 3.5516605166051667e-06,
      "loss": 1.4805,
      "step": 884
    },
    {
      "epoch": 1.1014312383322962,
      "grad_norm": 2.8858790397644043,
      "learning_rate": 3.5493542435424357e-06,
      "loss": 1.4339,
      "step": 885
    },
    {
      "epoch": 1.102675793403858,
      "grad_norm": 2.897688865661621,
      "learning_rate": 3.547047970479705e-06,
      "loss": 1.4663,
      "step": 886
    },
    {
      "epoch": 1.10392034847542,
      "grad_norm": 2.8516011238098145,
      "learning_rate": 3.5447416974169745e-06,
      "loss": 1.4946,
      "step": 887
    },
    {
      "epoch": 1.105164903546982,
      "grad_norm": 2.801990270614624,
      "learning_rate": 3.5424354243542435e-06,
      "loss": 1.4764,
      "step": 888
    },
    {
      "epoch": 1.1064094586185438,
      "grad_norm": 2.8651299476623535,
      "learning_rate": 3.5401291512915133e-06,
      "loss": 1.471,
      "step": 889
    },
    {
      "epoch": 1.1076540136901059,
      "grad_norm": 2.8247878551483154,
      "learning_rate": 3.5378228782287828e-06,
      "loss": 1.5494,
      "step": 890
    },
    {
      "epoch": 1.1076540136901059,
      "eval_loss": 1.6814671754837036,
      "eval_runtime": 49.7658,
      "eval_samples_per_second": 20.094,
      "eval_steps_per_second": 0.844,
      "step": 890
    },
    {
      "epoch": 1.1088985687616677,
      "grad_norm": 2.750497579574585,
      "learning_rate": 3.535516605166052e-06,
      "loss": 1.4578,
      "step": 891
    },
    {
      "epoch": 1.1101431238332296,
      "grad_norm": 2.7336409091949463,
      "learning_rate": 3.533210332103321e-06,
      "loss": 1.4486,
      "step": 892
    },
    {
      "epoch": 1.1113876789047916,
      "grad_norm": 3.0764975547790527,
      "learning_rate": 3.5309040590405906e-06,
      "loss": 1.4661,
      "step": 893
    },
    {
      "epoch": 1.1126322339763535,
      "grad_norm": 2.7930431365966797,
      "learning_rate": 3.5285977859778605e-06,
      "loss": 1.4378,
      "step": 894
    },
    {
      "epoch": 1.1138767890479153,
      "grad_norm": 2.944173574447632,
      "learning_rate": 3.5262915129151295e-06,
      "loss": 1.4639,
      "step": 895
    },
    {
      "epoch": 1.1151213441194774,
      "grad_norm": 3.0577549934387207,
      "learning_rate": 3.523985239852399e-06,
      "loss": 1.4668,
      "step": 896
    },
    {
      "epoch": 1.1163658991910392,
      "grad_norm": 2.805238962173462,
      "learning_rate": 3.5216789667896683e-06,
      "loss": 1.4531,
      "step": 897
    },
    {
      "epoch": 1.117610454262601,
      "grad_norm": 2.8995230197906494,
      "learning_rate": 3.5193726937269373e-06,
      "loss": 1.5016,
      "step": 898
    },
    {
      "epoch": 1.1188550093341632,
      "grad_norm": 2.8598949909210205,
      "learning_rate": 3.5170664206642067e-06,
      "loss": 1.4276,
      "step": 899
    },
    {
      "epoch": 1.120099564405725,
      "grad_norm": 2.8928749561309814,
      "learning_rate": 3.5147601476014766e-06,
      "loss": 1.4888,
      "step": 900
    },
    {
      "epoch": 1.120099564405725,
      "eval_loss": 1.6829349994659424,
      "eval_runtime": 46.7111,
      "eval_samples_per_second": 21.408,
      "eval_steps_per_second": 0.899,
      "step": 900
    },
    {
      "epoch": 1.1213441194772868,
      "grad_norm": 2.9442391395568848,
      "learning_rate": 3.5124538745387456e-06,
      "loss": 1.4636,
      "step": 901
    },
    {
      "epoch": 1.1225886745488487,
      "grad_norm": 3.1975884437561035,
      "learning_rate": 3.510147601476015e-06,
      "loss": 1.4654,
      "step": 902
    },
    {
      "epoch": 1.1238332296204108,
      "grad_norm": 2.915771722793579,
      "learning_rate": 3.5078413284132844e-06,
      "loss": 1.501,
      "step": 903
    },
    {
      "epoch": 1.1250777846919726,
      "grad_norm": 2.880690097808838,
      "learning_rate": 3.505535055350554e-06,
      "loss": 1.4629,
      "step": 904
    },
    {
      "epoch": 1.1263223397635345,
      "grad_norm": 2.868868350982666,
      "learning_rate": 3.503228782287823e-06,
      "loss": 1.4926,
      "step": 905
    },
    {
      "epoch": 1.1275668948350965,
      "grad_norm": 3.1113288402557373,
      "learning_rate": 3.5009225092250927e-06,
      "loss": 1.4572,
      "step": 906
    },
    {
      "epoch": 1.1288114499066584,
      "grad_norm": 2.904810667037964,
      "learning_rate": 3.498616236162362e-06,
      "loss": 1.4151,
      "step": 907
    },
    {
      "epoch": 1.1300560049782202,
      "grad_norm": 2.906148672103882,
      "learning_rate": 3.496309963099631e-06,
      "loss": 1.4713,
      "step": 908
    },
    {
      "epoch": 1.1313005600497823,
      "grad_norm": 2.9063987731933594,
      "learning_rate": 3.4940036900369005e-06,
      "loss": 1.5141,
      "step": 909
    },
    {
      "epoch": 1.1325451151213441,
      "grad_norm": 2.7462973594665527,
      "learning_rate": 3.4916974169741704e-06,
      "loss": 1.4396,
      "step": 910
    },
    {
      "epoch": 1.1325451151213441,
      "eval_loss": 1.6758365631103516,
      "eval_runtime": 41.9838,
      "eval_samples_per_second": 23.819,
      "eval_steps_per_second": 1.0,
      "step": 910
    },
    {
      "epoch": 1.133789670192906,
      "grad_norm": 2.854579448699951,
      "learning_rate": 3.4893911439114394e-06,
      "loss": 1.4471,
      "step": 911
    },
    {
      "epoch": 1.135034225264468,
      "grad_norm": 2.8700456619262695,
      "learning_rate": 3.487084870848709e-06,
      "loss": 1.4237,
      "step": 912
    },
    {
      "epoch": 1.13627878033603,
      "grad_norm": 2.883131742477417,
      "learning_rate": 3.4847785977859782e-06,
      "loss": 1.4014,
      "step": 913
    },
    {
      "epoch": 1.1375233354075918,
      "grad_norm": 2.9069855213165283,
      "learning_rate": 3.4824723247232472e-06,
      "loss": 1.396,
      "step": 914
    },
    {
      "epoch": 1.1387678904791536,
      "grad_norm": 2.9044485092163086,
      "learning_rate": 3.4801660516605166e-06,
      "loss": 1.4516,
      "step": 915
    },
    {
      "epoch": 1.1400124455507157,
      "grad_norm": 2.9939963817596436,
      "learning_rate": 3.4778597785977865e-06,
      "loss": 1.4347,
      "step": 916
    },
    {
      "epoch": 1.1412570006222775,
      "grad_norm": 2.762651205062866,
      "learning_rate": 3.475553505535056e-06,
      "loss": 1.4278,
      "step": 917
    },
    {
      "epoch": 1.1425015556938394,
      "grad_norm": 2.84051775932312,
      "learning_rate": 3.473247232472325e-06,
      "loss": 1.5154,
      "step": 918
    },
    {
      "epoch": 1.1437461107654014,
      "grad_norm": 2.7700562477111816,
      "learning_rate": 3.4709409594095943e-06,
      "loss": 1.4296,
      "step": 919
    },
    {
      "epoch": 1.1449906658369633,
      "grad_norm": 2.8805646896362305,
      "learning_rate": 3.4686346863468638e-06,
      "loss": 1.4716,
      "step": 920
    },
    {
      "epoch": 1.1449906658369633,
      "eval_loss": 1.672538161277771,
      "eval_runtime": 44.9663,
      "eval_samples_per_second": 22.239,
      "eval_steps_per_second": 0.934,
      "step": 920
    },
    {
      "epoch": 1.1462352209085251,
      "grad_norm": 2.8342535495758057,
      "learning_rate": 3.4663284132841328e-06,
      "loss": 1.4461,
      "step": 921
    },
    {
      "epoch": 1.1474797759800872,
      "grad_norm": 3.000023603439331,
      "learning_rate": 3.4640221402214026e-06,
      "loss": 1.4244,
      "step": 922
    },
    {
      "epoch": 1.148724331051649,
      "grad_norm": 2.7098913192749023,
      "learning_rate": 3.461715867158672e-06,
      "loss": 1.3943,
      "step": 923
    },
    {
      "epoch": 1.149968886123211,
      "grad_norm": 2.8422343730926514,
      "learning_rate": 3.459409594095941e-06,
      "loss": 1.456,
      "step": 924
    },
    {
      "epoch": 1.151213441194773,
      "grad_norm": 2.7740139961242676,
      "learning_rate": 3.4571033210332105e-06,
      "loss": 1.4299,
      "step": 925
    },
    {
      "epoch": 1.1524579962663348,
      "grad_norm": 2.792193651199341,
      "learning_rate": 3.4547970479704803e-06,
      "loss": 1.4255,
      "step": 926
    },
    {
      "epoch": 1.1537025513378967,
      "grad_norm": 2.786365032196045,
      "learning_rate": 3.4524907749077493e-06,
      "loss": 1.4674,
      "step": 927
    },
    {
      "epoch": 1.1549471064094585,
      "grad_norm": 2.791139841079712,
      "learning_rate": 3.4501845018450187e-06,
      "loss": 1.4219,
      "step": 928
    },
    {
      "epoch": 1.1561916614810206,
      "grad_norm": 2.829876184463501,
      "learning_rate": 3.447878228782288e-06,
      "loss": 1.5115,
      "step": 929
    },
    {
      "epoch": 1.1574362165525824,
      "grad_norm": 2.8779923915863037,
      "learning_rate": 3.4455719557195576e-06,
      "loss": 1.4484,
      "step": 930
    },
    {
      "epoch": 1.1574362165525824,
      "eval_loss": 1.6719067096710205,
      "eval_runtime": 44.4657,
      "eval_samples_per_second": 22.489,
      "eval_steps_per_second": 0.945,
      "step": 930
    },
    {
      "epoch": 1.1586807716241443,
      "grad_norm": 2.7856647968292236,
      "learning_rate": 3.4432656826568266e-06,
      "loss": 1.4423,
      "step": 931
    },
    {
      "epoch": 1.1599253266957064,
      "grad_norm": 2.8183937072753906,
      "learning_rate": 3.4409594095940964e-06,
      "loss": 1.4492,
      "step": 932
    },
    {
      "epoch": 1.1611698817672682,
      "grad_norm": 2.874630928039551,
      "learning_rate": 3.438653136531366e-06,
      "loss": 1.4546,
      "step": 933
    },
    {
      "epoch": 1.16241443683883,
      "grad_norm": 2.7180988788604736,
      "learning_rate": 3.436346863468635e-06,
      "loss": 1.4402,
      "step": 934
    },
    {
      "epoch": 1.1636589919103921,
      "grad_norm": 2.682368516921997,
      "learning_rate": 3.4340405904059043e-06,
      "loss": 1.4442,
      "step": 935
    },
    {
      "epoch": 1.164903546981954,
      "grad_norm": 2.8584909439086914,
      "learning_rate": 3.4317343173431737e-06,
      "loss": 1.4707,
      "step": 936
    },
    {
      "epoch": 1.1661481020535158,
      "grad_norm": 3.074317693710327,
      "learning_rate": 3.4294280442804427e-06,
      "loss": 1.5241,
      "step": 937
    },
    {
      "epoch": 1.167392657125078,
      "grad_norm": 2.9097986221313477,
      "learning_rate": 3.4271217712177125e-06,
      "loss": 1.4518,
      "step": 938
    },
    {
      "epoch": 1.1686372121966397,
      "grad_norm": 2.9661238193511963,
      "learning_rate": 3.424815498154982e-06,
      "loss": 1.4665,
      "step": 939
    },
    {
      "epoch": 1.1698817672682016,
      "grad_norm": 2.8246936798095703,
      "learning_rate": 3.4225092250922514e-06,
      "loss": 1.4789,
      "step": 940
    },
    {
      "epoch": 1.1698817672682016,
      "eval_loss": 1.6716506481170654,
      "eval_runtime": 43.1065,
      "eval_samples_per_second": 23.198,
      "eval_steps_per_second": 0.974,
      "step": 940
    },
    {
      "epoch": 1.1711263223397634,
      "grad_norm": 2.9993770122528076,
      "learning_rate": 3.4202029520295204e-06,
      "loss": 1.5287,
      "step": 941
    },
    {
      "epoch": 1.1723708774113255,
      "grad_norm": 2.7746829986572266,
      "learning_rate": 3.41789667896679e-06,
      "loss": 1.4105,
      "step": 942
    },
    {
      "epoch": 1.1736154324828874,
      "grad_norm": 2.7988786697387695,
      "learning_rate": 3.4155904059040596e-06,
      "loss": 1.4543,
      "step": 943
    },
    {
      "epoch": 1.1748599875544492,
      "grad_norm": 2.8851349353790283,
      "learning_rate": 3.4132841328413286e-06,
      "loss": 1.4527,
      "step": 944
    },
    {
      "epoch": 1.1761045426260113,
      "grad_norm": 2.8439807891845703,
      "learning_rate": 3.410977859778598e-06,
      "loss": 1.4383,
      "step": 945
    },
    {
      "epoch": 1.1773490976975731,
      "grad_norm": 2.811246156692505,
      "learning_rate": 3.4086715867158675e-06,
      "loss": 1.4004,
      "step": 946
    },
    {
      "epoch": 1.178593652769135,
      "grad_norm": 2.8452935218811035,
      "learning_rate": 3.4063653136531365e-06,
      "loss": 1.4989,
      "step": 947
    },
    {
      "epoch": 1.179838207840697,
      "grad_norm": 2.7633206844329834,
      "learning_rate": 3.4040590405904063e-06,
      "loss": 1.4644,
      "step": 948
    },
    {
      "epoch": 1.1810827629122589,
      "grad_norm": 2.777296543121338,
      "learning_rate": 3.4017527675276758e-06,
      "loss": 1.4398,
      "step": 949
    },
    {
      "epoch": 1.1823273179838207,
      "grad_norm": 3.0879335403442383,
      "learning_rate": 3.3994464944649448e-06,
      "loss": 1.4901,
      "step": 950
    },
    {
      "epoch": 1.1823273179838207,
      "eval_loss": 1.6718789339065552,
      "eval_runtime": 43.7598,
      "eval_samples_per_second": 22.852,
      "eval_steps_per_second": 0.96,
      "step": 950
    },
    {
      "epoch": 1.1835718730553828,
      "grad_norm": 3.030923366546631,
      "learning_rate": 3.397140221402214e-06,
      "loss": 1.4992,
      "step": 951
    },
    {
      "epoch": 1.1848164281269447,
      "grad_norm": 2.7215065956115723,
      "learning_rate": 3.3948339483394836e-06,
      "loss": 1.4709,
      "step": 952
    },
    {
      "epoch": 1.1860609831985065,
      "grad_norm": 2.738738775253296,
      "learning_rate": 3.3925276752767534e-06,
      "loss": 1.4838,
      "step": 953
    },
    {
      "epoch": 1.1873055382700684,
      "grad_norm": 2.8057045936584473,
      "learning_rate": 3.3902214022140224e-06,
      "loss": 1.4879,
      "step": 954
    },
    {
      "epoch": 1.1885500933416304,
      "grad_norm": 2.846531867980957,
      "learning_rate": 3.387915129151292e-06,
      "loss": 1.4585,
      "step": 955
    },
    {
      "epoch": 1.1897946484131923,
      "grad_norm": 2.8825955390930176,
      "learning_rate": 3.3856088560885613e-06,
      "loss": 1.5483,
      "step": 956
    },
    {
      "epoch": 1.1910392034847541,
      "grad_norm": 2.720550775527954,
      "learning_rate": 3.3833025830258303e-06,
      "loss": 1.4356,
      "step": 957
    },
    {
      "epoch": 1.1922837585563162,
      "grad_norm": 2.7653985023498535,
      "learning_rate": 3.3809963099630997e-06,
      "loss": 1.4679,
      "step": 958
    },
    {
      "epoch": 1.193528313627878,
      "grad_norm": 2.868544101715088,
      "learning_rate": 3.3786900369003696e-06,
      "loss": 1.4492,
      "step": 959
    },
    {
      "epoch": 1.1947728686994399,
      "grad_norm": 2.877669334411621,
      "learning_rate": 3.3763837638376386e-06,
      "loss": 1.493,
      "step": 960
    },
    {
      "epoch": 1.1947728686994399,
      "eval_loss": 1.6707913875579834,
      "eval_runtime": 44.0558,
      "eval_samples_per_second": 22.698,
      "eval_steps_per_second": 0.953,
      "step": 960
    },
    {
      "epoch": 1.196017423771002,
      "grad_norm": 3.0898303985595703,
      "learning_rate": 3.374077490774908e-06,
      "loss": 1.4336,
      "step": 961
    },
    {
      "epoch": 1.1972619788425638,
      "grad_norm": 2.859119415283203,
      "learning_rate": 3.3717712177121774e-06,
      "loss": 1.4862,
      "step": 962
    },
    {
      "epoch": 1.1985065339141256,
      "grad_norm": 2.9307544231414795,
      "learning_rate": 3.3694649446494464e-06,
      "loss": 1.4977,
      "step": 963
    },
    {
      "epoch": 1.1997510889856877,
      "grad_norm": 2.785691976547241,
      "learning_rate": 3.3671586715867163e-06,
      "loss": 1.4203,
      "step": 964
    },
    {
      "epoch": 1.2009956440572496,
      "grad_norm": 2.762620449066162,
      "learning_rate": 3.3648523985239857e-06,
      "loss": 1.5147,
      "step": 965
    },
    {
      "epoch": 1.2022401991288114,
      "grad_norm": 2.8887553215026855,
      "learning_rate": 3.362546125461255e-06,
      "loss": 1.4435,
      "step": 966
    },
    {
      "epoch": 1.2034847542003733,
      "grad_norm": 2.8161838054656982,
      "learning_rate": 3.360239852398524e-06,
      "loss": 1.4436,
      "step": 967
    },
    {
      "epoch": 1.2047293092719353,
      "grad_norm": 3.0834996700286865,
      "learning_rate": 3.3579335793357935e-06,
      "loss": 1.4861,
      "step": 968
    },
    {
      "epoch": 1.2059738643434972,
      "grad_norm": 3.0328774452209473,
      "learning_rate": 3.3556273062730634e-06,
      "loss": 1.5064,
      "step": 969
    },
    {
      "epoch": 1.207218419415059,
      "grad_norm": 2.977429151535034,
      "learning_rate": 3.3533210332103324e-06,
      "loss": 1.4148,
      "step": 970
    },
    {
      "epoch": 1.207218419415059,
      "eval_loss": 1.672300100326538,
      "eval_runtime": 46.0257,
      "eval_samples_per_second": 21.727,
      "eval_steps_per_second": 0.913,
      "step": 970
    },
    {
      "epoch": 1.208462974486621,
      "grad_norm": 2.9340717792510986,
      "learning_rate": 3.3510147601476018e-06,
      "loss": 1.4726,
      "step": 971
    },
    {
      "epoch": 1.209707529558183,
      "grad_norm": 2.8549611568450928,
      "learning_rate": 3.348708487084871e-06,
      "loss": 1.4798,
      "step": 972
    },
    {
      "epoch": 1.2109520846297448,
      "grad_norm": 2.74188232421875,
      "learning_rate": 3.34640221402214e-06,
      "loss": 1.4231,
      "step": 973
    },
    {
      "epoch": 1.2121966397013069,
      "grad_norm": 2.8730149269104004,
      "learning_rate": 3.3440959409594096e-06,
      "loss": 1.4135,
      "step": 974
    },
    {
      "epoch": 1.2134411947728687,
      "grad_norm": 2.844428062438965,
      "learning_rate": 3.3417896678966795e-06,
      "loss": 1.4164,
      "step": 975
    },
    {
      "epoch": 1.2146857498444306,
      "grad_norm": 2.920804738998413,
      "learning_rate": 3.3394833948339485e-06,
      "loss": 1.4898,
      "step": 976
    },
    {
      "epoch": 1.2159303049159926,
      "grad_norm": 2.9339325428009033,
      "learning_rate": 3.337177121771218e-06,
      "loss": 1.5112,
      "step": 977
    },
    {
      "epoch": 1.2171748599875545,
      "grad_norm": 2.7855703830718994,
      "learning_rate": 3.3348708487084873e-06,
      "loss": 1.4209,
      "step": 978
    },
    {
      "epoch": 1.2184194150591163,
      "grad_norm": 2.9347097873687744,
      "learning_rate": 3.332564575645757e-06,
      "loss": 1.4345,
      "step": 979
    },
    {
      "epoch": 1.2196639701306782,
      "grad_norm": 2.9517159461975098,
      "learning_rate": 3.330258302583026e-06,
      "loss": 1.4873,
      "step": 980
    },
    {
      "epoch": 1.2196639701306782,
      "eval_loss": 1.6751104593276978,
      "eval_runtime": 45.8657,
      "eval_samples_per_second": 21.803,
      "eval_steps_per_second": 0.916,
      "step": 980
    },
    {
      "epoch": 1.2209085252022402,
      "grad_norm": 3.043290853500366,
      "learning_rate": 3.3279520295202956e-06,
      "loss": 1.5121,
      "step": 981
    },
    {
      "epoch": 1.222153080273802,
      "grad_norm": 2.9060018062591553,
      "learning_rate": 3.325645756457565e-06,
      "loss": 1.4888,
      "step": 982
    },
    {
      "epoch": 1.223397635345364,
      "grad_norm": 3.186314344406128,
      "learning_rate": 3.323339483394834e-06,
      "loss": 1.4944,
      "step": 983
    },
    {
      "epoch": 1.224642190416926,
      "grad_norm": 3.2327520847320557,
      "learning_rate": 3.3210332103321034e-06,
      "loss": 1.515,
      "step": 984
    },
    {
      "epoch": 1.2258867454884879,
      "grad_norm": 2.9379491806030273,
      "learning_rate": 3.3187269372693733e-06,
      "loss": 1.4475,
      "step": 985
    },
    {
      "epoch": 1.2271313005600497,
      "grad_norm": 2.732428789138794,
      "learning_rate": 3.3164206642066423e-06,
      "loss": 1.4595,
      "step": 986
    },
    {
      "epoch": 1.2283758556316118,
      "grad_norm": 2.745485782623291,
      "learning_rate": 3.3141143911439117e-06,
      "loss": 1.4716,
      "step": 987
    },
    {
      "epoch": 1.2296204107031736,
      "grad_norm": 3.172178268432617,
      "learning_rate": 3.311808118081181e-06,
      "loss": 1.4901,
      "step": 988
    },
    {
      "epoch": 1.2308649657747355,
      "grad_norm": 2.8245432376861572,
      "learning_rate": 3.3095018450184506e-06,
      "loss": 1.4738,
      "step": 989
    },
    {
      "epoch": 1.2321095208462975,
      "grad_norm": 2.8529610633850098,
      "learning_rate": 3.3071955719557196e-06,
      "loss": 1.4549,
      "step": 990
    },
    {
      "epoch": 1.2321095208462975,
      "eval_loss": 1.6769033670425415,
      "eval_runtime": 45.7162,
      "eval_samples_per_second": 21.874,
      "eval_steps_per_second": 0.919,
      "step": 990
    },
    {
      "epoch": 1.2333540759178594,
      "grad_norm": 2.7896978855133057,
      "learning_rate": 3.3048892988929894e-06,
      "loss": 1.4207,
      "step": 991
    },
    {
      "epoch": 1.2345986309894212,
      "grad_norm": 3.011547088623047,
      "learning_rate": 3.302583025830259e-06,
      "loss": 1.4615,
      "step": 992
    },
    {
      "epoch": 1.235843186060983,
      "grad_norm": 2.838911771774292,
      "learning_rate": 3.300276752767528e-06,
      "loss": 1.4539,
      "step": 993
    },
    {
      "epoch": 1.2370877411325452,
      "grad_norm": 2.731318235397339,
      "learning_rate": 3.2979704797047972e-06,
      "loss": 1.4452,
      "step": 994
    },
    {
      "epoch": 1.238332296204107,
      "grad_norm": 2.7334372997283936,
      "learning_rate": 3.2956642066420667e-06,
      "loss": 1.4658,
      "step": 995
    },
    {
      "epoch": 1.2395768512756689,
      "grad_norm": 2.8925108909606934,
      "learning_rate": 3.2933579335793357e-06,
      "loss": 1.4368,
      "step": 996
    },
    {
      "epoch": 1.240821406347231,
      "grad_norm": 2.8550665378570557,
      "learning_rate": 3.2910516605166055e-06,
      "loss": 1.4428,
      "step": 997
    },
    {
      "epoch": 1.2420659614187928,
      "grad_norm": 2.953028440475464,
      "learning_rate": 3.288745387453875e-06,
      "loss": 1.5585,
      "step": 998
    },
    {
      "epoch": 1.2433105164903546,
      "grad_norm": 2.833862781524658,
      "learning_rate": 3.286439114391144e-06,
      "loss": 1.5013,
      "step": 999
    },
    {
      "epoch": 1.2445550715619167,
      "grad_norm": 3.0078632831573486,
      "learning_rate": 3.2841328413284134e-06,
      "loss": 1.421,
      "step": 1000
    },
    {
      "epoch": 1.2445550715619167,
      "eval_loss": 1.6693010330200195,
      "eval_runtime": 46.1434,
      "eval_samples_per_second": 21.672,
      "eval_steps_per_second": 0.91,
      "step": 1000
    },
    {
      "epoch": 1.2457996266334785,
      "grad_norm": 3.1139776706695557,
      "learning_rate": 3.281826568265683e-06,
      "loss": 1.4494,
      "step": 1001
    },
    {
      "epoch": 1.2470441817050404,
      "grad_norm": 2.6980462074279785,
      "learning_rate": 3.2795202952029526e-06,
      "loss": 1.4689,
      "step": 1002
    },
    {
      "epoch": 1.2482887367766025,
      "grad_norm": 2.797846794128418,
      "learning_rate": 3.2772140221402216e-06,
      "loss": 1.3867,
      "step": 1003
    },
    {
      "epoch": 1.2495332918481643,
      "grad_norm": 2.949923515319824,
      "learning_rate": 3.274907749077491e-06,
      "loss": 1.4953,
      "step": 1004
    },
    {
      "epoch": 1.2507778469197262,
      "grad_norm": 3.0071775913238525,
      "learning_rate": 3.2726014760147605e-06,
      "loss": 1.507,
      "step": 1005
    },
    {
      "epoch": 1.252022401991288,
      "grad_norm": 2.9245765209198,
      "learning_rate": 3.2702952029520295e-06,
      "loss": 1.497,
      "step": 1006
    },
    {
      "epoch": 1.25326695706285,
      "grad_norm": 2.755397081375122,
      "learning_rate": 3.2679889298892993e-06,
      "loss": 1.4847,
      "step": 1007
    },
    {
      "epoch": 1.254511512134412,
      "grad_norm": 2.8560867309570312,
      "learning_rate": 3.2656826568265687e-06,
      "loss": 1.4446,
      "step": 1008
    },
    {
      "epoch": 1.255756067205974,
      "grad_norm": 2.805844783782959,
      "learning_rate": 3.2633763837638377e-06,
      "loss": 1.4877,
      "step": 1009
    },
    {
      "epoch": 1.2570006222775358,
      "grad_norm": 3.004056453704834,
      "learning_rate": 3.261070110701107e-06,
      "loss": 1.4706,
      "step": 1010
    },
    {
      "epoch": 1.2570006222775358,
      "eval_loss": 1.662219762802124,
      "eval_runtime": 42.45,
      "eval_samples_per_second": 23.557,
      "eval_steps_per_second": 0.989,
      "step": 1010
    },
    {
      "epoch": 1.2582451773490977,
      "grad_norm": 2.9530556201934814,
      "learning_rate": 3.2587638376383766e-06,
      "loss": 1.4623,
      "step": 1011
    },
    {
      "epoch": 1.2594897324206595,
      "grad_norm": 2.8305587768554688,
      "learning_rate": 3.2564575645756456e-06,
      "loss": 1.4659,
      "step": 1012
    },
    {
      "epoch": 1.2607342874922216,
      "grad_norm": 2.895482063293457,
      "learning_rate": 3.2541512915129154e-06,
      "loss": 1.4267,
      "step": 1013
    },
    {
      "epoch": 1.2619788425637835,
      "grad_norm": 2.9198293685913086,
      "learning_rate": 3.251845018450185e-06,
      "loss": 1.4226,
      "step": 1014
    },
    {
      "epoch": 1.2632233976353453,
      "grad_norm": 2.9762983322143555,
      "learning_rate": 3.2495387453874543e-06,
      "loss": 1.4666,
      "step": 1015
    },
    {
      "epoch": 1.2644679527069074,
      "grad_norm": 2.905694007873535,
      "learning_rate": 3.2472324723247233e-06,
      "loss": 1.445,
      "step": 1016
    },
    {
      "epoch": 1.2657125077784692,
      "grad_norm": 2.8624324798583984,
      "learning_rate": 3.244926199261993e-06,
      "loss": 1.4685,
      "step": 1017
    },
    {
      "epoch": 1.266957062850031,
      "grad_norm": 2.8351657390594482,
      "learning_rate": 3.2426199261992625e-06,
      "loss": 1.4477,
      "step": 1018
    },
    {
      "epoch": 1.268201617921593,
      "grad_norm": 3.0314555168151855,
      "learning_rate": 3.2403136531365315e-06,
      "loss": 1.4744,
      "step": 1019
    },
    {
      "epoch": 1.269446172993155,
      "grad_norm": 2.848743438720703,
      "learning_rate": 3.238007380073801e-06,
      "loss": 1.3832,
      "step": 1020
    },
    {
      "epoch": 1.269446172993155,
      "eval_loss": 1.6641864776611328,
      "eval_runtime": 44.6957,
      "eval_samples_per_second": 22.374,
      "eval_steps_per_second": 0.94,
      "step": 1020
    },
    {
      "epoch": 1.2706907280647168,
      "grad_norm": 2.8528616428375244,
      "learning_rate": 3.2357011070110704e-06,
      "loss": 1.486,
      "step": 1021
    },
    {
      "epoch": 1.271935283136279,
      "grad_norm": 3.064131259918213,
      "learning_rate": 3.2333948339483394e-06,
      "loss": 1.4449,
      "step": 1022
    },
    {
      "epoch": 1.2731798382078408,
      "grad_norm": 3.1047396659851074,
      "learning_rate": 3.2310885608856092e-06,
      "loss": 1.4796,
      "step": 1023
    },
    {
      "epoch": 1.2744243932794026,
      "grad_norm": 2.897571086883545,
      "learning_rate": 3.2287822878228787e-06,
      "loss": 1.4762,
      "step": 1024
    },
    {
      "epoch": 1.2756689483509644,
      "grad_norm": 2.914445161819458,
      "learning_rate": 3.2264760147601477e-06,
      "loss": 1.4315,
      "step": 1025
    },
    {
      "epoch": 1.2769135034225265,
      "grad_norm": 2.913105010986328,
      "learning_rate": 3.224169741697417e-06,
      "loss": 1.4314,
      "step": 1026
    },
    {
      "epoch": 1.2781580584940884,
      "grad_norm": 2.7426509857177734,
      "learning_rate": 3.2218634686346865e-06,
      "loss": 1.4006,
      "step": 1027
    },
    {
      "epoch": 1.2794026135656502,
      "grad_norm": 3.15177845954895,
      "learning_rate": 3.2195571955719564e-06,
      "loss": 1.4426,
      "step": 1028
    },
    {
      "epoch": 1.2806471686372123,
      "grad_norm": 2.9131808280944824,
      "learning_rate": 3.2172509225092254e-06,
      "loss": 1.5067,
      "step": 1029
    },
    {
      "epoch": 1.2818917237087741,
      "grad_norm": 2.9975693225860596,
      "learning_rate": 3.2149446494464948e-06,
      "loss": 1.4612,
      "step": 1030
    },
    {
      "epoch": 1.2818917237087741,
      "eval_loss": 1.6711934804916382,
      "eval_runtime": 50.154,
      "eval_samples_per_second": 19.939,
      "eval_steps_per_second": 0.837,
      "step": 1030
    },
    {
      "epoch": 1.283136278780336,
      "grad_norm": 2.9257564544677734,
      "learning_rate": 3.212638376383764e-06,
      "loss": 1.4383,
      "step": 1031
    },
    {
      "epoch": 1.2843808338518978,
      "grad_norm": 2.9188551902770996,
      "learning_rate": 3.210332103321033e-06,
      "loss": 1.5097,
      "step": 1032
    },
    {
      "epoch": 1.28562538892346,
      "grad_norm": 2.9625136852264404,
      "learning_rate": 3.208025830258303e-06,
      "loss": 1.4127,
      "step": 1033
    },
    {
      "epoch": 1.2868699439950217,
      "grad_norm": 2.6977076530456543,
      "learning_rate": 3.2057195571955725e-06,
      "loss": 1.4163,
      "step": 1034
    },
    {
      "epoch": 1.2881144990665838,
      "grad_norm": 2.711514949798584,
      "learning_rate": 3.2034132841328415e-06,
      "loss": 1.4292,
      "step": 1035
    },
    {
      "epoch": 1.2893590541381457,
      "grad_norm": 2.8522849082946777,
      "learning_rate": 3.201107011070111e-06,
      "loss": 1.4386,
      "step": 1036
    },
    {
      "epoch": 1.2906036092097075,
      "grad_norm": 3.0120058059692383,
      "learning_rate": 3.1988007380073803e-06,
      "loss": 1.4429,
      "step": 1037
    },
    {
      "epoch": 1.2918481642812694,
      "grad_norm": 2.865572929382324,
      "learning_rate": 3.1964944649446493e-06,
      "loss": 1.4766,
      "step": 1038
    },
    {
      "epoch": 1.2930927193528314,
      "grad_norm": 2.8616085052490234,
      "learning_rate": 3.194188191881919e-06,
      "loss": 1.4391,
      "step": 1039
    },
    {
      "epoch": 1.2943372744243933,
      "grad_norm": 2.7028727531433105,
      "learning_rate": 3.1918819188191886e-06,
      "loss": 1.4141,
      "step": 1040
    },
    {
      "epoch": 1.2943372744243933,
      "eval_loss": 1.6724774837493896,
      "eval_runtime": 50.8274,
      "eval_samples_per_second": 19.674,
      "eval_steps_per_second": 0.826,
      "step": 1040
    },
    {
      "epoch": 1.2955818294959551,
      "grad_norm": 2.8322980403900146,
      "learning_rate": 3.189575645756458e-06,
      "loss": 1.4789,
      "step": 1041
    },
    {
      "epoch": 1.2968263845675172,
      "grad_norm": 2.979027032852173,
      "learning_rate": 3.187269372693727e-06,
      "loss": 1.4945,
      "step": 1042
    },
    {
      "epoch": 1.298070939639079,
      "grad_norm": 3.0695791244506836,
      "learning_rate": 3.1849630996309964e-06,
      "loss": 1.5108,
      "step": 1043
    },
    {
      "epoch": 1.299315494710641,
      "grad_norm": 2.7639617919921875,
      "learning_rate": 3.1826568265682663e-06,
      "loss": 1.4381,
      "step": 1044
    },
    {
      "epoch": 1.3005600497822027,
      "grad_norm": 2.865220308303833,
      "learning_rate": 3.1803505535055353e-06,
      "loss": 1.4787,
      "step": 1045
    },
    {
      "epoch": 1.3018046048537648,
      "grad_norm": 2.9786012172698975,
      "learning_rate": 3.1780442804428047e-06,
      "loss": 1.4356,
      "step": 1046
    },
    {
      "epoch": 1.3030491599253267,
      "grad_norm": 2.867659568786621,
      "learning_rate": 3.175738007380074e-06,
      "loss": 1.4831,
      "step": 1047
    },
    {
      "epoch": 1.3042937149968887,
      "grad_norm": 2.7092227935791016,
      "learning_rate": 3.173431734317343e-06,
      "loss": 1.4189,
      "step": 1048
    },
    {
      "epoch": 1.3055382700684506,
      "grad_norm": 2.85172176361084,
      "learning_rate": 3.1711254612546125e-06,
      "loss": 1.4252,
      "step": 1049
    },
    {
      "epoch": 1.3067828251400124,
      "grad_norm": 2.9820401668548584,
      "learning_rate": 3.1688191881918824e-06,
      "loss": 1.4924,
      "step": 1050
    },
    {
      "epoch": 1.3067828251400124,
      "eval_loss": 1.672357201576233,
      "eval_runtime": 53.7158,
      "eval_samples_per_second": 18.616,
      "eval_steps_per_second": 0.782,
      "step": 1050
    },
    {
      "epoch": 1.3080273802115743,
      "grad_norm": 2.9170854091644287,
      "learning_rate": 3.166512915129152e-06,
      "loss": 1.4698,
      "step": 1051
    },
    {
      "epoch": 1.3092719352831363,
      "grad_norm": 2.9447641372680664,
      "learning_rate": 3.164206642066421e-06,
      "loss": 1.4571,
      "step": 1052
    },
    {
      "epoch": 1.3105164903546982,
      "grad_norm": 2.895512342453003,
      "learning_rate": 3.1619003690036902e-06,
      "loss": 1.4373,
      "step": 1053
    },
    {
      "epoch": 1.31176104542626,
      "grad_norm": 2.82749605178833,
      "learning_rate": 3.15959409594096e-06,
      "loss": 1.4487,
      "step": 1054
    },
    {
      "epoch": 1.3130056004978221,
      "grad_norm": 2.9615845680236816,
      "learning_rate": 3.157287822878229e-06,
      "loss": 1.4917,
      "step": 1055
    },
    {
      "epoch": 1.314250155569384,
      "grad_norm": 2.8264551162719727,
      "learning_rate": 3.1549815498154985e-06,
      "loss": 1.4197,
      "step": 1056
    },
    {
      "epoch": 1.3154947106409458,
      "grad_norm": 2.8582863807678223,
      "learning_rate": 3.152675276752768e-06,
      "loss": 1.4882,
      "step": 1057
    },
    {
      "epoch": 1.3167392657125077,
      "grad_norm": 3.049182653427124,
      "learning_rate": 3.150369003690037e-06,
      "loss": 1.5096,
      "step": 1058
    },
    {
      "epoch": 1.3179838207840697,
      "grad_norm": 2.803922653198242,
      "learning_rate": 3.1480627306273063e-06,
      "loss": 1.4287,
      "step": 1059
    },
    {
      "epoch": 1.3192283758556316,
      "grad_norm": 2.8965165615081787,
      "learning_rate": 3.145756457564576e-06,
      "loss": 1.4786,
      "step": 1060
    },
    {
      "epoch": 1.3192283758556316,
      "eval_loss": 1.6656724214553833,
      "eval_runtime": 53.0223,
      "eval_samples_per_second": 18.86,
      "eval_steps_per_second": 0.792,
      "step": 1060
    },
    {
      "epoch": 1.3204729309271936,
      "grad_norm": 3.2148702144622803,
      "learning_rate": 3.143450184501845e-06,
      "loss": 1.4997,
      "step": 1061
    },
    {
      "epoch": 1.3217174859987555,
      "grad_norm": 3.013700246810913,
      "learning_rate": 3.1411439114391146e-06,
      "loss": 1.4835,
      "step": 1062
    },
    {
      "epoch": 1.3229620410703173,
      "grad_norm": 2.879260540008545,
      "learning_rate": 3.138837638376384e-06,
      "loss": 1.42,
      "step": 1063
    },
    {
      "epoch": 1.3242065961418792,
      "grad_norm": 2.6866135597229004,
      "learning_rate": 3.1365313653136535e-06,
      "loss": 1.4037,
      "step": 1064
    },
    {
      "epoch": 1.3254511512134413,
      "grad_norm": 2.8592655658721924,
      "learning_rate": 3.1342250922509225e-06,
      "loss": 1.443,
      "step": 1065
    },
    {
      "epoch": 1.326695706285003,
      "grad_norm": 3.1136722564697266,
      "learning_rate": 3.1319188191881923e-06,
      "loss": 1.469,
      "step": 1066
    },
    {
      "epoch": 1.327940261356565,
      "grad_norm": 2.8858542442321777,
      "learning_rate": 3.1296125461254617e-06,
      "loss": 1.4525,
      "step": 1067
    },
    {
      "epoch": 1.329184816428127,
      "grad_norm": 3.01088809967041,
      "learning_rate": 3.1273062730627307e-06,
      "loss": 1.4017,
      "step": 1068
    },
    {
      "epoch": 1.3304293714996889,
      "grad_norm": 3.185218334197998,
      "learning_rate": 3.125e-06,
      "loss": 1.4936,
      "step": 1069
    },
    {
      "epoch": 1.3316739265712507,
      "grad_norm": 3.168509006500244,
      "learning_rate": 3.12269372693727e-06,
      "loss": 1.4512,
      "step": 1070
    },
    {
      "epoch": 1.3316739265712507,
      "eval_loss": 1.66367506980896,
      "eval_runtime": 51.2724,
      "eval_samples_per_second": 19.504,
      "eval_steps_per_second": 0.819,
      "step": 1070
    },
    {
      "epoch": 1.3329184816428126,
      "grad_norm": 3.036905527114868,
      "learning_rate": 3.120387453874539e-06,
      "loss": 1.4057,
      "step": 1071
    },
    {
      "epoch": 1.3341630367143746,
      "grad_norm": 3.114818572998047,
      "learning_rate": 3.1180811808118084e-06,
      "loss": 1.4687,
      "step": 1072
    },
    {
      "epoch": 1.3354075917859365,
      "grad_norm": 2.929135799407959,
      "learning_rate": 3.115774907749078e-06,
      "loss": 1.5013,
      "step": 1073
    },
    {
      "epoch": 1.3366521468574986,
      "grad_norm": 3.1101112365722656,
      "learning_rate": 3.113468634686347e-06,
      "loss": 1.4222,
      "step": 1074
    },
    {
      "epoch": 1.3378967019290604,
      "grad_norm": 2.9812445640563965,
      "learning_rate": 3.1111623616236163e-06,
      "loss": 1.4476,
      "step": 1075
    },
    {
      "epoch": 1.3391412570006223,
      "grad_norm": 2.7879486083984375,
      "learning_rate": 3.108856088560886e-06,
      "loss": 1.4228,
      "step": 1076
    },
    {
      "epoch": 1.340385812072184,
      "grad_norm": 2.9183881282806396,
      "learning_rate": 3.1065498154981555e-06,
      "loss": 1.4768,
      "step": 1077
    },
    {
      "epoch": 1.3416303671437462,
      "grad_norm": 3.136688709259033,
      "learning_rate": 3.1042435424354245e-06,
      "loss": 1.4867,
      "step": 1078
    },
    {
      "epoch": 1.342874922215308,
      "grad_norm": 3.087040424346924,
      "learning_rate": 3.101937269372694e-06,
      "loss": 1.4912,
      "step": 1079
    },
    {
      "epoch": 1.3441194772868699,
      "grad_norm": 2.981447696685791,
      "learning_rate": 3.0996309963099634e-06,
      "loss": 1.4133,
      "step": 1080
    },
    {
      "epoch": 1.3441194772868699,
      "eval_loss": 1.6594599485397339,
      "eval_runtime": 50.0245,
      "eval_samples_per_second": 19.99,
      "eval_steps_per_second": 0.84,
      "step": 1080
    },
    {
      "epoch": 1.345364032358432,
      "grad_norm": 3.1392548084259033,
      "learning_rate": 3.0973247232472324e-06,
      "loss": 1.4131,
      "step": 1081
    },
    {
      "epoch": 1.3466085874299938,
      "grad_norm": 2.813328742980957,
      "learning_rate": 3.0950184501845022e-06,
      "loss": 1.4461,
      "step": 1082
    },
    {
      "epoch": 1.3478531425015556,
      "grad_norm": 2.930349111557007,
      "learning_rate": 3.0927121771217716e-06,
      "loss": 1.4423,
      "step": 1083
    },
    {
      "epoch": 1.3490976975731175,
      "grad_norm": 2.8514888286590576,
      "learning_rate": 3.0904059040590406e-06,
      "loss": 1.4499,
      "step": 1084
    },
    {
      "epoch": 1.3503422526446796,
      "grad_norm": 2.8733901977539062,
      "learning_rate": 3.08809963099631e-06,
      "loss": 1.4425,
      "step": 1085
    },
    {
      "epoch": 1.3515868077162414,
      "grad_norm": 2.8833565711975098,
      "learning_rate": 3.0857933579335795e-06,
      "loss": 1.4808,
      "step": 1086
    },
    {
      "epoch": 1.3528313627878035,
      "grad_norm": 2.9774460792541504,
      "learning_rate": 3.0834870848708485e-06,
      "loss": 1.5147,
      "step": 1087
    },
    {
      "epoch": 1.3540759178593653,
      "grad_norm": 2.9540138244628906,
      "learning_rate": 3.0811808118081183e-06,
      "loss": 1.4577,
      "step": 1088
    },
    {
      "epoch": 1.3553204729309272,
      "grad_norm": 2.9065780639648438,
      "learning_rate": 3.0788745387453878e-06,
      "loss": 1.4564,
      "step": 1089
    },
    {
      "epoch": 1.356565028002489,
      "grad_norm": 3.2054378986358643,
      "learning_rate": 3.076568265682657e-06,
      "loss": 1.4435,
      "step": 1090
    },
    {
      "epoch": 1.356565028002489,
      "eval_loss": 1.6565171480178833,
      "eval_runtime": 51.0475,
      "eval_samples_per_second": 19.59,
      "eval_steps_per_second": 0.823,
      "step": 1090
    },
    {
      "epoch": 1.357809583074051,
      "grad_norm": 2.9422333240509033,
      "learning_rate": 3.074261992619926e-06,
      "loss": 1.4844,
      "step": 1091
    },
    {
      "epoch": 1.359054138145613,
      "grad_norm": 2.738215684890747,
      "learning_rate": 3.071955719557196e-06,
      "loss": 1.4349,
      "step": 1092
    },
    {
      "epoch": 1.3602986932171748,
      "grad_norm": 3.032463312149048,
      "learning_rate": 3.0696494464944655e-06,
      "loss": 1.4133,
      "step": 1093
    },
    {
      "epoch": 1.3615432482887369,
      "grad_norm": 2.8251936435699463,
      "learning_rate": 3.0673431734317345e-06,
      "loss": 1.4751,
      "step": 1094
    },
    {
      "epoch": 1.3627878033602987,
      "grad_norm": 2.8674609661102295,
      "learning_rate": 3.065036900369004e-06,
      "loss": 1.464,
      "step": 1095
    },
    {
      "epoch": 1.3640323584318605,
      "grad_norm": 2.8118209838867188,
      "learning_rate": 3.0627306273062733e-06,
      "loss": 1.476,
      "step": 1096
    },
    {
      "epoch": 1.3652769135034224,
      "grad_norm": 3.152803897857666,
      "learning_rate": 3.0604243542435423e-06,
      "loss": 1.4209,
      "step": 1097
    },
    {
      "epoch": 1.3665214685749845,
      "grad_norm": 2.8795511722564697,
      "learning_rate": 3.058118081180812e-06,
      "loss": 1.462,
      "step": 1098
    },
    {
      "epoch": 1.3677660236465463,
      "grad_norm": 2.897827625274658,
      "learning_rate": 3.0558118081180816e-06,
      "loss": 1.4194,
      "step": 1099
    },
    {
      "epoch": 1.3690105787181084,
      "grad_norm": 2.9295172691345215,
      "learning_rate": 3.053505535055351e-06,
      "loss": 1.4742,
      "step": 1100
    },
    {
      "epoch": 1.3690105787181084,
      "eval_loss": 1.6628323793411255,
      "eval_runtime": 52.0227,
      "eval_samples_per_second": 19.222,
      "eval_steps_per_second": 0.807,
      "step": 1100
    },
    {
      "epoch": 1.3702551337896702,
      "grad_norm": 2.79770565032959,
      "learning_rate": 3.05119926199262e-06,
      "loss": 1.3923,
      "step": 1101
    },
    {
      "epoch": 1.371499688861232,
      "grad_norm": 2.8151590824127197,
      "learning_rate": 3.0488929889298894e-06,
      "loss": 1.4396,
      "step": 1102
    },
    {
      "epoch": 1.372744243932794,
      "grad_norm": 2.9383020401000977,
      "learning_rate": 3.0465867158671593e-06,
      "loss": 1.476,
      "step": 1103
    },
    {
      "epoch": 1.373988799004356,
      "grad_norm": 2.8010449409484863,
      "learning_rate": 3.0442804428044283e-06,
      "loss": 1.4579,
      "step": 1104
    },
    {
      "epoch": 1.3752333540759178,
      "grad_norm": 2.8714241981506348,
      "learning_rate": 3.0419741697416977e-06,
      "loss": 1.4921,
      "step": 1105
    },
    {
      "epoch": 1.3764779091474797,
      "grad_norm": 3.0715506076812744,
      "learning_rate": 3.039667896678967e-06,
      "loss": 1.4102,
      "step": 1106
    },
    {
      "epoch": 1.3777224642190418,
      "grad_norm": 2.8336658477783203,
      "learning_rate": 3.037361623616236e-06,
      "loss": 1.4065,
      "step": 1107
    },
    {
      "epoch": 1.3789670192906036,
      "grad_norm": 2.833709716796875,
      "learning_rate": 3.035055350553506e-06,
      "loss": 1.3485,
      "step": 1108
    },
    {
      "epoch": 1.3802115743621655,
      "grad_norm": 2.970330238342285,
      "learning_rate": 3.0327490774907754e-06,
      "loss": 1.3942,
      "step": 1109
    },
    {
      "epoch": 1.3814561294337273,
      "grad_norm": 2.932877779006958,
      "learning_rate": 3.0304428044280444e-06,
      "loss": 1.4281,
      "step": 1110
    },
    {
      "epoch": 1.3814561294337273,
      "eval_loss": 1.6577523946762085,
      "eval_runtime": 42.2764,
      "eval_samples_per_second": 23.654,
      "eval_steps_per_second": 0.993,
      "step": 1110
    },
    {
      "epoch": 1.3827006845052894,
      "grad_norm": 3.1917965412139893,
      "learning_rate": 3.028136531365314e-06,
      "loss": 1.4662,
      "step": 1111
    },
    {
      "epoch": 1.3839452395768512,
      "grad_norm": 2.867098331451416,
      "learning_rate": 3.0258302583025832e-06,
      "loss": 1.4471,
      "step": 1112
    },
    {
      "epoch": 1.3851897946484133,
      "grad_norm": 2.8024351596832275,
      "learning_rate": 3.023523985239853e-06,
      "loss": 1.4211,
      "step": 1113
    },
    {
      "epoch": 1.3864343497199751,
      "grad_norm": 2.7674360275268555,
      "learning_rate": 3.021217712177122e-06,
      "loss": 1.3762,
      "step": 1114
    },
    {
      "epoch": 1.387678904791537,
      "grad_norm": 2.873400926589966,
      "learning_rate": 3.0189114391143915e-06,
      "loss": 1.4825,
      "step": 1115
    },
    {
      "epoch": 1.3889234598630988,
      "grad_norm": 2.9294228553771973,
      "learning_rate": 3.016605166051661e-06,
      "loss": 1.4855,
      "step": 1116
    },
    {
      "epoch": 1.390168014934661,
      "grad_norm": 2.8885910511016846,
      "learning_rate": 3.01429889298893e-06,
      "loss": 1.4196,
      "step": 1117
    },
    {
      "epoch": 1.3914125700062228,
      "grad_norm": 2.7579641342163086,
      "learning_rate": 3.0119926199261993e-06,
      "loss": 1.4174,
      "step": 1118
    },
    {
      "epoch": 1.3926571250777848,
      "grad_norm": 2.8475749492645264,
      "learning_rate": 3.009686346863469e-06,
      "loss": 1.4271,
      "step": 1119
    },
    {
      "epoch": 1.3939016801493467,
      "grad_norm": 2.8472487926483154,
      "learning_rate": 3.007380073800738e-06,
      "loss": 1.4649,
      "step": 1120
    },
    {
      "epoch": 1.3939016801493467,
      "eval_loss": 1.6544252634048462,
      "eval_runtime": 45.034,
      "eval_samples_per_second": 22.205,
      "eval_steps_per_second": 0.933,
      "step": 1120
    },
    {
      "epoch": 1.3951462352209085,
      "grad_norm": 2.916067123413086,
      "learning_rate": 3.0050738007380076e-06,
      "loss": 1.4134,
      "step": 1121
    },
    {
      "epoch": 1.3963907902924704,
      "grad_norm": 2.8867766857147217,
      "learning_rate": 3.002767527675277e-06,
      "loss": 1.4378,
      "step": 1122
    },
    {
      "epoch": 1.3976353453640322,
      "grad_norm": 2.844597101211548,
      "learning_rate": 3.000461254612546e-06,
      "loss": 1.4521,
      "step": 1123
    },
    {
      "epoch": 1.3988799004355943,
      "grad_norm": 2.850982427597046,
      "learning_rate": 2.998154981549816e-06,
      "loss": 1.3948,
      "step": 1124
    },
    {
      "epoch": 1.4001244555071561,
      "grad_norm": 2.813525438308716,
      "learning_rate": 2.9958487084870853e-06,
      "loss": 1.457,
      "step": 1125
    },
    {
      "epoch": 1.4013690105787182,
      "grad_norm": 2.9029061794281006,
      "learning_rate": 2.9935424354243547e-06,
      "loss": 1.4657,
      "step": 1126
    },
    {
      "epoch": 1.40261356565028,
      "grad_norm": 2.8818368911743164,
      "learning_rate": 2.9912361623616237e-06,
      "loss": 1.4283,
      "step": 1127
    },
    {
      "epoch": 1.403858120721842,
      "grad_norm": 2.8317973613739014,
      "learning_rate": 2.988929889298893e-06,
      "loss": 1.4019,
      "step": 1128
    },
    {
      "epoch": 1.4051026757934038,
      "grad_norm": 2.9725966453552246,
      "learning_rate": 2.986623616236163e-06,
      "loss": 1.4425,
      "step": 1129
    },
    {
      "epoch": 1.4063472308649658,
      "grad_norm": 2.977820634841919,
      "learning_rate": 2.984317343173432e-06,
      "loss": 1.4729,
      "step": 1130
    },
    {
      "epoch": 1.4063472308649658,
      "eval_loss": 1.6550770998001099,
      "eval_runtime": 47.3909,
      "eval_samples_per_second": 21.101,
      "eval_steps_per_second": 0.886,
      "step": 1130
    },
    {
      "epoch": 1.4075917859365277,
      "grad_norm": 2.845229148864746,
      "learning_rate": 2.9820110701107014e-06,
      "loss": 1.4531,
      "step": 1131
    },
    {
      "epoch": 1.4088363410080897,
      "grad_norm": 2.883481979370117,
      "learning_rate": 2.979704797047971e-06,
      "loss": 1.4103,
      "step": 1132
    },
    {
      "epoch": 1.4100808960796516,
      "grad_norm": 3.0005011558532715,
      "learning_rate": 2.97739852398524e-06,
      "loss": 1.4381,
      "step": 1133
    },
    {
      "epoch": 1.4113254511512134,
      "grad_norm": 2.88545560836792,
      "learning_rate": 2.9750922509225093e-06,
      "loss": 1.4285,
      "step": 1134
    },
    {
      "epoch": 1.4125700062227753,
      "grad_norm": 2.893871307373047,
      "learning_rate": 2.972785977859779e-06,
      "loss": 1.4578,
      "step": 1135
    },
    {
      "epoch": 1.4138145612943374,
      "grad_norm": 3.0424110889434814,
      "learning_rate": 2.970479704797048e-06,
      "loss": 1.474,
      "step": 1136
    },
    {
      "epoch": 1.4150591163658992,
      "grad_norm": 2.8640952110290527,
      "learning_rate": 2.9681734317343175e-06,
      "loss": 1.4857,
      "step": 1137
    },
    {
      "epoch": 1.416303671437461,
      "grad_norm": 3.0472874641418457,
      "learning_rate": 2.965867158671587e-06,
      "loss": 1.4251,
      "step": 1138
    },
    {
      "epoch": 1.4175482265090231,
      "grad_norm": 2.9340946674346924,
      "learning_rate": 2.9635608856088564e-06,
      "loss": 1.4469,
      "step": 1139
    },
    {
      "epoch": 1.418792781580585,
      "grad_norm": 2.9423937797546387,
      "learning_rate": 2.9612546125461254e-06,
      "loss": 1.4835,
      "step": 1140
    },
    {
      "epoch": 1.418792781580585,
      "eval_loss": 1.647971749305725,
      "eval_runtime": 44.6901,
      "eval_samples_per_second": 22.376,
      "eval_steps_per_second": 0.94,
      "step": 1140
    },
    {
      "epoch": 1.4200373366521468,
      "grad_norm": 2.9077932834625244,
      "learning_rate": 2.958948339483395e-06,
      "loss": 1.4231,
      "step": 1141
    },
    {
      "epoch": 1.4212818917237087,
      "grad_norm": 2.8576467037200928,
      "learning_rate": 2.9566420664206646e-06,
      "loss": 1.4136,
      "step": 1142
    },
    {
      "epoch": 1.4225264467952707,
      "grad_norm": 2.790850877761841,
      "learning_rate": 2.9543357933579336e-06,
      "loss": 1.4181,
      "step": 1143
    },
    {
      "epoch": 1.4237710018668326,
      "grad_norm": 2.889728307723999,
      "learning_rate": 2.952029520295203e-06,
      "loss": 1.4474,
      "step": 1144
    },
    {
      "epoch": 1.4250155569383947,
      "grad_norm": 2.7928731441497803,
      "learning_rate": 2.949723247232473e-06,
      "loss": 1.4235,
      "step": 1145
    },
    {
      "epoch": 1.4262601120099565,
      "grad_norm": 2.8135132789611816,
      "learning_rate": 2.947416974169742e-06,
      "loss": 1.4327,
      "step": 1146
    },
    {
      "epoch": 1.4275046670815184,
      "grad_norm": 2.851341485977173,
      "learning_rate": 2.9451107011070113e-06,
      "loss": 1.4637,
      "step": 1147
    },
    {
      "epoch": 1.4287492221530802,
      "grad_norm": 2.839298725128174,
      "learning_rate": 2.9428044280442807e-06,
      "loss": 1.4882,
      "step": 1148
    },
    {
      "epoch": 1.4299937772246423,
      "grad_norm": 2.8126413822174072,
      "learning_rate": 2.9404981549815497e-06,
      "loss": 1.4219,
      "step": 1149
    },
    {
      "epoch": 1.4312383322962041,
      "grad_norm": 2.892277479171753,
      "learning_rate": 2.938191881918819e-06,
      "loss": 1.4558,
      "step": 1150
    },
    {
      "epoch": 1.4312383322962041,
      "eval_loss": 1.6500873565673828,
      "eval_runtime": 45.6914,
      "eval_samples_per_second": 21.886,
      "eval_steps_per_second": 0.919,
      "step": 1150
    },
    {
      "epoch": 1.432482887367766,
      "grad_norm": 2.9965860843658447,
      "learning_rate": 2.935885608856089e-06,
      "loss": 1.4712,
      "step": 1151
    },
    {
      "epoch": 1.433727442439328,
      "grad_norm": 2.886841297149658,
      "learning_rate": 2.9335793357933584e-06,
      "loss": 1.4475,
      "step": 1152
    },
    {
      "epoch": 1.4349719975108899,
      "grad_norm": 2.8532493114471436,
      "learning_rate": 2.9312730627306274e-06,
      "loss": 1.4436,
      "step": 1153
    },
    {
      "epoch": 1.4362165525824517,
      "grad_norm": 2.873619794845581,
      "learning_rate": 2.928966789667897e-06,
      "loss": 1.4505,
      "step": 1154
    },
    {
      "epoch": 1.4374611076540136,
      "grad_norm": 2.8811511993408203,
      "learning_rate": 2.9266605166051663e-06,
      "loss": 1.4196,
      "step": 1155
    },
    {
      "epoch": 1.4387056627255757,
      "grad_norm": 2.8095715045928955,
      "learning_rate": 2.9243542435424353e-06,
      "loss": 1.46,
      "step": 1156
    },
    {
      "epoch": 1.4399502177971375,
      "grad_norm": 2.8669135570526123,
      "learning_rate": 2.922047970479705e-06,
      "loss": 1.4121,
      "step": 1157
    },
    {
      "epoch": 1.4411947728686996,
      "grad_norm": 2.9843199253082275,
      "learning_rate": 2.9197416974169746e-06,
      "loss": 1.4847,
      "step": 1158
    },
    {
      "epoch": 1.4424393279402614,
      "grad_norm": 2.92252516746521,
      "learning_rate": 2.9174354243542436e-06,
      "loss": 1.4334,
      "step": 1159
    },
    {
      "epoch": 1.4436838830118233,
      "grad_norm": 2.8623733520507812,
      "learning_rate": 2.915129151291513e-06,
      "loss": 1.4292,
      "step": 1160
    },
    {
      "epoch": 1.4436838830118233,
      "eval_loss": 1.6537351608276367,
      "eval_runtime": 45.0702,
      "eval_samples_per_second": 22.188,
      "eval_steps_per_second": 0.932,
      "step": 1160
    },
    {
      "epoch": 1.4449284380833851,
      "grad_norm": 2.9631102085113525,
      "learning_rate": 2.912822878228783e-06,
      "loss": 1.4855,
      "step": 1161
    },
    {
      "epoch": 1.4461729931549472,
      "grad_norm": 2.9020369052886963,
      "learning_rate": 2.9105166051660522e-06,
      "loss": 1.4229,
      "step": 1162
    },
    {
      "epoch": 1.447417548226509,
      "grad_norm": 2.7908735275268555,
      "learning_rate": 2.9082103321033212e-06,
      "loss": 1.4329,
      "step": 1163
    },
    {
      "epoch": 1.4486621032980709,
      "grad_norm": 2.8533918857574463,
      "learning_rate": 2.9059040590405907e-06,
      "loss": 1.447,
      "step": 1164
    },
    {
      "epoch": 1.449906658369633,
      "grad_norm": 2.7187347412109375,
      "learning_rate": 2.90359778597786e-06,
      "loss": 1.4167,
      "step": 1165
    },
    {
      "epoch": 1.4511512134411948,
      "grad_norm": 2.906304359436035,
      "learning_rate": 2.901291512915129e-06,
      "loss": 1.4229,
      "step": 1166
    },
    {
      "epoch": 1.4523957685127566,
      "grad_norm": 2.8748841285705566,
      "learning_rate": 2.898985239852399e-06,
      "loss": 1.4472,
      "step": 1167
    },
    {
      "epoch": 1.4536403235843185,
      "grad_norm": 2.925084114074707,
      "learning_rate": 2.8966789667896684e-06,
      "loss": 1.4257,
      "step": 1168
    },
    {
      "epoch": 1.4548848786558806,
      "grad_norm": 2.7901270389556885,
      "learning_rate": 2.8943726937269374e-06,
      "loss": 1.4698,
      "step": 1169
    },
    {
      "epoch": 1.4561294337274424,
      "grad_norm": 2.980588436126709,
      "learning_rate": 2.8920664206642068e-06,
      "loss": 1.4812,
      "step": 1170
    },
    {
      "epoch": 1.4561294337274424,
      "eval_loss": 1.6486413478851318,
      "eval_runtime": 51.0519,
      "eval_samples_per_second": 19.588,
      "eval_steps_per_second": 0.823,
      "step": 1170
    },
    {
      "epoch": 1.4573739887990045,
      "grad_norm": 3.0290639400482178,
      "learning_rate": 2.889760147601476e-06,
      "loss": 1.4307,
      "step": 1171
    },
    {
      "epoch": 1.4586185438705663,
      "grad_norm": 2.8323185443878174,
      "learning_rate": 2.887453874538745e-06,
      "loss": 1.412,
      "step": 1172
    },
    {
      "epoch": 1.4598630989421282,
      "grad_norm": 2.880540609359741,
      "learning_rate": 2.885147601476015e-06,
      "loss": 1.4255,
      "step": 1173
    },
    {
      "epoch": 1.46110765401369,
      "grad_norm": 2.817723512649536,
      "learning_rate": 2.8828413284132845e-06,
      "loss": 1.426,
      "step": 1174
    },
    {
      "epoch": 1.462352209085252,
      "grad_norm": 2.81062912940979,
      "learning_rate": 2.880535055350554e-06,
      "loss": 1.4441,
      "step": 1175
    },
    {
      "epoch": 1.463596764156814,
      "grad_norm": 2.9940171241760254,
      "learning_rate": 2.878228782287823e-06,
      "loss": 1.4418,
      "step": 1176
    },
    {
      "epoch": 1.4648413192283758,
      "grad_norm": 2.8164620399475098,
      "learning_rate": 2.8759225092250923e-06,
      "loss": 1.4671,
      "step": 1177
    },
    {
      "epoch": 1.4660858742999379,
      "grad_norm": 2.830596446990967,
      "learning_rate": 2.873616236162362e-06,
      "loss": 1.4699,
      "step": 1178
    },
    {
      "epoch": 1.4673304293714997,
      "grad_norm": 2.7963128089904785,
      "learning_rate": 2.871309963099631e-06,
      "loss": 1.4217,
      "step": 1179
    },
    {
      "epoch": 1.4685749844430616,
      "grad_norm": 2.8172154426574707,
      "learning_rate": 2.8690036900369006e-06,
      "loss": 1.4675,
      "step": 1180
    },
    {
      "epoch": 1.4685749844430616,
      "eval_loss": 1.6465628147125244,
      "eval_runtime": 45.7709,
      "eval_samples_per_second": 21.848,
      "eval_steps_per_second": 0.918,
      "step": 1180
    },
    {
      "epoch": 1.4698195395146234,
      "grad_norm": 2.886220693588257,
      "learning_rate": 2.86669741697417e-06,
      "loss": 1.4212,
      "step": 1181
    },
    {
      "epoch": 1.4710640945861855,
      "grad_norm": 2.925135374069214,
      "learning_rate": 2.864391143911439e-06,
      "loss": 1.4341,
      "step": 1182
    },
    {
      "epoch": 1.4723086496577473,
      "grad_norm": 3.0038998126983643,
      "learning_rate": 2.862084870848709e-06,
      "loss": 1.5181,
      "step": 1183
    },
    {
      "epoch": 1.4735532047293094,
      "grad_norm": 2.871967315673828,
      "learning_rate": 2.8597785977859783e-06,
      "loss": 1.4243,
      "step": 1184
    },
    {
      "epoch": 1.4747977598008712,
      "grad_norm": 2.966698408126831,
      "learning_rate": 2.8574723247232473e-06,
      "loss": 1.4968,
      "step": 1185
    },
    {
      "epoch": 1.476042314872433,
      "grad_norm": 2.8052289485931396,
      "learning_rate": 2.8551660516605167e-06,
      "loss": 1.3491,
      "step": 1186
    },
    {
      "epoch": 1.477286869943995,
      "grad_norm": 2.932821035385132,
      "learning_rate": 2.852859778597786e-06,
      "loss": 1.437,
      "step": 1187
    },
    {
      "epoch": 1.478531425015557,
      "grad_norm": 2.780231475830078,
      "learning_rate": 2.850553505535056e-06,
      "loss": 1.3966,
      "step": 1188
    },
    {
      "epoch": 1.4797759800871189,
      "grad_norm": 2.8993380069732666,
      "learning_rate": 2.848247232472325e-06,
      "loss": 1.4059,
      "step": 1189
    },
    {
      "epoch": 1.4810205351586807,
      "grad_norm": 3.020855665206909,
      "learning_rate": 2.8459409594095944e-06,
      "loss": 1.4047,
      "step": 1190
    },
    {
      "epoch": 1.4810205351586807,
      "eval_loss": 1.649813175201416,
      "eval_runtime": 48.4672,
      "eval_samples_per_second": 20.633,
      "eval_steps_per_second": 0.867,
      "step": 1190
    },
    {
      "epoch": 1.4822650902302428,
      "grad_norm": 2.9288201332092285,
      "learning_rate": 2.843634686346864e-06,
      "loss": 1.4481,
      "step": 1191
    },
    {
      "epoch": 1.4835096453018046,
      "grad_norm": 2.8591952323913574,
      "learning_rate": 2.841328413284133e-06,
      "loss": 1.4509,
      "step": 1192
    },
    {
      "epoch": 1.4847542003733665,
      "grad_norm": 2.869105339050293,
      "learning_rate": 2.8390221402214022e-06,
      "loss": 1.4371,
      "step": 1193
    },
    {
      "epoch": 1.4859987554449283,
      "grad_norm": 2.9339916706085205,
      "learning_rate": 2.836715867158672e-06,
      "loss": 1.4562,
      "step": 1194
    },
    {
      "epoch": 1.4872433105164904,
      "grad_norm": 3.259225606918335,
      "learning_rate": 2.834409594095941e-06,
      "loss": 1.4495,
      "step": 1195
    },
    {
      "epoch": 1.4884878655880522,
      "grad_norm": 3.1184468269348145,
      "learning_rate": 2.8321033210332105e-06,
      "loss": 1.4542,
      "step": 1196
    },
    {
      "epoch": 1.4897324206596143,
      "grad_norm": 2.828624963760376,
      "learning_rate": 2.82979704797048e-06,
      "loss": 1.4254,
      "step": 1197
    },
    {
      "epoch": 1.4909769757311762,
      "grad_norm": 2.80483078956604,
      "learning_rate": 2.827490774907749e-06,
      "loss": 1.4673,
      "step": 1198
    },
    {
      "epoch": 1.492221530802738,
      "grad_norm": 2.8064558506011963,
      "learning_rate": 2.8251845018450188e-06,
      "loss": 1.4469,
      "step": 1199
    },
    {
      "epoch": 1.4934660858742999,
      "grad_norm": 2.9313275814056396,
      "learning_rate": 2.822878228782288e-06,
      "loss": 1.5039,
      "step": 1200
    },
    {
      "epoch": 1.4934660858742999,
      "eval_loss": 1.6433771848678589,
      "eval_runtime": 45.402,
      "eval_samples_per_second": 22.025,
      "eval_steps_per_second": 0.925,
      "step": 1200
    },
    {
      "epoch": 1.494710640945862,
      "grad_norm": 2.7521519660949707,
      "learning_rate": 2.8205719557195576e-06,
      "loss": 1.4288,
      "step": 1201
    },
    {
      "epoch": 1.4959551960174238,
      "grad_norm": 2.8780877590179443,
      "learning_rate": 2.8182656826568266e-06,
      "loss": 1.4388,
      "step": 1202
    },
    {
      "epoch": 1.4971997510889856,
      "grad_norm": 2.8828067779541016,
      "learning_rate": 2.815959409594096e-06,
      "loss": 1.4212,
      "step": 1203
    },
    {
      "epoch": 1.4984443061605477,
      "grad_norm": 2.959860324859619,
      "learning_rate": 2.813653136531366e-06,
      "loss": 1.4021,
      "step": 1204
    },
    {
      "epoch": 1.4996888612321095,
      "grad_norm": 3.159862518310547,
      "learning_rate": 2.811346863468635e-06,
      "loss": 1.4543,
      "step": 1205
    },
    {
      "epoch": 1.5009334163036714,
      "grad_norm": 2.742307424545288,
      "learning_rate": 2.8090405904059043e-06,
      "loss": 1.4301,
      "step": 1206
    },
    {
      "epoch": 1.5021779713752332,
      "grad_norm": 2.9725406169891357,
      "learning_rate": 2.8067343173431737e-06,
      "loss": 1.412,
      "step": 1207
    },
    {
      "epoch": 1.5034225264467953,
      "grad_norm": 2.866685152053833,
      "learning_rate": 2.8044280442804427e-06,
      "loss": 1.4234,
      "step": 1208
    },
    {
      "epoch": 1.5046670815183572,
      "grad_norm": 2.8023900985717773,
      "learning_rate": 2.802121771217712e-06,
      "loss": 1.4927,
      "step": 1209
    },
    {
      "epoch": 1.5059116365899192,
      "grad_norm": 2.830970048904419,
      "learning_rate": 2.799815498154982e-06,
      "loss": 1.4373,
      "step": 1210
    },
    {
      "epoch": 1.5059116365899192,
      "eval_loss": 1.651065468788147,
      "eval_runtime": 45.9527,
      "eval_samples_per_second": 21.761,
      "eval_steps_per_second": 0.914,
      "step": 1210
    },
    {
      "epoch": 1.507156191661481,
      "grad_norm": 2.934990167617798,
      "learning_rate": 2.7975092250922514e-06,
      "loss": 1.485,
      "step": 1211
    },
    {
      "epoch": 1.508400746733043,
      "grad_norm": 2.8635311126708984,
      "learning_rate": 2.7952029520295204e-06,
      "loss": 1.4389,
      "step": 1212
    },
    {
      "epoch": 1.5096453018046048,
      "grad_norm": 2.868851661682129,
      "learning_rate": 2.79289667896679e-06,
      "loss": 1.4481,
      "step": 1213
    },
    {
      "epoch": 1.5108898568761666,
      "grad_norm": 2.9756863117218018,
      "learning_rate": 2.7905904059040597e-06,
      "loss": 1.4425,
      "step": 1214
    },
    {
      "epoch": 1.5121344119477287,
      "grad_norm": 2.8925342559814453,
      "learning_rate": 2.7882841328413287e-06,
      "loss": 1.4497,
      "step": 1215
    },
    {
      "epoch": 1.5133789670192908,
      "grad_norm": 2.886101007461548,
      "learning_rate": 2.785977859778598e-06,
      "loss": 1.4772,
      "step": 1216
    },
    {
      "epoch": 1.5146235220908526,
      "grad_norm": 2.8732335567474365,
      "learning_rate": 2.7836715867158675e-06,
      "loss": 1.4351,
      "step": 1217
    },
    {
      "epoch": 1.5158680771624145,
      "grad_norm": 2.938364028930664,
      "learning_rate": 2.7813653136531365e-06,
      "loss": 1.3822,
      "step": 1218
    },
    {
      "epoch": 1.5171126322339763,
      "grad_norm": 2.8434407711029053,
      "learning_rate": 2.779059040590406e-06,
      "loss": 1.3868,
      "step": 1219
    },
    {
      "epoch": 1.5183571873055381,
      "grad_norm": 2.9862983226776123,
      "learning_rate": 2.776752767527676e-06,
      "loss": 1.4333,
      "step": 1220
    },
    {
      "epoch": 1.5183571873055381,
      "eval_loss": 1.6455049514770508,
      "eval_runtime": 47.7092,
      "eval_samples_per_second": 20.96,
      "eval_steps_per_second": 0.88,
      "step": 1220
    },
    {
      "epoch": 1.5196017423771002,
      "grad_norm": 3.222745656967163,
      "learning_rate": 2.774446494464945e-06,
      "loss": 1.4321,
      "step": 1221
    },
    {
      "epoch": 1.520846297448662,
      "grad_norm": 2.8184573650360107,
      "learning_rate": 2.7721402214022142e-06,
      "loss": 1.4346,
      "step": 1222
    },
    {
      "epoch": 1.5220908525202241,
      "grad_norm": 2.867112636566162,
      "learning_rate": 2.7698339483394837e-06,
      "loss": 1.4359,
      "step": 1223
    },
    {
      "epoch": 1.523335407591786,
      "grad_norm": 3.063075304031372,
      "learning_rate": 2.767527675276753e-06,
      "loss": 1.3919,
      "step": 1224
    },
    {
      "epoch": 1.5245799626633478,
      "grad_norm": 2.830925941467285,
      "learning_rate": 2.765221402214022e-06,
      "loss": 1.4133,
      "step": 1225
    },
    {
      "epoch": 1.5258245177349097,
      "grad_norm": 2.937262535095215,
      "learning_rate": 2.762915129151292e-06,
      "loss": 1.4079,
      "step": 1226
    },
    {
      "epoch": 1.5270690728064715,
      "grad_norm": 2.9446723461151123,
      "learning_rate": 2.7606088560885613e-06,
      "loss": 1.4169,
      "step": 1227
    },
    {
      "epoch": 1.5283136278780336,
      "grad_norm": 2.8828818798065186,
      "learning_rate": 2.7583025830258303e-06,
      "loss": 1.3707,
      "step": 1228
    },
    {
      "epoch": 1.5295581829495957,
      "grad_norm": 2.8671786785125732,
      "learning_rate": 2.7559963099630998e-06,
      "loss": 1.4999,
      "step": 1229
    },
    {
      "epoch": 1.5308027380211575,
      "grad_norm": 3.20247745513916,
      "learning_rate": 2.753690036900369e-06,
      "loss": 1.3893,
      "step": 1230
    },
    {
      "epoch": 1.5308027380211575,
      "eval_loss": 1.6418426036834717,
      "eval_runtime": 50.6885,
      "eval_samples_per_second": 19.728,
      "eval_steps_per_second": 0.829,
      "step": 1230
    },
    {
      "epoch": 1.5320472930927194,
      "grad_norm": 2.7745611667633057,
      "learning_rate": 2.751383763837638e-06,
      "loss": 1.3804,
      "step": 1231
    },
    {
      "epoch": 1.5332918481642812,
      "grad_norm": 2.93796706199646,
      "learning_rate": 2.749077490774908e-06,
      "loss": 1.4082,
      "step": 1232
    },
    {
      "epoch": 1.534536403235843,
      "grad_norm": 2.9986002445220947,
      "learning_rate": 2.7467712177121775e-06,
      "loss": 1.5093,
      "step": 1233
    },
    {
      "epoch": 1.5357809583074051,
      "grad_norm": 2.8558406829833984,
      "learning_rate": 2.7444649446494465e-06,
      "loss": 1.3856,
      "step": 1234
    },
    {
      "epoch": 1.537025513378967,
      "grad_norm": 2.8558406829833984,
      "learning_rate": 2.7444649446494465e-06,
      "loss": 1.4818,
      "step": 1235
    },
    {
      "epoch": 1.538270068450529,
      "grad_norm": 2.7725794315338135,
      "learning_rate": 2.742158671586716e-06,
      "loss": 1.4256,
      "step": 1236
    },
    {
      "epoch": 1.539514623522091,
      "grad_norm": 2.9687700271606445,
      "learning_rate": 2.7398523985239857e-06,
      "loss": 1.4623,
      "step": 1237
    },
    {
      "epoch": 1.5407591785936527,
      "grad_norm": 2.709195613861084,
      "learning_rate": 2.737546125461255e-06,
      "loss": 1.4665,
      "step": 1238
    },
    {
      "epoch": 1.5420037336652146,
      "grad_norm": 2.9164133071899414,
      "learning_rate": 2.735239852398524e-06,
      "loss": 1.4623,
      "step": 1239
    },
    {
      "epoch": 1.5432482887367764,
      "grad_norm": 3.0910398960113525,
      "learning_rate": 2.7329335793357936e-06,
      "loss": 1.4771,
      "step": 1240
    },
    {
      "epoch": 1.5432482887367764,
      "eval_loss": 1.6488943099975586,
      "eval_runtime": 50.98,
      "eval_samples_per_second": 19.616,
      "eval_steps_per_second": 0.824,
      "step": 1240
    },
    {
      "epoch": 1.5444928438083385,
      "grad_norm": 2.9204413890838623,
      "learning_rate": 2.730627306273063e-06,
      "loss": 1.4769,
      "step": 1241
    },
    {
      "epoch": 1.5457373988799006,
      "grad_norm": 2.967817783355713,
      "learning_rate": 2.728321033210332e-06,
      "loss": 1.4374,
      "step": 1242
    },
    {
      "epoch": 1.5469819539514624,
      "grad_norm": 2.891555070877075,
      "learning_rate": 2.726014760147602e-06,
      "loss": 1.4116,
      "step": 1243
    },
    {
      "epoch": 1.5482265090230243,
      "grad_norm": 2.882413148880005,
      "learning_rate": 2.7237084870848713e-06,
      "loss": 1.4304,
      "step": 1244
    },
    {
      "epoch": 1.5494710640945861,
      "grad_norm": 2.720031499862671,
      "learning_rate": 2.7214022140221403e-06,
      "loss": 1.4035,
      "step": 1245
    },
    {
      "epoch": 1.550715619166148,
      "grad_norm": 2.890763282775879,
      "learning_rate": 2.7190959409594097e-06,
      "loss": 1.4276,
      "step": 1246
    },
    {
      "epoch": 1.55196017423771,
      "grad_norm": 2.702901840209961,
      "learning_rate": 2.716789667896679e-06,
      "loss": 1.3971,
      "step": 1247
    },
    {
      "epoch": 1.553204729309272,
      "grad_norm": 2.8923733234405518,
      "learning_rate": 2.714483394833948e-06,
      "loss": 1.4242,
      "step": 1248
    },
    {
      "epoch": 1.554449284380834,
      "grad_norm": 2.909700632095337,
      "learning_rate": 2.712177121771218e-06,
      "loss": 1.4599,
      "step": 1249
    },
    {
      "epoch": 1.5556938394523958,
      "grad_norm": 2.9248416423797607,
      "learning_rate": 2.7098708487084874e-06,
      "loss": 1.4367,
      "step": 1250
    },
    {
      "epoch": 1.5556938394523958,
      "eval_loss": 1.6400885581970215,
      "eval_runtime": 52.6283,
      "eval_samples_per_second": 19.001,
      "eval_steps_per_second": 0.798,
      "step": 1250
    },
    {
      "epoch": 1.5569383945239577,
      "grad_norm": 2.9018962383270264,
      "learning_rate": 2.707564575645757e-06,
      "loss": 1.482,
      "step": 1251
    },
    {
      "epoch": 1.5581829495955195,
      "grad_norm": 2.928920030593872,
      "learning_rate": 2.705258302583026e-06,
      "loss": 1.4326,
      "step": 1252
    },
    {
      "epoch": 1.5594275046670814,
      "grad_norm": 2.8409273624420166,
      "learning_rate": 2.7029520295202956e-06,
      "loss": 1.4206,
      "step": 1253
    },
    {
      "epoch": 1.5606720597386434,
      "grad_norm": 2.7638025283813477,
      "learning_rate": 2.700645756457565e-06,
      "loss": 1.3835,
      "step": 1254
    },
    {
      "epoch": 1.5619166148102055,
      "grad_norm": 3.0424954891204834,
      "learning_rate": 2.698339483394834e-06,
      "loss": 1.4695,
      "step": 1255
    },
    {
      "epoch": 1.5631611698817673,
      "grad_norm": 3.0798399448394775,
      "learning_rate": 2.6960332103321035e-06,
      "loss": 1.4698,
      "step": 1256
    },
    {
      "epoch": 1.5644057249533292,
      "grad_norm": 2.9255073070526123,
      "learning_rate": 2.693726937269373e-06,
      "loss": 1.4153,
      "step": 1257
    },
    {
      "epoch": 1.565650280024891,
      "grad_norm": 3.1047182083129883,
      "learning_rate": 2.691420664206642e-06,
      "loss": 1.4391,
      "step": 1258
    },
    {
      "epoch": 1.5668948350964529,
      "grad_norm": 2.8893303871154785,
      "learning_rate": 2.6891143911439118e-06,
      "loss": 1.3855,
      "step": 1259
    },
    {
      "epoch": 1.568139390168015,
      "grad_norm": 2.886103868484497,
      "learning_rate": 2.686808118081181e-06,
      "loss": 1.4359,
      "step": 1260
    },
    {
      "epoch": 1.568139390168015,
      "eval_loss": 1.640094518661499,
      "eval_runtime": 49.7531,
      "eval_samples_per_second": 20.099,
      "eval_steps_per_second": 0.844,
      "step": 1260
    },
    {
      "epoch": 1.5693839452395768,
      "grad_norm": 3.337623357772827,
      "learning_rate": 2.6845018450184506e-06,
      "loss": 1.5257,
      "step": 1261
    },
    {
      "epoch": 1.5706285003111389,
      "grad_norm": 3.0222229957580566,
      "learning_rate": 2.6821955719557196e-06,
      "loss": 1.4221,
      "step": 1262
    },
    {
      "epoch": 1.5718730553827007,
      "grad_norm": 2.967501401901245,
      "learning_rate": 2.679889298892989e-06,
      "loss": 1.5488,
      "step": 1263
    },
    {
      "epoch": 1.5731176104542626,
      "grad_norm": 3.064924716949463,
      "learning_rate": 2.677583025830259e-06,
      "loss": 1.4424,
      "step": 1264
    },
    {
      "epoch": 1.5743621655258244,
      "grad_norm": 2.851073741912842,
      "learning_rate": 2.675276752767528e-06,
      "loss": 1.4266,
      "step": 1265
    },
    {
      "epoch": 1.5756067205973865,
      "grad_norm": 2.922013282775879,
      "learning_rate": 2.6729704797047973e-06,
      "loss": 1.4069,
      "step": 1266
    },
    {
      "epoch": 1.5768512756689483,
      "grad_norm": 2.7312874794006348,
      "learning_rate": 2.6706642066420667e-06,
      "loss": 1.3591,
      "step": 1267
    },
    {
      "epoch": 1.5780958307405104,
      "grad_norm": 3.37092924118042,
      "learning_rate": 2.6683579335793357e-06,
      "loss": 1.4512,
      "step": 1268
    },
    {
      "epoch": 1.5793403858120723,
      "grad_norm": 2.899977684020996,
      "learning_rate": 2.6660516605166056e-06,
      "loss": 1.4177,
      "step": 1269
    },
    {
      "epoch": 1.580584940883634,
      "grad_norm": 2.962331771850586,
      "learning_rate": 2.663745387453875e-06,
      "loss": 1.4522,
      "step": 1270
    },
    {
      "epoch": 1.580584940883634,
      "eval_loss": 1.643125057220459,
      "eval_runtime": 49.7864,
      "eval_samples_per_second": 20.086,
      "eval_steps_per_second": 0.844,
      "step": 1270
    },
    {
      "epoch": 1.581829495955196,
      "grad_norm": 3.224000930786133,
      "learning_rate": 2.661439114391144e-06,
      "loss": 1.4427,
      "step": 1271
    },
    {
      "epoch": 1.5830740510267578,
      "grad_norm": 3.090118169784546,
      "learning_rate": 2.6591328413284134e-06,
      "loss": 1.4308,
      "step": 1272
    },
    {
      "epoch": 1.5843186060983199,
      "grad_norm": 2.877392530441284,
      "learning_rate": 2.656826568265683e-06,
      "loss": 1.4277,
      "step": 1273
    },
    {
      "epoch": 1.5855631611698817,
      "grad_norm": 3.2313079833984375,
      "learning_rate": 2.6545202952029527e-06,
      "loss": 1.4342,
      "step": 1274
    },
    {
      "epoch": 1.5868077162414438,
      "grad_norm": 3.0011425018310547,
      "learning_rate": 2.6522140221402217e-06,
      "loss": 1.4545,
      "step": 1275
    },
    {
      "epoch": 1.5880522713130056,
      "grad_norm": 2.802729845046997,
      "learning_rate": 2.649907749077491e-06,
      "loss": 1.4286,
      "step": 1276
    },
    {
      "epoch": 1.5892968263845675,
      "grad_norm": 3.168306350708008,
      "learning_rate": 2.6476014760147605e-06,
      "loss": 1.4726,
      "step": 1277
    },
    {
      "epoch": 1.5905413814561293,
      "grad_norm": 2.8712313175201416,
      "learning_rate": 2.6452952029520295e-06,
      "loss": 1.424,
      "step": 1278
    },
    {
      "epoch": 1.5917859365276914,
      "grad_norm": 3.0059421062469482,
      "learning_rate": 2.642988929889299e-06,
      "loss": 1.4461,
      "step": 1279
    },
    {
      "epoch": 1.5930304915992533,
      "grad_norm": 2.808736562728882,
      "learning_rate": 2.640682656826569e-06,
      "loss": 1.4507,
      "step": 1280
    },
    {
      "epoch": 1.5930304915992533,
      "eval_loss": 1.6343430280685425,
      "eval_runtime": 52.0843,
      "eval_samples_per_second": 19.2,
      "eval_steps_per_second": 0.806,
      "step": 1280
    },
    {
      "epoch": 1.5942750466708153,
      "grad_norm": 2.65305757522583,
      "learning_rate": 2.638376383763838e-06,
      "loss": 1.3912,
      "step": 1281
    },
    {
      "epoch": 1.5955196017423772,
      "grad_norm": 2.9121081829071045,
      "learning_rate": 2.6360701107011072e-06,
      "loss": 1.3855,
      "step": 1282
    },
    {
      "epoch": 1.596764156813939,
      "grad_norm": 3.035590410232544,
      "learning_rate": 2.6337638376383766e-06,
      "loss": 1.4341,
      "step": 1283
    },
    {
      "epoch": 1.5980087118855009,
      "grad_norm": 2.829501152038574,
      "learning_rate": 2.6314575645756456e-06,
      "loss": 1.4061,
      "step": 1284
    },
    {
      "epoch": 1.5992532669570627,
      "grad_norm": 2.9735665321350098,
      "learning_rate": 2.629151291512915e-06,
      "loss": 1.3729,
      "step": 1285
    },
    {
      "epoch": 1.6004978220286248,
      "grad_norm": 3.0455307960510254,
      "learning_rate": 2.626845018450185e-06,
      "loss": 1.4475,
      "step": 1286
    },
    {
      "epoch": 1.6017423771001866,
      "grad_norm": 3.0120978355407715,
      "learning_rate": 2.6245387453874543e-06,
      "loss": 1.4747,
      "step": 1287
    },
    {
      "epoch": 1.6029869321717487,
      "grad_norm": 2.8382294178009033,
      "learning_rate": 2.6222324723247233e-06,
      "loss": 1.4107,
      "step": 1288
    },
    {
      "epoch": 1.6042314872433105,
      "grad_norm": 2.9768004417419434,
      "learning_rate": 2.6199261992619928e-06,
      "loss": 1.5052,
      "step": 1289
    },
    {
      "epoch": 1.6054760423148724,
      "grad_norm": 2.9081623554229736,
      "learning_rate": 2.6176199261992626e-06,
      "loss": 1.4094,
      "step": 1290
    },
    {
      "epoch": 1.6054760423148724,
      "eval_loss": 1.6426081657409668,
      "eval_runtime": 49.497,
      "eval_samples_per_second": 20.203,
      "eval_steps_per_second": 0.849,
      "step": 1290
    },
    {
      "epoch": 1.6067205973864342,
      "grad_norm": 2.9973902702331543,
      "learning_rate": 2.6153136531365316e-06,
      "loss": 1.4339,
      "step": 1291
    },
    {
      "epoch": 1.6079651524579963,
      "grad_norm": 2.91597843170166,
      "learning_rate": 2.613007380073801e-06,
      "loss": 1.4457,
      "step": 1292
    },
    {
      "epoch": 1.6092097075295582,
      "grad_norm": 2.8896684646606445,
      "learning_rate": 2.6107011070110704e-06,
      "loss": 1.4149,
      "step": 1293
    },
    {
      "epoch": 1.6104542626011202,
      "grad_norm": 2.8653416633605957,
      "learning_rate": 2.6083948339483394e-06,
      "loss": 1.4667,
      "step": 1294
    },
    {
      "epoch": 1.611698817672682,
      "grad_norm": 3.1000678539276123,
      "learning_rate": 2.606088560885609e-06,
      "loss": 1.4776,
      "step": 1295
    },
    {
      "epoch": 1.612943372744244,
      "grad_norm": 2.879951000213623,
      "learning_rate": 2.6037822878228787e-06,
      "loss": 1.4861,
      "step": 1296
    },
    {
      "epoch": 1.6141879278158058,
      "grad_norm": 2.852196455001831,
      "learning_rate": 2.6014760147601477e-06,
      "loss": 1.4013,
      "step": 1297
    },
    {
      "epoch": 1.6154324828873676,
      "grad_norm": 2.9815022945404053,
      "learning_rate": 2.599169741697417e-06,
      "loss": 1.5116,
      "step": 1298
    },
    {
      "epoch": 1.6166770379589297,
      "grad_norm": 2.8925321102142334,
      "learning_rate": 2.5968634686346866e-06,
      "loss": 1.4761,
      "step": 1299
    },
    {
      "epoch": 1.6179215930304915,
      "grad_norm": 2.8847744464874268,
      "learning_rate": 2.594557195571956e-06,
      "loss": 1.4389,
      "step": 1300
    },
    {
      "epoch": 1.6179215930304915,
      "eval_loss": 1.6421130895614624,
      "eval_runtime": 53.2733,
      "eval_samples_per_second": 18.771,
      "eval_steps_per_second": 0.788,
      "step": 1300
    },
    {
      "epoch": 1.6191661481020536,
      "grad_norm": 2.8704094886779785,
      "learning_rate": 2.592250922509225e-06,
      "loss": 1.423,
      "step": 1301
    },
    {
      "epoch": 1.6204107031736155,
      "grad_norm": 3.0075790882110596,
      "learning_rate": 2.589944649446495e-06,
      "loss": 1.4343,
      "step": 1302
    },
    {
      "epoch": 1.6216552582451773,
      "grad_norm": 2.8250370025634766,
      "learning_rate": 2.5876383763837643e-06,
      "loss": 1.3954,
      "step": 1303
    },
    {
      "epoch": 1.6228998133167392,
      "grad_norm": 2.8975305557250977,
      "learning_rate": 2.5853321033210333e-06,
      "loss": 1.441,
      "step": 1304
    },
    {
      "epoch": 1.6241443683883012,
      "grad_norm": 3.0084152221679688,
      "learning_rate": 2.5830258302583027e-06,
      "loss": 1.4447,
      "step": 1305
    },
    {
      "epoch": 1.625388923459863,
      "grad_norm": 2.7868149280548096,
      "learning_rate": 2.5807195571955725e-06,
      "loss": 1.4517,
      "step": 1306
    },
    {
      "epoch": 1.6266334785314251,
      "grad_norm": 2.963439464569092,
      "learning_rate": 2.5784132841328415e-06,
      "loss": 1.4578,
      "step": 1307
    },
    {
      "epoch": 1.627878033602987,
      "grad_norm": 3.0648446083068848,
      "learning_rate": 2.576107011070111e-06,
      "loss": 1.5124,
      "step": 1308
    },
    {
      "epoch": 1.6291225886745488,
      "grad_norm": 2.768556594848633,
      "learning_rate": 2.5738007380073804e-06,
      "loss": 1.4145,
      "step": 1309
    },
    {
      "epoch": 1.6303671437461107,
      "grad_norm": 2.780244827270508,
      "learning_rate": 2.5714944649446494e-06,
      "loss": 1.4298,
      "step": 1310
    },
    {
      "epoch": 1.6303671437461107,
      "eval_loss": 1.6375432014465332,
      "eval_runtime": 43.006,
      "eval_samples_per_second": 23.253,
      "eval_steps_per_second": 0.977,
      "step": 1310
    },
    {
      "epoch": 1.6316116988176725,
      "grad_norm": 2.7755584716796875,
      "learning_rate": 2.5691881918819188e-06,
      "loss": 1.44,
      "step": 1311
    },
    {
      "epoch": 1.6328562538892346,
      "grad_norm": 2.8583791255950928,
      "learning_rate": 2.5668819188191886e-06,
      "loss": 1.4375,
      "step": 1312
    },
    {
      "epoch": 1.6341008089607967,
      "grad_norm": 2.8298938274383545,
      "learning_rate": 2.564575645756458e-06,
      "loss": 1.4171,
      "step": 1313
    },
    {
      "epoch": 1.6353453640323585,
      "grad_norm": 2.79626727104187,
      "learning_rate": 2.562269372693727e-06,
      "loss": 1.3764,
      "step": 1314
    },
    {
      "epoch": 1.6365899191039204,
      "grad_norm": 2.743941307067871,
      "learning_rate": 2.5599630996309965e-06,
      "loss": 1.3855,
      "step": 1315
    },
    {
      "epoch": 1.6378344741754822,
      "grad_norm": 2.950843572616577,
      "learning_rate": 2.557656826568266e-06,
      "loss": 1.4521,
      "step": 1316
    },
    {
      "epoch": 1.639079029247044,
      "grad_norm": 3.0037457942962646,
      "learning_rate": 2.555350553505535e-06,
      "loss": 1.4246,
      "step": 1317
    },
    {
      "epoch": 1.6403235843186061,
      "grad_norm": 2.9074928760528564,
      "learning_rate": 2.5530442804428047e-06,
      "loss": 1.4269,
      "step": 1318
    },
    {
      "epoch": 1.641568139390168,
      "grad_norm": 2.915112257003784,
      "learning_rate": 2.550738007380074e-06,
      "loss": 1.4262,
      "step": 1319
    },
    {
      "epoch": 1.64281269446173,
      "grad_norm": 2.7829935550689697,
      "learning_rate": 2.548431734317343e-06,
      "loss": 1.4143,
      "step": 1320
    },
    {
      "epoch": 1.64281269446173,
      "eval_loss": 1.6351630687713623,
      "eval_runtime": 45.9445,
      "eval_samples_per_second": 21.765,
      "eval_steps_per_second": 0.914,
      "step": 1320
    },
    {
      "epoch": 1.644057249533292,
      "grad_norm": 2.8766672611236572,
      "learning_rate": 2.5461254612546126e-06,
      "loss": 1.3596,
      "step": 1321
    },
    {
      "epoch": 1.6453018046048538,
      "grad_norm": 2.9509501457214355,
      "learning_rate": 2.543819188191882e-06,
      "loss": 1.3766,
      "step": 1322
    },
    {
      "epoch": 1.6465463596764156,
      "grad_norm": 2.993206262588501,
      "learning_rate": 2.541512915129152e-06,
      "loss": 1.4226,
      "step": 1323
    },
    {
      "epoch": 1.6477909147479775,
      "grad_norm": 2.872122049331665,
      "learning_rate": 2.539206642066421e-06,
      "loss": 1.4096,
      "step": 1324
    },
    {
      "epoch": 1.6490354698195395,
      "grad_norm": 3.1725332736968994,
      "learning_rate": 2.5369003690036903e-06,
      "loss": 1.4616,
      "step": 1325
    },
    {
      "epoch": 1.6502800248911016,
      "grad_norm": 2.920024871826172,
      "learning_rate": 2.5345940959409597e-06,
      "loss": 1.4581,
      "step": 1326
    },
    {
      "epoch": 1.6515245799626634,
      "grad_norm": 2.9344615936279297,
      "learning_rate": 2.5322878228782287e-06,
      "loss": 1.4416,
      "step": 1327
    },
    {
      "epoch": 1.6527691350342253,
      "grad_norm": 2.8990018367767334,
      "learning_rate": 2.5299815498154986e-06,
      "loss": 1.4658,
      "step": 1328
    },
    {
      "epoch": 1.6540136901057871,
      "grad_norm": 2.9203572273254395,
      "learning_rate": 2.527675276752768e-06,
      "loss": 1.4132,
      "step": 1329
    },
    {
      "epoch": 1.655258245177349,
      "grad_norm": 3.0077338218688965,
      "learning_rate": 2.525369003690037e-06,
      "loss": 1.4959,
      "step": 1330
    },
    {
      "epoch": 1.655258245177349,
      "eval_loss": 1.6384899616241455,
      "eval_runtime": 44.8728,
      "eval_samples_per_second": 22.285,
      "eval_steps_per_second": 0.936,
      "step": 1330
    },
    {
      "epoch": 1.656502800248911,
      "grad_norm": 2.98171329498291,
      "learning_rate": 2.5230627306273064e-06,
      "loss": 1.3821,
      "step": 1331
    },
    {
      "epoch": 1.657747355320473,
      "grad_norm": 2.848473310470581,
      "learning_rate": 2.520756457564576e-06,
      "loss": 1.4157,
      "step": 1332
    },
    {
      "epoch": 1.658991910392035,
      "grad_norm": 2.8811120986938477,
      "learning_rate": 2.518450184501845e-06,
      "loss": 1.4507,
      "step": 1333
    },
    {
      "epoch": 1.6602364654635968,
      "grad_norm": 2.8421835899353027,
      "learning_rate": 2.5161439114391147e-06,
      "loss": 1.415,
      "step": 1334
    },
    {
      "epoch": 1.6614810205351587,
      "grad_norm": 2.8382010459899902,
      "learning_rate": 2.513837638376384e-06,
      "loss": 1.3882,
      "step": 1335
    },
    {
      "epoch": 1.6627255756067205,
      "grad_norm": 3.009395122528076,
      "learning_rate": 2.5115313653136535e-06,
      "loss": 1.3916,
      "step": 1336
    },
    {
      "epoch": 1.6639701306782824,
      "grad_norm": 2.8583426475524902,
      "learning_rate": 2.5092250922509225e-06,
      "loss": 1.4196,
      "step": 1337
    },
    {
      "epoch": 1.6652146857498444,
      "grad_norm": 2.8368632793426514,
      "learning_rate": 2.506918819188192e-06,
      "loss": 1.3973,
      "step": 1338
    },
    {
      "epoch": 1.6664592408214065,
      "grad_norm": 2.9369115829467773,
      "learning_rate": 2.5046125461254618e-06,
      "loss": 1.4193,
      "step": 1339
    },
    {
      "epoch": 1.6677037958929684,
      "grad_norm": 2.870661973953247,
      "learning_rate": 2.5023062730627308e-06,
      "loss": 1.4212,
      "step": 1340
    },
    {
      "epoch": 1.6677037958929684,
      "eval_loss": 1.6332666873931885,
      "eval_runtime": 45.9731,
      "eval_samples_per_second": 21.752,
      "eval_steps_per_second": 0.914,
      "step": 1340
    },
    {
      "epoch": 1.6689483509645302,
      "grad_norm": 2.908878803253174,
      "learning_rate": 2.5e-06,
      "loss": 1.4083,
      "step": 1341
    },
    {
      "epoch": 1.670192906036092,
      "grad_norm": 2.833280563354492,
      "learning_rate": 2.4976937269372696e-06,
      "loss": 1.3911,
      "step": 1342
    },
    {
      "epoch": 1.671437461107654,
      "grad_norm": 2.7213969230651855,
      "learning_rate": 2.495387453874539e-06,
      "loss": 1.3997,
      "step": 1343
    },
    {
      "epoch": 1.672682016179216,
      "grad_norm": 2.9431257247924805,
      "learning_rate": 2.4930811808118085e-06,
      "loss": 1.4476,
      "step": 1344
    },
    {
      "epoch": 1.6739265712507778,
      "grad_norm": 2.8822503089904785,
      "learning_rate": 2.490774907749078e-06,
      "loss": 1.468,
      "step": 1345
    },
    {
      "epoch": 1.6751711263223399,
      "grad_norm": 2.836256504058838,
      "learning_rate": 2.488468634686347e-06,
      "loss": 1.4392,
      "step": 1346
    },
    {
      "epoch": 1.6764156813939017,
      "grad_norm": 2.8763248920440674,
      "learning_rate": 2.4861623616236163e-06,
      "loss": 1.4834,
      "step": 1347
    },
    {
      "epoch": 1.6776602364654636,
      "grad_norm": 2.884368658065796,
      "learning_rate": 2.4838560885608857e-06,
      "loss": 1.4323,
      "step": 1348
    },
    {
      "epoch": 1.6789047915370254,
      "grad_norm": 2.9247467517852783,
      "learning_rate": 2.481549815498155e-06,
      "loss": 1.4483,
      "step": 1349
    },
    {
      "epoch": 1.6801493466085873,
      "grad_norm": 3.007827043533325,
      "learning_rate": 2.4792435424354246e-06,
      "loss": 1.4802,
      "step": 1350
    },
    {
      "epoch": 1.6801493466085873,
      "eval_loss": 1.6364624500274658,
      "eval_runtime": 45.8512,
      "eval_samples_per_second": 21.81,
      "eval_steps_per_second": 0.916,
      "step": 1350
    },
    {
      "epoch": 1.6813939016801493,
      "grad_norm": 2.8283681869506836,
      "learning_rate": 2.476937269372694e-06,
      "loss": 1.3845,
      "step": 1351
    },
    {
      "epoch": 1.6826384567517114,
      "grad_norm": 3.0330536365509033,
      "learning_rate": 2.4746309963099634e-06,
      "loss": 1.3922,
      "step": 1352
    },
    {
      "epoch": 1.6838830118232733,
      "grad_norm": 3.1264121532440186,
      "learning_rate": 2.472324723247233e-06,
      "loss": 1.4498,
      "step": 1353
    },
    {
      "epoch": 1.6851275668948351,
      "grad_norm": 2.756582021713257,
      "learning_rate": 2.470018450184502e-06,
      "loss": 1.3814,
      "step": 1354
    },
    {
      "epoch": 1.686372121966397,
      "grad_norm": 2.8777999877929688,
      "learning_rate": 2.4677121771217713e-06,
      "loss": 1.3932,
      "step": 1355
    },
    {
      "epoch": 1.6876166770379588,
      "grad_norm": 2.9220948219299316,
      "learning_rate": 2.4654059040590407e-06,
      "loss": 1.3972,
      "step": 1356
    },
    {
      "epoch": 1.6888612321095209,
      "grad_norm": 2.818316698074341,
      "learning_rate": 2.46309963099631e-06,
      "loss": 1.3438,
      "step": 1357
    },
    {
      "epoch": 1.6901057871810827,
      "grad_norm": 2.9452595710754395,
      "learning_rate": 2.4607933579335795e-06,
      "loss": 1.4417,
      "step": 1358
    },
    {
      "epoch": 1.6913503422526448,
      "grad_norm": 3.1954009532928467,
      "learning_rate": 2.458487084870849e-06,
      "loss": 1.4563,
      "step": 1359
    },
    {
      "epoch": 1.6925948973242066,
      "grad_norm": 2.8596904277801514,
      "learning_rate": 2.4561808118081184e-06,
      "loss": 1.4252,
      "step": 1360
    },
    {
      "epoch": 1.6925948973242066,
      "eval_loss": 1.6394789218902588,
      "eval_runtime": 49.4542,
      "eval_samples_per_second": 20.221,
      "eval_steps_per_second": 0.849,
      "step": 1360
    },
    {
      "epoch": 1.6938394523957685,
      "grad_norm": 3.1115806102752686,
      "learning_rate": 2.453874538745388e-06,
      "loss": 1.4094,
      "step": 1361
    },
    {
      "epoch": 1.6950840074673303,
      "grad_norm": 3.2042553424835205,
      "learning_rate": 2.451568265682657e-06,
      "loss": 1.412,
      "step": 1362
    },
    {
      "epoch": 1.6963285625388922,
      "grad_norm": 3.1195313930511475,
      "learning_rate": 2.4492619926199267e-06,
      "loss": 1.4928,
      "step": 1363
    },
    {
      "epoch": 1.6975731176104543,
      "grad_norm": 2.992426633834839,
      "learning_rate": 2.4469557195571957e-06,
      "loss": 1.5128,
      "step": 1364
    },
    {
      "epoch": 1.6988176726820163,
      "grad_norm": 2.8993239402770996,
      "learning_rate": 2.444649446494465e-06,
      "loss": 1.4387,
      "step": 1365
    },
    {
      "epoch": 1.7000622277535782,
      "grad_norm": 2.923661708831787,
      "learning_rate": 2.4423431734317345e-06,
      "loss": 1.4179,
      "step": 1366
    },
    {
      "epoch": 1.70130678282514,
      "grad_norm": 2.822744131088257,
      "learning_rate": 2.440036900369004e-06,
      "loss": 1.4615,
      "step": 1367
    },
    {
      "epoch": 1.7025513378967019,
      "grad_norm": 2.769036293029785,
      "learning_rate": 2.437730627306273e-06,
      "loss": 1.4654,
      "step": 1368
    },
    {
      "epoch": 1.7037958929682637,
      "grad_norm": 2.798102855682373,
      "learning_rate": 2.4354243542435428e-06,
      "loss": 1.351,
      "step": 1369
    },
    {
      "epoch": 1.7050404480398258,
      "grad_norm": 2.8862292766571045,
      "learning_rate": 2.4331180811808118e-06,
      "loss": 1.4554,
      "step": 1370
    },
    {
      "epoch": 1.7050404480398258,
      "eval_loss": 1.6349775791168213,
      "eval_runtime": 46.8676,
      "eval_samples_per_second": 21.337,
      "eval_steps_per_second": 0.896,
      "step": 1370
    },
    {
      "epoch": 1.7062850031113876,
      "grad_norm": 2.9363768100738525,
      "learning_rate": 2.4308118081180816e-06,
      "loss": 1.4002,
      "step": 1371
    },
    {
      "epoch": 1.7075295581829497,
      "grad_norm": 2.893996000289917,
      "learning_rate": 2.4285055350553506e-06,
      "loss": 1.3996,
      "step": 1372
    },
    {
      "epoch": 1.7087741132545116,
      "grad_norm": 2.9740145206451416,
      "learning_rate": 2.42619926199262e-06,
      "loss": 1.4479,
      "step": 1373
    },
    {
      "epoch": 1.7100186683260734,
      "grad_norm": 2.7912323474884033,
      "learning_rate": 2.4238929889298895e-06,
      "loss": 1.4292,
      "step": 1374
    },
    {
      "epoch": 1.7112632233976353,
      "grad_norm": 2.780423164367676,
      "learning_rate": 2.421586715867159e-06,
      "loss": 1.3784,
      "step": 1375
    },
    {
      "epoch": 1.712507778469197,
      "grad_norm": 2.824495792388916,
      "learning_rate": 2.4192804428044283e-06,
      "loss": 1.4788,
      "step": 1376
    },
    {
      "epoch": 1.7137523335407592,
      "grad_norm": 2.8135807514190674,
      "learning_rate": 2.4169741697416977e-06,
      "loss": 1.3823,
      "step": 1377
    },
    {
      "epoch": 1.7149968886123212,
      "grad_norm": 2.7912282943725586,
      "learning_rate": 2.4146678966789667e-06,
      "loss": 1.3853,
      "step": 1378
    },
    {
      "epoch": 1.716241443683883,
      "grad_norm": 2.9185361862182617,
      "learning_rate": 2.4123616236162366e-06,
      "loss": 1.3905,
      "step": 1379
    },
    {
      "epoch": 1.717485998755445,
      "grad_norm": 2.9129741191864014,
      "learning_rate": 2.4100553505535056e-06,
      "loss": 1.369,
      "step": 1380
    },
    {
      "epoch": 1.717485998755445,
      "eval_loss": 1.6308677196502686,
      "eval_runtime": 46.1789,
      "eval_samples_per_second": 21.655,
      "eval_steps_per_second": 0.91,
      "step": 1380
    },
    {
      "epoch": 1.7187305538270068,
      "grad_norm": 3.103837013244629,
      "learning_rate": 2.407749077490775e-06,
      "loss": 1.4413,
      "step": 1381
    },
    {
      "epoch": 1.7199751088985686,
      "grad_norm": 3.030013084411621,
      "learning_rate": 2.4054428044280444e-06,
      "loss": 1.4523,
      "step": 1382
    },
    {
      "epoch": 1.7212196639701307,
      "grad_norm": 2.939122200012207,
      "learning_rate": 2.403136531365314e-06,
      "loss": 1.4349,
      "step": 1383
    },
    {
      "epoch": 1.7224642190416926,
      "grad_norm": 2.780301570892334,
      "learning_rate": 2.4008302583025833e-06,
      "loss": 1.4054,
      "step": 1384
    },
    {
      "epoch": 1.7237087741132546,
      "grad_norm": 2.8379433155059814,
      "learning_rate": 2.3985239852398527e-06,
      "loss": 1.4338,
      "step": 1385
    },
    {
      "epoch": 1.7249533291848165,
      "grad_norm": 2.77071475982666,
      "learning_rate": 2.3962177121771217e-06,
      "loss": 1.396,
      "step": 1386
    },
    {
      "epoch": 1.7261978842563783,
      "grad_norm": 2.815927267074585,
      "learning_rate": 2.3939114391143915e-06,
      "loss": 1.4436,
      "step": 1387
    },
    {
      "epoch": 1.7274424393279402,
      "grad_norm": 3.002138614654541,
      "learning_rate": 2.3916051660516605e-06,
      "loss": 1.4325,
      "step": 1388
    },
    {
      "epoch": 1.728686994399502,
      "grad_norm": 2.901615619659424,
      "learning_rate": 2.3892988929889304e-06,
      "loss": 1.3761,
      "step": 1389
    },
    {
      "epoch": 1.729931549471064,
      "grad_norm": 2.9884517192840576,
      "learning_rate": 2.3869926199261994e-06,
      "loss": 1.4541,
      "step": 1390
    },
    {
      "epoch": 1.729931549471064,
      "eval_loss": 1.634454369544983,
      "eval_runtime": 47.1589,
      "eval_samples_per_second": 21.205,
      "eval_steps_per_second": 0.891,
      "step": 1390
    },
    {
      "epoch": 1.7311761045426262,
      "grad_norm": 2.942561626434326,
      "learning_rate": 2.384686346863469e-06,
      "loss": 1.4087,
      "step": 1391
    },
    {
      "epoch": 1.732420659614188,
      "grad_norm": 3.0410168170928955,
      "learning_rate": 2.3823800738007382e-06,
      "loss": 1.4453,
      "step": 1392
    },
    {
      "epoch": 1.7336652146857499,
      "grad_norm": 2.8488197326660156,
      "learning_rate": 2.3800738007380077e-06,
      "loss": 1.4642,
      "step": 1393
    },
    {
      "epoch": 1.7349097697573117,
      "grad_norm": 2.8784000873565674,
      "learning_rate": 2.377767527675277e-06,
      "loss": 1.4334,
      "step": 1394
    },
    {
      "epoch": 1.7361543248288736,
      "grad_norm": 2.818960428237915,
      "learning_rate": 2.3754612546125465e-06,
      "loss": 1.4579,
      "step": 1395
    },
    {
      "epoch": 1.7373988799004356,
      "grad_norm": 2.9459662437438965,
      "learning_rate": 2.3731549815498155e-06,
      "loss": 1.4524,
      "step": 1396
    },
    {
      "epoch": 1.7386434349719975,
      "grad_norm": 2.829437732696533,
      "learning_rate": 2.3708487084870853e-06,
      "loss": 1.4765,
      "step": 1397
    },
    {
      "epoch": 1.7398879900435595,
      "grad_norm": 2.8156938552856445,
      "learning_rate": 2.3685424354243543e-06,
      "loss": 1.3722,
      "step": 1398
    },
    {
      "epoch": 1.7411325451151214,
      "grad_norm": 2.8707408905029297,
      "learning_rate": 2.3662361623616238e-06,
      "loss": 1.4209,
      "step": 1399
    },
    {
      "epoch": 1.7423771001866832,
      "grad_norm": 3.141056776046753,
      "learning_rate": 2.363929889298893e-06,
      "loss": 1.4622,
      "step": 1400
    },
    {
      "epoch": 1.7423771001866832,
      "eval_loss": 1.6302117109298706,
      "eval_runtime": 45.7354,
      "eval_samples_per_second": 21.865,
      "eval_steps_per_second": 0.918,
      "step": 1400
    },
    {
      "epoch": 1.743621655258245,
      "grad_norm": 2.8620946407318115,
      "learning_rate": 2.3616236162361626e-06,
      "loss": 1.4243,
      "step": 1401
    },
    {
      "epoch": 1.744866210329807,
      "grad_norm": 2.927598237991333,
      "learning_rate": 2.359317343173432e-06,
      "loss": 1.441,
      "step": 1402
    },
    {
      "epoch": 1.746110765401369,
      "grad_norm": 3.0190820693969727,
      "learning_rate": 2.3570110701107015e-06,
      "loss": 1.407,
      "step": 1403
    },
    {
      "epoch": 1.747355320472931,
      "grad_norm": 2.891575574874878,
      "learning_rate": 2.3547047970479705e-06,
      "loss": 1.4215,
      "step": 1404
    },
    {
      "epoch": 1.748599875544493,
      "grad_norm": 2.9068562984466553,
      "learning_rate": 2.3523985239852403e-06,
      "loss": 1.4028,
      "step": 1405
    },
    {
      "epoch": 1.7498444306160548,
      "grad_norm": 2.8855676651000977,
      "learning_rate": 2.3500922509225093e-06,
      "loss": 1.4561,
      "step": 1406
    },
    {
      "epoch": 1.7510889856876166,
      "grad_norm": 2.9122142791748047,
      "learning_rate": 2.3477859778597787e-06,
      "loss": 1.3914,
      "step": 1407
    },
    {
      "epoch": 1.7523335407591785,
      "grad_norm": 2.9551825523376465,
      "learning_rate": 2.345479704797048e-06,
      "loss": 1.4017,
      "step": 1408
    },
    {
      "epoch": 1.7535780958307405,
      "grad_norm": 2.846301555633545,
      "learning_rate": 2.3431734317343176e-06,
      "loss": 1.3984,
      "step": 1409
    },
    {
      "epoch": 1.7548226509023024,
      "grad_norm": 3.2388343811035156,
      "learning_rate": 2.340867158671587e-06,
      "loss": 1.4908,
      "step": 1410
    },
    {
      "epoch": 1.7548226509023024,
      "eval_loss": 1.6310688257217407,
      "eval_runtime": 43.6166,
      "eval_samples_per_second": 22.927,
      "eval_steps_per_second": 0.963,
      "step": 1410
    },
    {
      "epoch": 1.7560672059738645,
      "grad_norm": 2.8560962677001953,
      "learning_rate": 2.3385608856088564e-06,
      "loss": 1.4064,
      "step": 1411
    },
    {
      "epoch": 1.7573117610454263,
      "grad_norm": 2.8909051418304443,
      "learning_rate": 2.336254612546126e-06,
      "loss": 1.4395,
      "step": 1412
    },
    {
      "epoch": 1.7585563161169881,
      "grad_norm": 2.973435401916504,
      "learning_rate": 2.333948339483395e-06,
      "loss": 1.424,
      "step": 1413
    },
    {
      "epoch": 1.75980087118855,
      "grad_norm": 3.0337705612182617,
      "learning_rate": 2.3316420664206643e-06,
      "loss": 1.4188,
      "step": 1414
    },
    {
      "epoch": 1.7610454262601118,
      "grad_norm": 2.7573161125183105,
      "learning_rate": 2.3293357933579337e-06,
      "loss": 1.3815,
      "step": 1415
    },
    {
      "epoch": 1.762289981331674,
      "grad_norm": 2.7978484630584717,
      "learning_rate": 2.327029520295203e-06,
      "loss": 1.4007,
      "step": 1416
    },
    {
      "epoch": 1.763534536403236,
      "grad_norm": 2.9547295570373535,
      "learning_rate": 2.3247232472324725e-06,
      "loss": 1.3924,
      "step": 1417
    },
    {
      "epoch": 1.7647790914747978,
      "grad_norm": 2.941542863845825,
      "learning_rate": 2.322416974169742e-06,
      "loss": 1.4434,
      "step": 1418
    },
    {
      "epoch": 1.7660236465463597,
      "grad_norm": 2.855360984802246,
      "learning_rate": 2.3201107011070114e-06,
      "loss": 1.3852,
      "step": 1419
    },
    {
      "epoch": 1.7672682016179215,
      "grad_norm": 2.87410044670105,
      "learning_rate": 2.317804428044281e-06,
      "loss": 1.4014,
      "step": 1420
    },
    {
      "epoch": 1.7672682016179215,
      "eval_loss": 1.6320465803146362,
      "eval_runtime": 47.5665,
      "eval_samples_per_second": 21.023,
      "eval_steps_per_second": 0.883,
      "step": 1420
    },
    {
      "epoch": 1.7685127566894834,
      "grad_norm": 2.84712290763855,
      "learning_rate": 2.31549815498155e-06,
      "loss": 1.4049,
      "step": 1421
    },
    {
      "epoch": 1.7697573117610454,
      "grad_norm": 2.924795627593994,
      "learning_rate": 2.3131918819188192e-06,
      "loss": 1.3823,
      "step": 1422
    },
    {
      "epoch": 1.7710018668326073,
      "grad_norm": 2.9930474758148193,
      "learning_rate": 2.3108856088560886e-06,
      "loss": 1.4226,
      "step": 1423
    },
    {
      "epoch": 1.7722464219041694,
      "grad_norm": 2.975088596343994,
      "learning_rate": 2.308579335793358e-06,
      "loss": 1.448,
      "step": 1424
    },
    {
      "epoch": 1.7734909769757312,
      "grad_norm": 2.958648443222046,
      "learning_rate": 2.3062730627306275e-06,
      "loss": 1.421,
      "step": 1425
    },
    {
      "epoch": 1.774735532047293,
      "grad_norm": 3.057056427001953,
      "learning_rate": 2.303966789667897e-06,
      "loss": 1.4383,
      "step": 1426
    },
    {
      "epoch": 1.775980087118855,
      "grad_norm": 2.8630926609039307,
      "learning_rate": 2.3016605166051663e-06,
      "loss": 1.3638,
      "step": 1427
    },
    {
      "epoch": 1.777224642190417,
      "grad_norm": 2.9939122200012207,
      "learning_rate": 2.2993542435424358e-06,
      "loss": 1.363,
      "step": 1428
    },
    {
      "epoch": 1.7784691972619788,
      "grad_norm": 2.8689146041870117,
      "learning_rate": 2.2970479704797048e-06,
      "loss": 1.4288,
      "step": 1429
    },
    {
      "epoch": 1.779713752333541,
      "grad_norm": 3.0293314456939697,
      "learning_rate": 2.294741697416974e-06,
      "loss": 1.5169,
      "step": 1430
    },
    {
      "epoch": 1.779713752333541,
      "eval_loss": 1.6306153535842896,
      "eval_runtime": 46.5436,
      "eval_samples_per_second": 21.485,
      "eval_steps_per_second": 0.902,
      "step": 1430
    },
    {
      "epoch": 1.7809583074051027,
      "grad_norm": 2.9598498344421387,
      "learning_rate": 2.2924354243542436e-06,
      "loss": 1.4094,
      "step": 1431
    },
    {
      "epoch": 1.7822028624766646,
      "grad_norm": 2.7908709049224854,
      "learning_rate": 2.290129151291513e-06,
      "loss": 1.4045,
      "step": 1432
    },
    {
      "epoch": 1.7834474175482264,
      "grad_norm": 2.9249887466430664,
      "learning_rate": 2.2878228782287825e-06,
      "loss": 1.4238,
      "step": 1433
    },
    {
      "epoch": 1.7846919726197883,
      "grad_norm": 2.9420599937438965,
      "learning_rate": 2.285516605166052e-06,
      "loss": 1.3984,
      "step": 1434
    },
    {
      "epoch": 1.7859365276913504,
      "grad_norm": 2.89164137840271,
      "learning_rate": 2.2832103321033213e-06,
      "loss": 1.4121,
      "step": 1435
    },
    {
      "epoch": 1.7871810827629122,
      "grad_norm": 3.031681537628174,
      "learning_rate": 2.2809040590405907e-06,
      "loss": 1.4333,
      "step": 1436
    },
    {
      "epoch": 1.7884256378344743,
      "grad_norm": 2.9249119758605957,
      "learning_rate": 2.2785977859778597e-06,
      "loss": 1.4223,
      "step": 1437
    },
    {
      "epoch": 1.7896701929060361,
      "grad_norm": 2.9735820293426514,
      "learning_rate": 2.2762915129151296e-06,
      "loss": 1.4092,
      "step": 1438
    },
    {
      "epoch": 1.790914747977598,
      "grad_norm": 2.992908000946045,
      "learning_rate": 2.2739852398523986e-06,
      "loss": 1.4088,
      "step": 1439
    },
    {
      "epoch": 1.7921593030491598,
      "grad_norm": 2.9289679527282715,
      "learning_rate": 2.271678966789668e-06,
      "loss": 1.455,
      "step": 1440
    },
    {
      "epoch": 1.7921593030491598,
      "eval_loss": 1.6259137392044067,
      "eval_runtime": 48.3913,
      "eval_samples_per_second": 20.665,
      "eval_steps_per_second": 0.868,
      "step": 1440
    },
    {
      "epoch": 1.793403858120722,
      "grad_norm": 2.9257709980010986,
      "learning_rate": 2.2693726937269374e-06,
      "loss": 1.3674,
      "step": 1441
    },
    {
      "epoch": 1.7946484131922837,
      "grad_norm": 2.86993408203125,
      "learning_rate": 2.267066420664207e-06,
      "loss": 1.4333,
      "step": 1442
    },
    {
      "epoch": 1.7958929682638458,
      "grad_norm": 2.9061667919158936,
      "learning_rate": 2.2647601476014763e-06,
      "loss": 1.439,
      "step": 1443
    },
    {
      "epoch": 1.7971375233354077,
      "grad_norm": 2.8412883281707764,
      "learning_rate": 2.2624538745387457e-06,
      "loss": 1.3728,
      "step": 1444
    },
    {
      "epoch": 1.7983820784069695,
      "grad_norm": 3.0139172077178955,
      "learning_rate": 2.2601476014760147e-06,
      "loss": 1.4635,
      "step": 1445
    },
    {
      "epoch": 1.7996266334785314,
      "grad_norm": 3.016864538192749,
      "learning_rate": 2.2578413284132845e-06,
      "loss": 1.4729,
      "step": 1446
    },
    {
      "epoch": 1.8008711885500932,
      "grad_norm": 2.9747543334960938,
      "learning_rate": 2.2555350553505535e-06,
      "loss": 1.4484,
      "step": 1447
    },
    {
      "epoch": 1.8021157436216553,
      "grad_norm": 3.039520740509033,
      "learning_rate": 2.253228782287823e-06,
      "loss": 1.4188,
      "step": 1448
    },
    {
      "epoch": 1.8033602986932171,
      "grad_norm": 3.0893354415893555,
      "learning_rate": 2.2509225092250924e-06,
      "loss": 1.4346,
      "step": 1449
    },
    {
      "epoch": 1.8046048537647792,
      "grad_norm": 2.9363458156585693,
      "learning_rate": 2.248616236162362e-06,
      "loss": 1.4305,
      "step": 1450
    },
    {
      "epoch": 1.8046048537647792,
      "eval_loss": 1.6296759843826294,
      "eval_runtime": 49.8814,
      "eval_samples_per_second": 20.048,
      "eval_steps_per_second": 0.842,
      "step": 1450
    },
    {
      "epoch": 1.805849408836341,
      "grad_norm": 2.941295623779297,
      "learning_rate": 2.2463099630996312e-06,
      "loss": 1.3816,
      "step": 1451
    },
    {
      "epoch": 1.807093963907903,
      "grad_norm": 3.3323709964752197,
      "learning_rate": 2.2440036900369006e-06,
      "loss": 1.4556,
      "step": 1452
    },
    {
      "epoch": 1.8083385189794647,
      "grad_norm": 2.942725658416748,
      "learning_rate": 2.2416974169741696e-06,
      "loss": 1.4102,
      "step": 1453
    },
    {
      "epoch": 1.8095830740510268,
      "grad_norm": 3.136852979660034,
      "learning_rate": 2.2393911439114395e-06,
      "loss": 1.4187,
      "step": 1454
    },
    {
      "epoch": 1.8108276291225887,
      "grad_norm": 2.959941864013672,
      "learning_rate": 2.2370848708487085e-06,
      "loss": 1.4008,
      "step": 1455
    },
    {
      "epoch": 1.8120721841941507,
      "grad_norm": 3.136758804321289,
      "learning_rate": 2.2347785977859783e-06,
      "loss": 1.4283,
      "step": 1456
    },
    {
      "epoch": 1.8133167392657126,
      "grad_norm": 2.973801612854004,
      "learning_rate": 2.2324723247232473e-06,
      "loss": 1.4911,
      "step": 1457
    },
    {
      "epoch": 1.8145612943372744,
      "grad_norm": 2.96920108795166,
      "learning_rate": 2.2301660516605168e-06,
      "loss": 1.4012,
      "step": 1458
    },
    {
      "epoch": 1.8158058494088363,
      "grad_norm": 3.128460645675659,
      "learning_rate": 2.227859778597786e-06,
      "loss": 1.4134,
      "step": 1459
    },
    {
      "epoch": 1.8170504044803981,
      "grad_norm": 2.8887293338775635,
      "learning_rate": 2.2255535055350556e-06,
      "loss": 1.3961,
      "step": 1460
    },
    {
      "epoch": 1.8170504044803981,
      "eval_loss": 1.6299411058425903,
      "eval_runtime": 47.4462,
      "eval_samples_per_second": 21.077,
      "eval_steps_per_second": 0.885,
      "step": 1460
    },
    {
      "epoch": 1.8182949595519602,
      "grad_norm": 3.1253349781036377,
      "learning_rate": 2.2232472324723246e-06,
      "loss": 1.462,
      "step": 1461
    },
    {
      "epoch": 1.819539514623522,
      "grad_norm": 3.0835256576538086,
      "learning_rate": 2.2209409594095944e-06,
      "loss": 1.4127,
      "step": 1462
    },
    {
      "epoch": 1.820784069695084,
      "grad_norm": 3.2318263053894043,
      "learning_rate": 2.2186346863468634e-06,
      "loss": 1.3559,
      "step": 1463
    },
    {
      "epoch": 1.822028624766646,
      "grad_norm": 3.0926361083984375,
      "learning_rate": 2.2163284132841333e-06,
      "loss": 1.3967,
      "step": 1464
    },
    {
      "epoch": 1.8232731798382078,
      "grad_norm": 2.927354097366333,
      "learning_rate": 2.2140221402214023e-06,
      "loss": 1.4169,
      "step": 1465
    },
    {
      "epoch": 1.8245177349097697,
      "grad_norm": 3.00901460647583,
      "learning_rate": 2.2117158671586717e-06,
      "loss": 1.4099,
      "step": 1466
    },
    {
      "epoch": 1.8257622899813317,
      "grad_norm": 3.255394458770752,
      "learning_rate": 2.209409594095941e-06,
      "loss": 1.4229,
      "step": 1467
    },
    {
      "epoch": 1.8270068450528936,
      "grad_norm": 3.0004467964172363,
      "learning_rate": 2.2071033210332106e-06,
      "loss": 1.4089,
      "step": 1468
    },
    {
      "epoch": 1.8282514001244556,
      "grad_norm": 2.873079299926758,
      "learning_rate": 2.20479704797048e-06,
      "loss": 1.4043,
      "step": 1469
    },
    {
      "epoch": 1.8294959551960175,
      "grad_norm": 2.924314498901367,
      "learning_rate": 2.2024907749077494e-06,
      "loss": 1.389,
      "step": 1470
    },
    {
      "epoch": 1.8294959551960175,
      "eval_loss": 1.6256897449493408,
      "eval_runtime": 45.593,
      "eval_samples_per_second": 21.933,
      "eval_steps_per_second": 0.921,
      "step": 1470
    },
    {
      "epoch": 1.8307405102675793,
      "grad_norm": 2.975888252258301,
      "learning_rate": 2.2001845018450184e-06,
      "loss": 1.3983,
      "step": 1471
    },
    {
      "epoch": 1.8319850653391412,
      "grad_norm": 2.9005095958709717,
      "learning_rate": 2.1978782287822883e-06,
      "loss": 1.4247,
      "step": 1472
    },
    {
      "epoch": 1.833229620410703,
      "grad_norm": 2.952568292617798,
      "learning_rate": 2.1955719557195573e-06,
      "loss": 1.4195,
      "step": 1473
    },
    {
      "epoch": 1.834474175482265,
      "grad_norm": 3.21946120262146,
      "learning_rate": 2.1932656826568267e-06,
      "loss": 1.4282,
      "step": 1474
    },
    {
      "epoch": 1.8357187305538272,
      "grad_norm": 2.847078800201416,
      "learning_rate": 2.190959409594096e-06,
      "loss": 1.3733,
      "step": 1475
    },
    {
      "epoch": 1.836963285625389,
      "grad_norm": 2.9820823669433594,
      "learning_rate": 2.1886531365313655e-06,
      "loss": 1.4494,
      "step": 1476
    },
    {
      "epoch": 1.8382078406969509,
      "grad_norm": 2.9088616371154785,
      "learning_rate": 2.186346863468635e-06,
      "loss": 1.3263,
      "step": 1477
    },
    {
      "epoch": 1.8394523957685127,
      "grad_norm": 2.9320101737976074,
      "learning_rate": 2.1840405904059044e-06,
      "loss": 1.3931,
      "step": 1478
    },
    {
      "epoch": 1.8406969508400746,
      "grad_norm": 2.8825459480285645,
      "learning_rate": 2.1817343173431734e-06,
      "loss": 1.4188,
      "step": 1479
    },
    {
      "epoch": 1.8419415059116366,
      "grad_norm": 2.9527266025543213,
      "learning_rate": 2.1794280442804432e-06,
      "loss": 1.4245,
      "step": 1480
    },
    {
      "epoch": 1.8419415059116366,
      "eval_loss": 1.623984456062317,
      "eval_runtime": 45.3346,
      "eval_samples_per_second": 22.058,
      "eval_steps_per_second": 0.926,
      "step": 1480
    },
    {
      "epoch": 1.8431860609831985,
      "grad_norm": 2.951578140258789,
      "learning_rate": 2.177121771217712e-06,
      "loss": 1.3723,
      "step": 1481
    },
    {
      "epoch": 1.8444306160547606,
      "grad_norm": 2.841216564178467,
      "learning_rate": 2.1748154981549816e-06,
      "loss": 1.3893,
      "step": 1482
    },
    {
      "epoch": 1.8456751711263224,
      "grad_norm": 2.933821201324463,
      "learning_rate": 2.172509225092251e-06,
      "loss": 1.4196,
      "step": 1483
    },
    {
      "epoch": 1.8469197261978842,
      "grad_norm": 3.2382006645202637,
      "learning_rate": 2.1702029520295205e-06,
      "loss": 1.3512,
      "step": 1484
    },
    {
      "epoch": 1.848164281269446,
      "grad_norm": 2.972015380859375,
      "learning_rate": 2.16789667896679e-06,
      "loss": 1.3762,
      "step": 1485
    },
    {
      "epoch": 1.849408836341008,
      "grad_norm": 2.8245530128479004,
      "learning_rate": 2.1655904059040593e-06,
      "loss": 1.3758,
      "step": 1486
    },
    {
      "epoch": 1.85065339141257,
      "grad_norm": 3.108802556991577,
      "learning_rate": 2.1632841328413287e-06,
      "loss": 1.4454,
      "step": 1487
    },
    {
      "epoch": 1.851897946484132,
      "grad_norm": 3.196927547454834,
      "learning_rate": 2.160977859778598e-06,
      "loss": 1.4548,
      "step": 1488
    },
    {
      "epoch": 1.853142501555694,
      "grad_norm": 2.9764368534088135,
      "learning_rate": 2.158671586715867e-06,
      "loss": 1.4335,
      "step": 1489
    },
    {
      "epoch": 1.8543870566272558,
      "grad_norm": 2.8558847904205322,
      "learning_rate": 2.1563653136531366e-06,
      "loss": 1.4468,
      "step": 1490
    },
    {
      "epoch": 1.8543870566272558,
      "eval_loss": 1.6222788095474243,
      "eval_runtime": 50.0333,
      "eval_samples_per_second": 19.987,
      "eval_steps_per_second": 0.839,
      "step": 1490
    },
    {
      "epoch": 1.8556316116988176,
      "grad_norm": 2.875429153442383,
      "learning_rate": 2.154059040590406e-06,
      "loss": 1.4258,
      "step": 1491
    },
    {
      "epoch": 1.8568761667703795,
      "grad_norm": 2.993828773498535,
      "learning_rate": 2.1517527675276754e-06,
      "loss": 1.3982,
      "step": 1492
    },
    {
      "epoch": 1.8581207218419415,
      "grad_norm": 2.9214375019073486,
      "learning_rate": 2.149446494464945e-06,
      "loss": 1.4978,
      "step": 1493
    },
    {
      "epoch": 1.8593652769135034,
      "grad_norm": 2.9349868297576904,
      "learning_rate": 2.1471402214022143e-06,
      "loss": 1.4385,
      "step": 1494
    },
    {
      "epoch": 1.8606098319850655,
      "grad_norm": 2.926352024078369,
      "learning_rate": 2.1448339483394837e-06,
      "loss": 1.3913,
      "step": 1495
    },
    {
      "epoch": 1.8618543870566273,
      "grad_norm": 2.8497722148895264,
      "learning_rate": 2.142527675276753e-06,
      "loss": 1.3915,
      "step": 1496
    },
    {
      "epoch": 1.8630989421281892,
      "grad_norm": 2.8016490936279297,
      "learning_rate": 2.140221402214022e-06,
      "loss": 1.3867,
      "step": 1497
    },
    {
      "epoch": 1.864343497199751,
      "grad_norm": 2.919970989227295,
      "learning_rate": 2.1379151291512916e-06,
      "loss": 1.3784,
      "step": 1498
    },
    {
      "epoch": 1.8655880522713129,
      "grad_norm": 3.095557689666748,
      "learning_rate": 2.135608856088561e-06,
      "loss": 1.4297,
      "step": 1499
    },
    {
      "epoch": 1.866832607342875,
      "grad_norm": 3.001741647720337,
      "learning_rate": 2.1333025830258304e-06,
      "loss": 1.4259,
      "step": 1500
    },
    {
      "epoch": 1.866832607342875,
      "eval_loss": 1.6227651834487915,
      "eval_runtime": 48.9527,
      "eval_samples_per_second": 20.428,
      "eval_steps_per_second": 0.858,
      "step": 1500
    },
    {
      "epoch": 1.868077162414437,
      "grad_norm": 2.906398057937622,
      "learning_rate": 2.1309963099631e-06,
      "loss": 1.4209,
      "step": 1501
    },
    {
      "epoch": 1.8693217174859988,
      "grad_norm": 3.006981134414673,
      "learning_rate": 2.1286900369003692e-06,
      "loss": 1.4008,
      "step": 1502
    },
    {
      "epoch": 1.8705662725575607,
      "grad_norm": 2.848041534423828,
      "learning_rate": 2.1263837638376387e-06,
      "loss": 1.4402,
      "step": 1503
    },
    {
      "epoch": 1.8718108276291225,
      "grad_norm": 2.9804189205169678,
      "learning_rate": 2.124077490774908e-06,
      "loss": 1.3883,
      "step": 1504
    },
    {
      "epoch": 1.8730553827006844,
      "grad_norm": 2.9567763805389404,
      "learning_rate": 2.1217712177121775e-06,
      "loss": 1.452,
      "step": 1505
    },
    {
      "epoch": 1.8742999377722465,
      "grad_norm": 2.842329978942871,
      "learning_rate": 2.1194649446494465e-06,
      "loss": 1.3862,
      "step": 1506
    },
    {
      "epoch": 1.8755444928438083,
      "grad_norm": 2.8646347522735596,
      "learning_rate": 2.117158671586716e-06,
      "loss": 1.4487,
      "step": 1507
    },
    {
      "epoch": 1.8767890479153704,
      "grad_norm": 2.9858622550964355,
      "learning_rate": 2.1148523985239854e-06,
      "loss": 1.4656,
      "step": 1508
    },
    {
      "epoch": 1.8780336029869322,
      "grad_norm": 3.1869702339172363,
      "learning_rate": 2.1125461254612548e-06,
      "loss": 1.4804,
      "step": 1509
    },
    {
      "epoch": 1.879278158058494,
      "grad_norm": 2.987708806991577,
      "learning_rate": 2.110239852398524e-06,
      "loss": 1.4597,
      "step": 1510
    },
    {
      "epoch": 1.879278158058494,
      "eval_loss": 1.6230274438858032,
      "eval_runtime": 43.1679,
      "eval_samples_per_second": 23.165,
      "eval_steps_per_second": 0.973,
      "step": 1510
    },
    {
      "epoch": 1.880522713130056,
      "grad_norm": 2.9133615493774414,
      "learning_rate": 2.1079335793357936e-06,
      "loss": 1.438,
      "step": 1511
    },
    {
      "epoch": 1.8817672682016178,
      "grad_norm": 2.992630958557129,
      "learning_rate": 2.1056273062730626e-06,
      "loss": 1.3445,
      "step": 1512
    },
    {
      "epoch": 1.8830118232731798,
      "grad_norm": 2.8437397480010986,
      "learning_rate": 2.1033210332103325e-06,
      "loss": 1.4519,
      "step": 1513
    },
    {
      "epoch": 1.884256378344742,
      "grad_norm": 2.7574918270111084,
      "learning_rate": 2.1010147601476015e-06,
      "loss": 1.398,
      "step": 1514
    },
    {
      "epoch": 1.8855009334163038,
      "grad_norm": 2.9313275814056396,
      "learning_rate": 2.098708487084871e-06,
      "loss": 1.4079,
      "step": 1515
    },
    {
      "epoch": 1.8867454884878656,
      "grad_norm": 2.9336822032928467,
      "learning_rate": 2.0964022140221403e-06,
      "loss": 1.4588,
      "step": 1516
    },
    {
      "epoch": 1.8879900435594275,
      "grad_norm": 2.7718148231506348,
      "learning_rate": 2.0940959409594097e-06,
      "loss": 1.398,
      "step": 1517
    },
    {
      "epoch": 1.8892345986309893,
      "grad_norm": 2.9897115230560303,
      "learning_rate": 2.091789667896679e-06,
      "loss": 1.4558,
      "step": 1518
    },
    {
      "epoch": 1.8904791537025514,
      "grad_norm": 2.8779306411743164,
      "learning_rate": 2.0894833948339486e-06,
      "loss": 1.3617,
      "step": 1519
    },
    {
      "epoch": 1.8917237087741132,
      "grad_norm": 3.035209894180298,
      "learning_rate": 2.0871771217712176e-06,
      "loss": 1.4149,
      "step": 1520
    },
    {
      "epoch": 1.8917237087741132,
      "eval_loss": 1.6238139867782593,
      "eval_runtime": 47.5895,
      "eval_samples_per_second": 21.013,
      "eval_steps_per_second": 0.883,
      "step": 1520
    },
    {
      "epoch": 1.8929682638456753,
      "grad_norm": 2.9098663330078125,
      "learning_rate": 2.0848708487084874e-06,
      "loss": 1.4138,
      "step": 1521
    },
    {
      "epoch": 1.8942128189172371,
      "grad_norm": 2.8637537956237793,
      "learning_rate": 2.0825645756457564e-06,
      "loss": 1.3726,
      "step": 1522
    },
    {
      "epoch": 1.895457373988799,
      "grad_norm": 2.981131076812744,
      "learning_rate": 2.0802583025830263e-06,
      "loss": 1.4156,
      "step": 1523
    },
    {
      "epoch": 1.8967019290603608,
      "grad_norm": 2.9190938472747803,
      "learning_rate": 2.0779520295202953e-06,
      "loss": 1.3988,
      "step": 1524
    },
    {
      "epoch": 1.8979464841319227,
      "grad_norm": 2.937696695327759,
      "learning_rate": 2.0756457564575647e-06,
      "loss": 1.3746,
      "step": 1525
    },
    {
      "epoch": 1.8991910392034848,
      "grad_norm": 3.0422637462615967,
      "learning_rate": 2.073339483394834e-06,
      "loss": 1.4064,
      "step": 1526
    },
    {
      "epoch": 1.9004355942750468,
      "grad_norm": 2.920274496078491,
      "learning_rate": 2.0710332103321035e-06,
      "loss": 1.416,
      "step": 1527
    },
    {
      "epoch": 1.9016801493466087,
      "grad_norm": 2.917694091796875,
      "learning_rate": 2.0687269372693725e-06,
      "loss": 1.455,
      "step": 1528
    },
    {
      "epoch": 1.9029247044181705,
      "grad_norm": 2.961634635925293,
      "learning_rate": 2.0664206642066424e-06,
      "loss": 1.3617,
      "step": 1529
    },
    {
      "epoch": 1.9041692594897324,
      "grad_norm": 3.014674186706543,
      "learning_rate": 2.0641143911439114e-06,
      "loss": 1.4464,
      "step": 1530
    },
    {
      "epoch": 1.9041692594897324,
      "eval_loss": 1.6272473335266113,
      "eval_runtime": 49.0723,
      "eval_samples_per_second": 20.378,
      "eval_steps_per_second": 0.856,
      "step": 1530
    },
    {
      "epoch": 1.9054138145612942,
      "grad_norm": 3.0149717330932617,
      "learning_rate": 2.0618081180811812e-06,
      "loss": 1.4478,
      "step": 1531
    },
    {
      "epoch": 1.9066583696328563,
      "grad_norm": 2.9122891426086426,
      "learning_rate": 2.0595018450184502e-06,
      "loss": 1.4161,
      "step": 1532
    },
    {
      "epoch": 1.9079029247044181,
      "grad_norm": 2.873610496520996,
      "learning_rate": 2.0571955719557197e-06,
      "loss": 1.4102,
      "step": 1533
    },
    {
      "epoch": 1.9091474797759802,
      "grad_norm": 3.173130512237549,
      "learning_rate": 2.054889298892989e-06,
      "loss": 1.4761,
      "step": 1534
    },
    {
      "epoch": 1.910392034847542,
      "grad_norm": 3.0038845539093018,
      "learning_rate": 2.0525830258302585e-06,
      "loss": 1.411,
      "step": 1535
    },
    {
      "epoch": 1.911636589919104,
      "grad_norm": 3.0535807609558105,
      "learning_rate": 2.050276752767528e-06,
      "loss": 1.3858,
      "step": 1536
    },
    {
      "epoch": 1.9128811449906657,
      "grad_norm": 2.9695370197296143,
      "learning_rate": 2.0479704797047974e-06,
      "loss": 1.4343,
      "step": 1537
    },
    {
      "epoch": 1.9141257000622276,
      "grad_norm": 2.9697799682617188,
      "learning_rate": 2.0456642066420664e-06,
      "loss": 1.4218,
      "step": 1538
    },
    {
      "epoch": 1.9153702551337897,
      "grad_norm": 2.903224468231201,
      "learning_rate": 2.043357933579336e-06,
      "loss": 1.4614,
      "step": 1539
    },
    {
      "epoch": 1.9166148102053517,
      "grad_norm": 2.789318084716797,
      "learning_rate": 2.041051660516605e-06,
      "loss": 1.3156,
      "step": 1540
    },
    {
      "epoch": 1.9166148102053517,
      "eval_loss": 1.6211179494857788,
      "eval_runtime": 47.0639,
      "eval_samples_per_second": 21.248,
      "eval_steps_per_second": 0.892,
      "step": 1540
    },
    {
      "epoch": 1.9178593652769136,
      "grad_norm": 3.066250801086426,
      "learning_rate": 2.0387453874538746e-06,
      "loss": 1.4439,
      "step": 1541
    },
    {
      "epoch": 1.9191039203484754,
      "grad_norm": 2.8898353576660156,
      "learning_rate": 2.036439114391144e-06,
      "loss": 1.3716,
      "step": 1542
    },
    {
      "epoch": 1.9203484754200373,
      "grad_norm": 2.975554943084717,
      "learning_rate": 2.0341328413284135e-06,
      "loss": 1.4008,
      "step": 1543
    },
    {
      "epoch": 1.9215930304915991,
      "grad_norm": 3.0215322971343994,
      "learning_rate": 2.031826568265683e-06,
      "loss": 1.3787,
      "step": 1544
    },
    {
      "epoch": 1.9228375855631612,
      "grad_norm": 3.074124574661255,
      "learning_rate": 2.0295202952029523e-06,
      "loss": 1.4528,
      "step": 1545
    },
    {
      "epoch": 1.924082140634723,
      "grad_norm": 2.8828608989715576,
      "learning_rate": 2.0272140221402213e-06,
      "loss": 1.3934,
      "step": 1546
    },
    {
      "epoch": 1.9253266957062851,
      "grad_norm": 2.8467960357666016,
      "learning_rate": 2.024907749077491e-06,
      "loss": 1.414,
      "step": 1547
    },
    {
      "epoch": 1.926571250777847,
      "grad_norm": 2.9690022468566895,
      "learning_rate": 2.02260147601476e-06,
      "loss": 1.4024,
      "step": 1548
    },
    {
      "epoch": 1.9278158058494088,
      "grad_norm": 2.913491725921631,
      "learning_rate": 2.02029520295203e-06,
      "loss": 1.4166,
      "step": 1549
    },
    {
      "epoch": 1.9290603609209707,
      "grad_norm": 2.9665236473083496,
      "learning_rate": 2.017988929889299e-06,
      "loss": 1.3949,
      "step": 1550
    },
    {
      "epoch": 1.9290603609209707,
      "eval_loss": 1.6198350191116333,
      "eval_runtime": 48.1578,
      "eval_samples_per_second": 20.765,
      "eval_steps_per_second": 0.872,
      "step": 1550
    },
    {
      "epoch": 1.9303049159925325,
      "grad_norm": 3.0438289642333984,
      "learning_rate": 2.0156826568265684e-06,
      "loss": 1.4364,
      "step": 1551
    },
    {
      "epoch": 1.9315494710640946,
      "grad_norm": 2.9459996223449707,
      "learning_rate": 2.013376383763838e-06,
      "loss": 1.4102,
      "step": 1552
    },
    {
      "epoch": 1.9327940261356567,
      "grad_norm": 2.898162364959717,
      "learning_rate": 2.0110701107011073e-06,
      "loss": 1.3856,
      "step": 1553
    },
    {
      "epoch": 1.9340385812072185,
      "grad_norm": 3.0183422565460205,
      "learning_rate": 2.0087638376383767e-06,
      "loss": 1.4351,
      "step": 1554
    },
    {
      "epoch": 1.9352831362787803,
      "grad_norm": 3.0031955242156982,
      "learning_rate": 2.006457564575646e-06,
      "loss": 1.4383,
      "step": 1555
    },
    {
      "epoch": 1.9365276913503422,
      "grad_norm": 3.0006372928619385,
      "learning_rate": 2.004151291512915e-06,
      "loss": 1.4132,
      "step": 1556
    },
    {
      "epoch": 1.937772246421904,
      "grad_norm": 2.8013830184936523,
      "learning_rate": 2.0018450184501845e-06,
      "loss": 1.3738,
      "step": 1557
    },
    {
      "epoch": 1.9390168014934661,
      "grad_norm": 2.8243088722229004,
      "learning_rate": 1.999538745387454e-06,
      "loss": 1.3414,
      "step": 1558
    },
    {
      "epoch": 1.940261356565028,
      "grad_norm": 2.864657163619995,
      "learning_rate": 1.9972324723247234e-06,
      "loss": 1.3452,
      "step": 1559
    },
    {
      "epoch": 1.94150591163659,
      "grad_norm": 2.888096570968628,
      "learning_rate": 1.994926199261993e-06,
      "loss": 1.3919,
      "step": 1560
    },
    {
      "epoch": 1.94150591163659,
      "eval_loss": 1.6243816614151,
      "eval_runtime": 48.9206,
      "eval_samples_per_second": 20.441,
      "eval_steps_per_second": 0.859,
      "step": 1560
    },
    {
      "epoch": 1.9427504667081519,
      "grad_norm": 2.871358871459961,
      "learning_rate": 1.9926199261992622e-06,
      "loss": 1.4041,
      "step": 1561
    },
    {
      "epoch": 1.9439950217797137,
      "grad_norm": 2.8249785900115967,
      "learning_rate": 1.9903136531365317e-06,
      "loss": 1.3993,
      "step": 1562
    },
    {
      "epoch": 1.9452395768512756,
      "grad_norm": 2.8441622257232666,
      "learning_rate": 1.988007380073801e-06,
      "loss": 1.4237,
      "step": 1563
    },
    {
      "epoch": 1.9464841319228374,
      "grad_norm": 2.849748373031616,
      "learning_rate": 1.98570110701107e-06,
      "loss": 1.4004,
      "step": 1564
    },
    {
      "epoch": 1.9477286869943995,
      "grad_norm": 3.0251729488372803,
      "learning_rate": 1.9833948339483395e-06,
      "loss": 1.3856,
      "step": 1565
    },
    {
      "epoch": 1.9489732420659616,
      "grad_norm": 3.016796827316284,
      "learning_rate": 1.981088560885609e-06,
      "loss": 1.4412,
      "step": 1566
    },
    {
      "epoch": 1.9502177971375234,
      "grad_norm": 2.8642990589141846,
      "learning_rate": 1.9787822878228783e-06,
      "loss": 1.3923,
      "step": 1567
    },
    {
      "epoch": 1.9514623522090853,
      "grad_norm": 3.1414499282836914,
      "learning_rate": 1.9764760147601478e-06,
      "loss": 1.4632,
      "step": 1568
    },
    {
      "epoch": 1.952706907280647,
      "grad_norm": 3.001513957977295,
      "learning_rate": 1.974169741697417e-06,
      "loss": 1.4113,
      "step": 1569
    },
    {
      "epoch": 1.953951462352209,
      "grad_norm": 3.0168347358703613,
      "learning_rate": 1.9718634686346866e-06,
      "loss": 1.4621,
      "step": 1570
    },
    {
      "epoch": 1.953951462352209,
      "eval_loss": 1.6229702234268188,
      "eval_runtime": 47.8157,
      "eval_samples_per_second": 20.914,
      "eval_steps_per_second": 0.878,
      "step": 1570
    },
    {
      "epoch": 1.955196017423771,
      "grad_norm": 2.82951283454895,
      "learning_rate": 1.969557195571956e-06,
      "loss": 1.4078,
      "step": 1571
    },
    {
      "epoch": 1.9564405724953329,
      "grad_norm": 2.948712110519409,
      "learning_rate": 1.9672509225092255e-06,
      "loss": 1.3425,
      "step": 1572
    },
    {
      "epoch": 1.957685127566895,
      "grad_norm": 2.8607540130615234,
      "learning_rate": 1.9649446494464945e-06,
      "loss": 1.3507,
      "step": 1573
    },
    {
      "epoch": 1.9589296826384568,
      "grad_norm": 2.852407455444336,
      "learning_rate": 1.962638376383764e-06,
      "loss": 1.4256,
      "step": 1574
    },
    {
      "epoch": 1.9601742377100186,
      "grad_norm": 2.7342076301574707,
      "learning_rate": 1.9603321033210333e-06,
      "loss": 1.334,
      "step": 1575
    },
    {
      "epoch": 1.9614187927815805,
      "grad_norm": 2.900651454925537,
      "learning_rate": 1.9580258302583027e-06,
      "loss": 1.3887,
      "step": 1576
    },
    {
      "epoch": 1.9626633478531423,
      "grad_norm": 2.8871212005615234,
      "learning_rate": 1.955719557195572e-06,
      "loss": 1.4563,
      "step": 1577
    },
    {
      "epoch": 1.9639079029247044,
      "grad_norm": 2.9238333702087402,
      "learning_rate": 1.9534132841328416e-06,
      "loss": 1.4203,
      "step": 1578
    },
    {
      "epoch": 1.9651524579962665,
      "grad_norm": 2.9352502822875977,
      "learning_rate": 1.951107011070111e-06,
      "loss": 1.4425,
      "step": 1579
    },
    {
      "epoch": 1.9663970130678283,
      "grad_norm": 2.7428557872772217,
      "learning_rate": 1.9488007380073804e-06,
      "loss": 1.3578,
      "step": 1580
    },
    {
      "epoch": 1.9663970130678283,
      "eval_loss": 1.6171667575836182,
      "eval_runtime": 52.2103,
      "eval_samples_per_second": 19.153,
      "eval_steps_per_second": 0.804,
      "step": 1580
    },
    {
      "epoch": 1.9676415681393902,
      "grad_norm": 2.86552095413208,
      "learning_rate": 1.9464944649446494e-06,
      "loss": 1.4064,
      "step": 1581
    },
    {
      "epoch": 1.968886123210952,
      "grad_norm": 2.848663330078125,
      "learning_rate": 1.944188191881919e-06,
      "loss": 1.4168,
      "step": 1582
    },
    {
      "epoch": 1.9701306782825139,
      "grad_norm": 2.948513984680176,
      "learning_rate": 1.9418819188191883e-06,
      "loss": 1.4197,
      "step": 1583
    },
    {
      "epoch": 1.971375233354076,
      "grad_norm": 2.9728989601135254,
      "learning_rate": 1.9395756457564577e-06,
      "loss": 1.4375,
      "step": 1584
    },
    {
      "epoch": 1.9726197884256378,
      "grad_norm": 2.849546194076538,
      "learning_rate": 1.937269372693727e-06,
      "loss": 1.4057,
      "step": 1585
    },
    {
      "epoch": 1.9738643434971999,
      "grad_norm": 2.9513893127441406,
      "learning_rate": 1.9349630996309965e-06,
      "loss": 1.3849,
      "step": 1586
    },
    {
      "epoch": 1.9751088985687617,
      "grad_norm": 3.021533250808716,
      "learning_rate": 1.932656826568266e-06,
      "loss": 1.4118,
      "step": 1587
    },
    {
      "epoch": 1.9763534536403236,
      "grad_norm": 2.8262035846710205,
      "learning_rate": 1.9303505535055354e-06,
      "loss": 1.3507,
      "step": 1588
    },
    {
      "epoch": 1.9775980087118854,
      "grad_norm": 2.976349115371704,
      "learning_rate": 1.9280442804428044e-06,
      "loss": 1.3975,
      "step": 1589
    },
    {
      "epoch": 1.9788425637834475,
      "grad_norm": 2.988787889480591,
      "learning_rate": 1.925738007380074e-06,
      "loss": 1.3816,
      "step": 1590
    },
    {
      "epoch": 1.9788425637834475,
      "eval_loss": 1.6183580160140991,
      "eval_runtime": 47.1659,
      "eval_samples_per_second": 21.202,
      "eval_steps_per_second": 0.89,
      "step": 1590
    },
    {
      "epoch": 1.9800871188550093,
      "grad_norm": 2.9827136993408203,
      "learning_rate": 1.9234317343173432e-06,
      "loss": 1.4502,
      "step": 1591
    },
    {
      "epoch": 1.9813316739265714,
      "grad_norm": 2.9239609241485596,
      "learning_rate": 1.9211254612546126e-06,
      "loss": 1.4047,
      "step": 1592
    },
    {
      "epoch": 1.9825762289981332,
      "grad_norm": 2.916612386703491,
      "learning_rate": 1.918819188191882e-06,
      "loss": 1.4188,
      "step": 1593
    },
    {
      "epoch": 1.983820784069695,
      "grad_norm": 2.894780158996582,
      "learning_rate": 1.9165129151291515e-06,
      "loss": 1.3911,
      "step": 1594
    },
    {
      "epoch": 1.985065339141257,
      "grad_norm": 2.8818533420562744,
      "learning_rate": 1.914206642066421e-06,
      "loss": 1.4115,
      "step": 1595
    },
    {
      "epoch": 1.9863098942128188,
      "grad_norm": 2.8122730255126953,
      "learning_rate": 1.9119003690036903e-06,
      "loss": 1.4147,
      "step": 1596
    },
    {
      "epoch": 1.9875544492843809,
      "grad_norm": 3.0528125762939453,
      "learning_rate": 1.9095940959409593e-06,
      "loss": 1.3903,
      "step": 1597
    },
    {
      "epoch": 1.9887990043559427,
      "grad_norm": 2.944221019744873,
      "learning_rate": 1.907287822878229e-06,
      "loss": 1.4044,
      "step": 1598
    },
    {
      "epoch": 1.9900435594275048,
      "grad_norm": 2.785045862197876,
      "learning_rate": 1.9049815498154984e-06,
      "loss": 1.4238,
      "step": 1599
    },
    {
      "epoch": 1.9912881144990666,
      "grad_norm": 2.9685428142547607,
      "learning_rate": 1.9026752767527676e-06,
      "loss": 1.4523,
      "step": 1600
    },
    {
      "epoch": 1.9912881144990666,
      "eval_loss": 1.6160162687301636,
      "eval_runtime": 47.5378,
      "eval_samples_per_second": 21.036,
      "eval_steps_per_second": 0.884,
      "step": 1600
    },
    {
      "epoch": 1.9925326695706285,
      "grad_norm": 3.050858736038208,
      "learning_rate": 1.900369003690037e-06,
      "loss": 1.3857,
      "step": 1601
    },
    {
      "epoch": 1.9937772246421903,
      "grad_norm": 2.978673219680786,
      "learning_rate": 1.8980627306273065e-06,
      "loss": 1.3948,
      "step": 1602
    },
    {
      "epoch": 1.9950217797137524,
      "grad_norm": 2.8683884143829346,
      "learning_rate": 1.8957564575645759e-06,
      "loss": 1.383,
      "step": 1603
    },
    {
      "epoch": 1.9962663347853142,
      "grad_norm": 2.836437225341797,
      "learning_rate": 1.8934501845018453e-06,
      "loss": 1.4177,
      "step": 1604
    },
    {
      "epoch": 1.9975108898568763,
      "grad_norm": 2.8872523307800293,
      "learning_rate": 1.8911439114391145e-06,
      "loss": 1.3652,
      "step": 1605
    },
    {
      "epoch": 1.9987554449284382,
      "grad_norm": 2.9714925289154053,
      "learning_rate": 1.888837638376384e-06,
      "loss": 1.4065,
      "step": 1606
    },
    {
      "epoch": 2.0,
      "grad_norm": 2.8581159114837646,
      "learning_rate": 1.8865313653136534e-06,
      "loss": 1.4139,
      "step": 1607
    },
    {
      "epoch": 2.001244555071562,
      "grad_norm": 2.9979677200317383,
      "learning_rate": 1.8842250922509226e-06,
      "loss": 1.3277,
      "step": 1608
    },
    {
      "epoch": 2.0024891101431237,
      "grad_norm": 3.1065833568573,
      "learning_rate": 1.881918819188192e-06,
      "loss": 1.4148,
      "step": 1609
    },
    {
      "epoch": 2.0037336652146855,
      "grad_norm": 3.072094678878784,
      "learning_rate": 1.8796125461254614e-06,
      "loss": 1.359,
      "step": 1610
    },
    {
      "epoch": 2.0037336652146855,
      "eval_loss": 1.623691201210022,
      "eval_runtime": 43.8246,
      "eval_samples_per_second": 22.818,
      "eval_steps_per_second": 0.958,
      "step": 1610
    },
    {
      "epoch": 2.004978220286248,
      "grad_norm": 3.1043107509613037,
      "learning_rate": 1.8773062730627308e-06,
      "loss": 1.2892,
      "step": 1611
    },
    {
      "epoch": 2.0062227753578097,
      "grad_norm": 3.19337797164917,
      "learning_rate": 1.8750000000000003e-06,
      "loss": 1.3369,
      "step": 1612
    },
    {
      "epoch": 2.0074673304293715,
      "grad_norm": 3.238434314727783,
      "learning_rate": 1.8726937269372695e-06,
      "loss": 1.3339,
      "step": 1613
    },
    {
      "epoch": 2.0087118855009334,
      "grad_norm": 3.206210136413574,
      "learning_rate": 1.8703874538745389e-06,
      "loss": 1.3078,
      "step": 1614
    },
    {
      "epoch": 2.0099564405724952,
      "grad_norm": 3.3336570262908936,
      "learning_rate": 1.8680811808118083e-06,
      "loss": 1.3766,
      "step": 1615
    },
    {
      "epoch": 2.011200995644057,
      "grad_norm": 3.3670783042907715,
      "learning_rate": 1.8657749077490777e-06,
      "loss": 1.3779,
      "step": 1616
    },
    {
      "epoch": 2.0124455507156194,
      "grad_norm": 3.2298123836517334,
      "learning_rate": 1.863468634686347e-06,
      "loss": 1.3725,
      "step": 1617
    },
    {
      "epoch": 2.013690105787181,
      "grad_norm": 3.251560688018799,
      "learning_rate": 1.8611623616236164e-06,
      "loss": 1.3197,
      "step": 1618
    },
    {
      "epoch": 2.014934660858743,
      "grad_norm": 3.377767562866211,
      "learning_rate": 1.8588560885608858e-06,
      "loss": 1.3178,
      "step": 1619
    },
    {
      "epoch": 2.016179215930305,
      "grad_norm": 3.141448736190796,
      "learning_rate": 1.856549815498155e-06,
      "loss": 1.3402,
      "step": 1620
    },
    {
      "epoch": 2.016179215930305,
      "eval_loss": 1.6248869895935059,
      "eval_runtime": 52.3566,
      "eval_samples_per_second": 19.1,
      "eval_steps_per_second": 0.802,
      "step": 1620
    },
    {
      "epoch": 2.0174237710018668,
      "grad_norm": 3.2168662548065186,
      "learning_rate": 1.8542435424354244e-06,
      "loss": 1.3757,
      "step": 1621
    },
    {
      "epoch": 2.0186683260734286,
      "grad_norm": 3.399155616760254,
      "learning_rate": 1.8519372693726939e-06,
      "loss": 1.3949,
      "step": 1622
    },
    {
      "epoch": 2.0199128811449905,
      "grad_norm": 3.3192570209503174,
      "learning_rate": 1.8496309963099633e-06,
      "loss": 1.3109,
      "step": 1623
    },
    {
      "epoch": 2.0211574362165527,
      "grad_norm": 3.4673702716827393,
      "learning_rate": 1.8473247232472327e-06,
      "loss": 1.4209,
      "step": 1624
    },
    {
      "epoch": 2.0224019912881146,
      "grad_norm": 3.3889009952545166,
      "learning_rate": 1.845018450184502e-06,
      "loss": 1.3648,
      "step": 1625
    },
    {
      "epoch": 2.0236465463596764,
      "grad_norm": 3.3962509632110596,
      "learning_rate": 1.8427121771217713e-06,
      "loss": 1.432,
      "step": 1626
    },
    {
      "epoch": 2.0248911014312383,
      "grad_norm": 3.2030723094940186,
      "learning_rate": 1.8404059040590408e-06,
      "loss": 1.3882,
      "step": 1627
    },
    {
      "epoch": 2.0261356565028,
      "grad_norm": 3.262387990951538,
      "learning_rate": 1.83809963099631e-06,
      "loss": 1.372,
      "step": 1628
    },
    {
      "epoch": 2.027380211574362,
      "grad_norm": 3.2742042541503906,
      "learning_rate": 1.8357933579335796e-06,
      "loss": 1.3777,
      "step": 1629
    },
    {
      "epoch": 2.0286247666459243,
      "grad_norm": 3.147533893585205,
      "learning_rate": 1.8334870848708488e-06,
      "loss": 1.3462,
      "step": 1630
    },
    {
      "epoch": 2.0286247666459243,
      "eval_loss": 1.6222952604293823,
      "eval_runtime": 54.0648,
      "eval_samples_per_second": 18.496,
      "eval_steps_per_second": 0.777,
      "step": 1630
    },
    {
      "epoch": 2.029869321717486,
      "grad_norm": 3.2916760444641113,
      "learning_rate": 1.8311808118081182e-06,
      "loss": 1.3186,
      "step": 1631
    },
    {
      "epoch": 2.031113876789048,
      "grad_norm": 3.2263071537017822,
      "learning_rate": 1.8288745387453877e-06,
      "loss": 1.3411,
      "step": 1632
    },
    {
      "epoch": 2.03235843186061,
      "grad_norm": 3.3505237102508545,
      "learning_rate": 1.8265682656826569e-06,
      "loss": 1.3726,
      "step": 1633
    },
    {
      "epoch": 2.0336029869321717,
      "grad_norm": 3.3842458724975586,
      "learning_rate": 1.8242619926199265e-06,
      "loss": 1.3931,
      "step": 1634
    },
    {
      "epoch": 2.0348475420037335,
      "grad_norm": 3.261672019958496,
      "learning_rate": 1.8219557195571957e-06,
      "loss": 1.3413,
      "step": 1635
    },
    {
      "epoch": 2.0360920970752954,
      "grad_norm": 3.310851573944092,
      "learning_rate": 1.819649446494465e-06,
      "loss": 1.3712,
      "step": 1636
    },
    {
      "epoch": 2.0373366521468577,
      "grad_norm": 3.1508402824401855,
      "learning_rate": 1.8173431734317346e-06,
      "loss": 1.3332,
      "step": 1637
    },
    {
      "epoch": 2.0385812072184195,
      "grad_norm": 3.197026014328003,
      "learning_rate": 1.8150369003690038e-06,
      "loss": 1.3787,
      "step": 1638
    },
    {
      "epoch": 2.0398257622899814,
      "grad_norm": 3.2075631618499756,
      "learning_rate": 1.812730627306273e-06,
      "loss": 1.3838,
      "step": 1639
    },
    {
      "epoch": 2.041070317361543,
      "grad_norm": 3.1499602794647217,
      "learning_rate": 1.8104243542435426e-06,
      "loss": 1.3945,
      "step": 1640
    },
    {
      "epoch": 2.041070317361543,
      "eval_loss": 1.6184145212173462,
      "eval_runtime": 50.1434,
      "eval_samples_per_second": 19.943,
      "eval_steps_per_second": 0.838,
      "step": 1640
    },
    {
      "epoch": 2.042314872433105,
      "grad_norm": 3.1790080070495605,
      "learning_rate": 1.8081180811808118e-06,
      "loss": 1.3261,
      "step": 1641
    },
    {
      "epoch": 2.043559427504667,
      "grad_norm": 3.137925148010254,
      "learning_rate": 1.8058118081180815e-06,
      "loss": 1.3328,
      "step": 1642
    },
    {
      "epoch": 2.044803982576229,
      "grad_norm": 3.2554683685302734,
      "learning_rate": 1.8035055350553507e-06,
      "loss": 1.3293,
      "step": 1643
    },
    {
      "epoch": 2.046048537647791,
      "grad_norm": 3.3206231594085693,
      "learning_rate": 1.8011992619926199e-06,
      "loss": 1.3415,
      "step": 1644
    },
    {
      "epoch": 2.047293092719353,
      "grad_norm": 3.390183210372925,
      "learning_rate": 1.7988929889298895e-06,
      "loss": 1.3855,
      "step": 1645
    },
    {
      "epoch": 2.0485376477909147,
      "grad_norm": 3.5058743953704834,
      "learning_rate": 1.7965867158671587e-06,
      "loss": 1.3517,
      "step": 1646
    },
    {
      "epoch": 2.0497822028624766,
      "grad_norm": 3.34566593170166,
      "learning_rate": 1.7942804428044284e-06,
      "loss": 1.343,
      "step": 1647
    },
    {
      "epoch": 2.0510267579340384,
      "grad_norm": 3.3177366256713867,
      "learning_rate": 1.7919741697416976e-06,
      "loss": 1.3202,
      "step": 1648
    },
    {
      "epoch": 2.0522713130056003,
      "grad_norm": 3.260061025619507,
      "learning_rate": 1.7896678966789668e-06,
      "loss": 1.39,
      "step": 1649
    },
    {
      "epoch": 2.0535158680771626,
      "grad_norm": 3.246882677078247,
      "learning_rate": 1.7873616236162364e-06,
      "loss": 1.3591,
      "step": 1650
    },
    {
      "epoch": 2.0535158680771626,
      "eval_loss": 1.6233465671539307,
      "eval_runtime": 54.5853,
      "eval_samples_per_second": 18.32,
      "eval_steps_per_second": 0.769,
      "step": 1650
    },
    {
      "epoch": 2.0547604231487244,
      "grad_norm": 3.2423501014709473,
      "learning_rate": 1.7850553505535056e-06,
      "loss": 1.3653,
      "step": 1651
    },
    {
      "epoch": 2.0560049782202863,
      "grad_norm": 3.2285611629486084,
      "learning_rate": 1.7827490774907748e-06,
      "loss": 1.356,
      "step": 1652
    },
    {
      "epoch": 2.057249533291848,
      "grad_norm": 3.3764569759368896,
      "learning_rate": 1.7804428044280445e-06,
      "loss": 1.3982,
      "step": 1653
    },
    {
      "epoch": 2.05849408836341,
      "grad_norm": 3.390240430831909,
      "learning_rate": 1.7781365313653137e-06,
      "loss": 1.3606,
      "step": 1654
    },
    {
      "epoch": 2.059738643434972,
      "grad_norm": 3.301675319671631,
      "learning_rate": 1.7758302583025833e-06,
      "loss": 1.3707,
      "step": 1655
    },
    {
      "epoch": 2.060983198506534,
      "grad_norm": 3.2020740509033203,
      "learning_rate": 1.7735239852398525e-06,
      "loss": 1.2709,
      "step": 1656
    },
    {
      "epoch": 2.062227753578096,
      "grad_norm": 3.5533366203308105,
      "learning_rate": 1.7712177121771217e-06,
      "loss": 1.3734,
      "step": 1657
    },
    {
      "epoch": 2.063472308649658,
      "grad_norm": 3.539336681365967,
      "learning_rate": 1.7689114391143914e-06,
      "loss": 1.39,
      "step": 1658
    },
    {
      "epoch": 2.0647168637212197,
      "grad_norm": 3.355268716812134,
      "learning_rate": 1.7666051660516606e-06,
      "loss": 1.3949,
      "step": 1659
    },
    {
      "epoch": 2.0659614187927815,
      "grad_norm": 3.1635212898254395,
      "learning_rate": 1.7642988929889302e-06,
      "loss": 1.3321,
      "step": 1660
    },
    {
      "epoch": 2.0659614187927815,
      "eval_loss": 1.6189814805984497,
      "eval_runtime": 50.8958,
      "eval_samples_per_second": 19.648,
      "eval_steps_per_second": 0.825,
      "step": 1660
    },
    {
      "epoch": 2.0672059738643433,
      "grad_norm": 3.4177732467651367,
      "learning_rate": 1.7619926199261994e-06,
      "loss": 1.3824,
      "step": 1661
    },
    {
      "epoch": 2.068450528935905,
      "grad_norm": 3.3664519786834717,
      "learning_rate": 1.7596863468634687e-06,
      "loss": 1.3587,
      "step": 1662
    },
    {
      "epoch": 2.0696950840074675,
      "grad_norm": 3.2475030422210693,
      "learning_rate": 1.7573800738007383e-06,
      "loss": 1.3446,
      "step": 1663
    },
    {
      "epoch": 2.0709396390790293,
      "grad_norm": 3.354497194290161,
      "learning_rate": 1.7550738007380075e-06,
      "loss": 1.366,
      "step": 1664
    },
    {
      "epoch": 2.072184194150591,
      "grad_norm": 3.247157573699951,
      "learning_rate": 1.752767527675277e-06,
      "loss": 1.3831,
      "step": 1665
    },
    {
      "epoch": 2.073428749222153,
      "grad_norm": 3.3339738845825195,
      "learning_rate": 1.7504612546125463e-06,
      "loss": 1.3362,
      "step": 1666
    },
    {
      "epoch": 2.074673304293715,
      "grad_norm": 3.289401054382324,
      "learning_rate": 1.7481549815498156e-06,
      "loss": 1.3549,
      "step": 1667
    },
    {
      "epoch": 2.0759178593652767,
      "grad_norm": 3.5802483558654785,
      "learning_rate": 1.7458487084870852e-06,
      "loss": 1.3826,
      "step": 1668
    },
    {
      "epoch": 2.077162414436839,
      "grad_norm": 3.389045476913452,
      "learning_rate": 1.7435424354243544e-06,
      "loss": 1.3561,
      "step": 1669
    },
    {
      "epoch": 2.078406969508401,
      "grad_norm": 3.104553461074829,
      "learning_rate": 1.7412361623616236e-06,
      "loss": 1.313,
      "step": 1670
    },
    {
      "epoch": 2.078406969508401,
      "eval_loss": 1.6231248378753662,
      "eval_runtime": 47.7675,
      "eval_samples_per_second": 20.935,
      "eval_steps_per_second": 0.879,
      "step": 1670
    },
    {
      "epoch": 2.0796515245799627,
      "grad_norm": 3.247082233428955,
      "learning_rate": 1.7389298892988932e-06,
      "loss": 1.3544,
      "step": 1671
    },
    {
      "epoch": 2.0808960796515246,
      "grad_norm": 3.2196743488311768,
      "learning_rate": 1.7366236162361625e-06,
      "loss": 1.3013,
      "step": 1672
    },
    {
      "epoch": 2.0821406347230864,
      "grad_norm": 3.4537713527679443,
      "learning_rate": 1.7343173431734319e-06,
      "loss": 1.3486,
      "step": 1673
    },
    {
      "epoch": 2.0833851897946483,
      "grad_norm": 3.1996660232543945,
      "learning_rate": 1.7320110701107013e-06,
      "loss": 1.3524,
      "step": 1674
    },
    {
      "epoch": 2.08462974486621,
      "grad_norm": 3.4143941402435303,
      "learning_rate": 1.7297047970479705e-06,
      "loss": 1.4062,
      "step": 1675
    },
    {
      "epoch": 2.0858742999377724,
      "grad_norm": 3.329657793045044,
      "learning_rate": 1.7273985239852401e-06,
      "loss": 1.336,
      "step": 1676
    },
    {
      "epoch": 2.0871188550093343,
      "grad_norm": 3.490600347518921,
      "learning_rate": 1.7250922509225094e-06,
      "loss": 1.3845,
      "step": 1677
    },
    {
      "epoch": 2.088363410080896,
      "grad_norm": 3.2943880558013916,
      "learning_rate": 1.7227859778597788e-06,
      "loss": 1.3952,
      "step": 1678
    },
    {
      "epoch": 2.089607965152458,
      "grad_norm": 3.3676629066467285,
      "learning_rate": 1.7204797047970482e-06,
      "loss": 1.3816,
      "step": 1679
    },
    {
      "epoch": 2.09085252022402,
      "grad_norm": 3.3164591789245605,
      "learning_rate": 1.7181734317343174e-06,
      "loss": 1.4124,
      "step": 1680
    },
    {
      "epoch": 2.09085252022402,
      "eval_loss": 1.6207176446914673,
      "eval_runtime": 48.9133,
      "eval_samples_per_second": 20.444,
      "eval_steps_per_second": 0.859,
      "step": 1680
    },
    {
      "epoch": 2.0920970752955816,
      "grad_norm": 3.3276405334472656,
      "learning_rate": 1.7158671586715868e-06,
      "loss": 1.4032,
      "step": 1681
    },
    {
      "epoch": 2.093341630367144,
      "grad_norm": 3.2198235988616943,
      "learning_rate": 1.7135608856088563e-06,
      "loss": 1.3483,
      "step": 1682
    },
    {
      "epoch": 2.094586185438706,
      "grad_norm": 3.1618118286132812,
      "learning_rate": 1.7112546125461257e-06,
      "loss": 1.3478,
      "step": 1683
    },
    {
      "epoch": 2.0958307405102676,
      "grad_norm": 3.1534817218780518,
      "learning_rate": 1.708948339483395e-06,
      "loss": 1.3727,
      "step": 1684
    },
    {
      "epoch": 2.0970752955818295,
      "grad_norm": 3.230872631072998,
      "learning_rate": 1.7066420664206643e-06,
      "loss": 1.3604,
      "step": 1685
    },
    {
      "epoch": 2.0983198506533913,
      "grad_norm": 3.088921308517456,
      "learning_rate": 1.7043357933579337e-06,
      "loss": 1.3425,
      "step": 1686
    },
    {
      "epoch": 2.099564405724953,
      "grad_norm": 3.3279967308044434,
      "learning_rate": 1.7020295202952032e-06,
      "loss": 1.3365,
      "step": 1687
    },
    {
      "epoch": 2.100808960796515,
      "grad_norm": 3.342949628829956,
      "learning_rate": 1.6997232472324724e-06,
      "loss": 1.3641,
      "step": 1688
    },
    {
      "epoch": 2.1020535158680773,
      "grad_norm": 3.5477232933044434,
      "learning_rate": 1.6974169741697418e-06,
      "loss": 1.376,
      "step": 1689
    },
    {
      "epoch": 2.103298070939639,
      "grad_norm": 3.363581895828247,
      "learning_rate": 1.6951107011070112e-06,
      "loss": 1.3251,
      "step": 1690
    },
    {
      "epoch": 2.103298070939639,
      "eval_loss": 1.6247185468673706,
      "eval_runtime": 45.6567,
      "eval_samples_per_second": 21.903,
      "eval_steps_per_second": 0.92,
      "step": 1690
    },
    {
      "epoch": 2.104542626011201,
      "grad_norm": 3.526014566421509,
      "learning_rate": 1.6928044280442806e-06,
      "loss": 1.3873,
      "step": 1691
    },
    {
      "epoch": 2.105787181082763,
      "grad_norm": 3.36954665184021,
      "learning_rate": 1.6904981549815499e-06,
      "loss": 1.3708,
      "step": 1692
    },
    {
      "epoch": 2.1070317361543247,
      "grad_norm": 3.2767856121063232,
      "learning_rate": 1.6881918819188193e-06,
      "loss": 1.3022,
      "step": 1693
    },
    {
      "epoch": 2.1082762912258866,
      "grad_norm": 3.5752694606781006,
      "learning_rate": 1.6858856088560887e-06,
      "loss": 1.3486,
      "step": 1694
    },
    {
      "epoch": 2.109520846297449,
      "grad_norm": 3.247654676437378,
      "learning_rate": 1.6835793357933581e-06,
      "loss": 1.3567,
      "step": 1695
    },
    {
      "epoch": 2.1107654013690107,
      "grad_norm": 3.328662633895874,
      "learning_rate": 1.6812730627306275e-06,
      "loss": 1.3694,
      "step": 1696
    },
    {
      "epoch": 2.1120099564405725,
      "grad_norm": 3.2084357738494873,
      "learning_rate": 1.6789667896678968e-06,
      "loss": 1.3467,
      "step": 1697
    },
    {
      "epoch": 2.1132545115121344,
      "grad_norm": 3.4850761890411377,
      "learning_rate": 1.6766605166051662e-06,
      "loss": 1.3635,
      "step": 1698
    },
    {
      "epoch": 2.1144990665836962,
      "grad_norm": 3.40693998336792,
      "learning_rate": 1.6743542435424356e-06,
      "loss": 1.3589,
      "step": 1699
    },
    {
      "epoch": 2.115743621655258,
      "grad_norm": 3.430428981781006,
      "learning_rate": 1.6720479704797048e-06,
      "loss": 1.4017,
      "step": 1700
    },
    {
      "epoch": 2.115743621655258,
      "eval_loss": 1.623623251914978,
      "eval_runtime": 46.6136,
      "eval_samples_per_second": 21.453,
      "eval_steps_per_second": 0.901,
      "step": 1700
    },
    {
      "epoch": 2.11698817672682,
      "grad_norm": 3.4609930515289307,
      "learning_rate": 1.6697416974169742e-06,
      "loss": 1.3514,
      "step": 1701
    },
    {
      "epoch": 2.1182327317983822,
      "grad_norm": 3.4114904403686523,
      "learning_rate": 1.6674354243542437e-06,
      "loss": 1.4017,
      "step": 1702
    },
    {
      "epoch": 2.119477286869944,
      "grad_norm": 3.4334566593170166,
      "learning_rate": 1.665129151291513e-06,
      "loss": 1.3339,
      "step": 1703
    },
    {
      "epoch": 2.120721841941506,
      "grad_norm": 3.3242125511169434,
      "learning_rate": 1.6628228782287825e-06,
      "loss": 1.3739,
      "step": 1704
    },
    {
      "epoch": 2.1219663970130678,
      "grad_norm": 3.2655038833618164,
      "learning_rate": 1.6605166051660517e-06,
      "loss": 1.3473,
      "step": 1705
    },
    {
      "epoch": 2.1232109520846296,
      "grad_norm": 3.2794158458709717,
      "learning_rate": 1.6582103321033211e-06,
      "loss": 1.3431,
      "step": 1706
    },
    {
      "epoch": 2.1244555071561915,
      "grad_norm": 3.4481866359710693,
      "learning_rate": 1.6559040590405906e-06,
      "loss": 1.3343,
      "step": 1707
    },
    {
      "epoch": 2.1257000622277538,
      "grad_norm": 3.250974178314209,
      "learning_rate": 1.6535977859778598e-06,
      "loss": 1.3255,
      "step": 1708
    },
    {
      "epoch": 2.1269446172993156,
      "grad_norm": 3.2407114505767822,
      "learning_rate": 1.6512915129151294e-06,
      "loss": 1.3587,
      "step": 1709
    },
    {
      "epoch": 2.1281891723708775,
      "grad_norm": 3.4453704357147217,
      "learning_rate": 1.6489852398523986e-06,
      "loss": 1.3937,
      "step": 1710
    },
    {
      "epoch": 2.1281891723708775,
      "eval_loss": 1.6269443035125732,
      "eval_runtime": 42.294,
      "eval_samples_per_second": 23.644,
      "eval_steps_per_second": 0.993,
      "step": 1710
    },
    {
      "epoch": 2.1294337274424393,
      "grad_norm": 3.4088101387023926,
      "learning_rate": 1.6466789667896678e-06,
      "loss": 1.389,
      "step": 1711
    },
    {
      "epoch": 2.130678282514001,
      "grad_norm": 3.3215279579162598,
      "learning_rate": 1.6443726937269375e-06,
      "loss": 1.3685,
      "step": 1712
    },
    {
      "epoch": 2.131922837585563,
      "grad_norm": 3.3293471336364746,
      "learning_rate": 1.6420664206642067e-06,
      "loss": 1.3714,
      "step": 1713
    },
    {
      "epoch": 2.133167392657125,
      "grad_norm": 3.232369899749756,
      "learning_rate": 1.6397601476014763e-06,
      "loss": 1.3614,
      "step": 1714
    },
    {
      "epoch": 2.134411947728687,
      "grad_norm": 3.3835854530334473,
      "learning_rate": 1.6374538745387455e-06,
      "loss": 1.3509,
      "step": 1715
    },
    {
      "epoch": 2.135656502800249,
      "grad_norm": 3.2008566856384277,
      "learning_rate": 1.6351476014760147e-06,
      "loss": 1.2935,
      "step": 1716
    },
    {
      "epoch": 2.136901057871811,
      "grad_norm": 3.4421873092651367,
      "learning_rate": 1.6328413284132844e-06,
      "loss": 1.3442,
      "step": 1717
    },
    {
      "epoch": 2.1381456129433727,
      "grad_norm": 3.379856586456299,
      "learning_rate": 1.6305350553505536e-06,
      "loss": 1.3917,
      "step": 1718
    },
    {
      "epoch": 2.1393901680149345,
      "grad_norm": 3.5044374465942383,
      "learning_rate": 1.6282287822878228e-06,
      "loss": 1.4052,
      "step": 1719
    },
    {
      "epoch": 2.1406347230864964,
      "grad_norm": 3.3533782958984375,
      "learning_rate": 1.6259225092250924e-06,
      "loss": 1.3435,
      "step": 1720
    },
    {
      "epoch": 2.1406347230864964,
      "eval_loss": 1.6230002641677856,
      "eval_runtime": 45.9961,
      "eval_samples_per_second": 21.741,
      "eval_steps_per_second": 0.913,
      "step": 1720
    },
    {
      "epoch": 2.1418792781580587,
      "grad_norm": 3.356597661972046,
      "learning_rate": 1.6236162361623616e-06,
      "loss": 1.3176,
      "step": 1721
    },
    {
      "epoch": 2.1431238332296205,
      "grad_norm": 3.204878807067871,
      "learning_rate": 1.6213099630996313e-06,
      "loss": 1.3553,
      "step": 1722
    },
    {
      "epoch": 2.1443683883011824,
      "grad_norm": 3.2749500274658203,
      "learning_rate": 1.6190036900369005e-06,
      "loss": 1.327,
      "step": 1723
    },
    {
      "epoch": 2.145612943372744,
      "grad_norm": 3.3262641429901123,
      "learning_rate": 1.6166974169741697e-06,
      "loss": 1.3453,
      "step": 1724
    },
    {
      "epoch": 2.146857498444306,
      "grad_norm": 3.351846218109131,
      "learning_rate": 1.6143911439114393e-06,
      "loss": 1.334,
      "step": 1725
    },
    {
      "epoch": 2.148102053515868,
      "grad_norm": 3.2748796939849854,
      "learning_rate": 1.6120848708487085e-06,
      "loss": 1.3178,
      "step": 1726
    },
    {
      "epoch": 2.1493466085874298,
      "grad_norm": 3.261385917663574,
      "learning_rate": 1.6097785977859782e-06,
      "loss": 1.317,
      "step": 1727
    },
    {
      "epoch": 2.150591163658992,
      "grad_norm": 3.4880056381225586,
      "learning_rate": 1.6074723247232474e-06,
      "loss": 1.377,
      "step": 1728
    },
    {
      "epoch": 2.151835718730554,
      "grad_norm": 3.3214828968048096,
      "learning_rate": 1.6051660516605166e-06,
      "loss": 1.3757,
      "step": 1729
    },
    {
      "epoch": 2.1530802738021158,
      "grad_norm": 3.3787684440612793,
      "learning_rate": 1.6028597785977862e-06,
      "loss": 1.3458,
      "step": 1730
    },
    {
      "epoch": 2.1530802738021158,
      "eval_loss": 1.6261134147644043,
      "eval_runtime": 45.3233,
      "eval_samples_per_second": 22.064,
      "eval_steps_per_second": 0.927,
      "step": 1730
    },
    {
      "epoch": 2.1543248288736776,
      "grad_norm": 3.423396587371826,
      "learning_rate": 1.6005535055350554e-06,
      "loss": 1.3543,
      "step": 1731
    },
    {
      "epoch": 2.1555693839452394,
      "grad_norm": 3.432394504547119,
      "learning_rate": 1.5982472324723247e-06,
      "loss": 1.3548,
      "step": 1732
    },
    {
      "epoch": 2.1568139390168013,
      "grad_norm": 3.3628666400909424,
      "learning_rate": 1.5959409594095943e-06,
      "loss": 1.3619,
      "step": 1733
    },
    {
      "epoch": 2.1580584940883636,
      "grad_norm": 3.3048722743988037,
      "learning_rate": 1.5936346863468635e-06,
      "loss": 1.3565,
      "step": 1734
    },
    {
      "epoch": 2.1593030491599254,
      "grad_norm": 3.21997332572937,
      "learning_rate": 1.5913284132841331e-06,
      "loss": 1.3353,
      "step": 1735
    },
    {
      "epoch": 2.1605476042314873,
      "grad_norm": 3.272420644760132,
      "learning_rate": 1.5890221402214023e-06,
      "loss": 1.3155,
      "step": 1736
    },
    {
      "epoch": 2.161792159303049,
      "grad_norm": 3.246156692504883,
      "learning_rate": 1.5867158671586716e-06,
      "loss": 1.3636,
      "step": 1737
    },
    {
      "epoch": 2.163036714374611,
      "grad_norm": 3.3257296085357666,
      "learning_rate": 1.5844095940959412e-06,
      "loss": 1.3748,
      "step": 1738
    },
    {
      "epoch": 2.164281269446173,
      "grad_norm": 3.298217535018921,
      "learning_rate": 1.5821033210332104e-06,
      "loss": 1.3171,
      "step": 1739
    },
    {
      "epoch": 2.1655258245177347,
      "grad_norm": 3.4946815967559814,
      "learning_rate": 1.57979704797048e-06,
      "loss": 1.3971,
      "step": 1740
    },
    {
      "epoch": 2.1655258245177347,
      "eval_loss": 1.6242249011993408,
      "eval_runtime": 46.2522,
      "eval_samples_per_second": 21.621,
      "eval_steps_per_second": 0.908,
      "step": 1740
    },
    {
      "epoch": 2.166770379589297,
      "grad_norm": 3.1652512550354004,
      "learning_rate": 1.5774907749077492e-06,
      "loss": 1.3332,
      "step": 1741
    },
    {
      "epoch": 2.168014934660859,
      "grad_norm": 3.292405843734741,
      "learning_rate": 1.5751845018450185e-06,
      "loss": 1.3105,
      "step": 1742
    },
    {
      "epoch": 2.1692594897324207,
      "grad_norm": 3.3213112354278564,
      "learning_rate": 1.572878228782288e-06,
      "loss": 1.372,
      "step": 1743
    },
    {
      "epoch": 2.1705040448039825,
      "grad_norm": 3.1998512744903564,
      "learning_rate": 1.5705719557195573e-06,
      "loss": 1.36,
      "step": 1744
    },
    {
      "epoch": 2.1717485998755444,
      "grad_norm": 3.449044942855835,
      "learning_rate": 1.5682656826568267e-06,
      "loss": 1.3262,
      "step": 1745
    },
    {
      "epoch": 2.172993154947106,
      "grad_norm": 3.435952663421631,
      "learning_rate": 1.5659594095940962e-06,
      "loss": 1.3473,
      "step": 1746
    },
    {
      "epoch": 2.1742377100186685,
      "grad_norm": 3.211683511734009,
      "learning_rate": 1.5636531365313654e-06,
      "loss": 1.3676,
      "step": 1747
    },
    {
      "epoch": 2.1754822650902303,
      "grad_norm": 3.3525032997131348,
      "learning_rate": 1.561346863468635e-06,
      "loss": 1.3792,
      "step": 1748
    },
    {
      "epoch": 2.176726820161792,
      "grad_norm": 3.3060429096221924,
      "learning_rate": 1.5590405904059042e-06,
      "loss": 1.3104,
      "step": 1749
    },
    {
      "epoch": 2.177971375233354,
      "grad_norm": 3.1698927879333496,
      "learning_rate": 1.5567343173431734e-06,
      "loss": 1.3142,
      "step": 1750
    },
    {
      "epoch": 2.177971375233354,
      "eval_loss": 1.6239022016525269,
      "eval_runtime": 47.4352,
      "eval_samples_per_second": 21.081,
      "eval_steps_per_second": 0.885,
      "step": 1750
    },
    {
      "epoch": 2.179215930304916,
      "grad_norm": 3.3830013275146484,
      "learning_rate": 1.554428044280443e-06,
      "loss": 1.38,
      "step": 1751
    },
    {
      "epoch": 2.1804604853764777,
      "grad_norm": 3.144509792327881,
      "learning_rate": 1.5521217712177123e-06,
      "loss": 1.3447,
      "step": 1752
    },
    {
      "epoch": 2.18170504044804,
      "grad_norm": 3.2411561012268066,
      "learning_rate": 1.5498154981549817e-06,
      "loss": 1.3709,
      "step": 1753
    },
    {
      "epoch": 2.182949595519602,
      "grad_norm": 3.19698429107666,
      "learning_rate": 1.5475092250922511e-06,
      "loss": 1.3068,
      "step": 1754
    },
    {
      "epoch": 2.1841941505911637,
      "grad_norm": 3.301417589187622,
      "learning_rate": 1.5452029520295203e-06,
      "loss": 1.3219,
      "step": 1755
    },
    {
      "epoch": 2.1854387056627256,
      "grad_norm": 3.3205950260162354,
      "learning_rate": 1.5428966789667897e-06,
      "loss": 1.333,
      "step": 1756
    },
    {
      "epoch": 2.1866832607342874,
      "grad_norm": 3.3267645835876465,
      "learning_rate": 1.5405904059040592e-06,
      "loss": 1.3536,
      "step": 1757
    },
    {
      "epoch": 2.1879278158058493,
      "grad_norm": 3.37038254737854,
      "learning_rate": 1.5382841328413286e-06,
      "loss": 1.3481,
      "step": 1758
    },
    {
      "epoch": 2.189172370877411,
      "grad_norm": 3.5459911823272705,
      "learning_rate": 1.535977859778598e-06,
      "loss": 1.3798,
      "step": 1759
    },
    {
      "epoch": 2.1904169259489734,
      "grad_norm": 3.40732741355896,
      "learning_rate": 1.5336715867158672e-06,
      "loss": 1.3672,
      "step": 1760
    },
    {
      "epoch": 2.1904169259489734,
      "eval_loss": 1.6204931735992432,
      "eval_runtime": 45.5831,
      "eval_samples_per_second": 21.938,
      "eval_steps_per_second": 0.921,
      "step": 1760
    },
    {
      "epoch": 2.1916614810205353,
      "grad_norm": 3.218372106552124,
      "learning_rate": 1.5313653136531366e-06,
      "loss": 1.3486,
      "step": 1761
    },
    {
      "epoch": 2.192906036092097,
      "grad_norm": 3.370919942855835,
      "learning_rate": 1.529059040590406e-06,
      "loss": 1.3927,
      "step": 1762
    },
    {
      "epoch": 2.194150591163659,
      "grad_norm": 3.3146965503692627,
      "learning_rate": 1.5267527675276755e-06,
      "loss": 1.3539,
      "step": 1763
    },
    {
      "epoch": 2.195395146235221,
      "grad_norm": 3.410720109939575,
      "learning_rate": 1.5244464944649447e-06,
      "loss": 1.3744,
      "step": 1764
    },
    {
      "epoch": 2.1966397013067827,
      "grad_norm": 3.2189249992370605,
      "learning_rate": 1.5221402214022141e-06,
      "loss": 1.36,
      "step": 1765
    },
    {
      "epoch": 2.197884256378345,
      "grad_norm": 3.149064064025879,
      "learning_rate": 1.5198339483394836e-06,
      "loss": 1.2923,
      "step": 1766
    },
    {
      "epoch": 2.199128811449907,
      "grad_norm": 3.1745471954345703,
      "learning_rate": 1.517527675276753e-06,
      "loss": 1.3394,
      "step": 1767
    },
    {
      "epoch": 2.2003733665214686,
      "grad_norm": 3.3207221031188965,
      "learning_rate": 1.5152214022140222e-06,
      "loss": 1.3309,
      "step": 1768
    },
    {
      "epoch": 2.2016179215930305,
      "grad_norm": 3.3948538303375244,
      "learning_rate": 1.5129151291512916e-06,
      "loss": 1.3425,
      "step": 1769
    },
    {
      "epoch": 2.2028624766645923,
      "grad_norm": 3.4169349670410156,
      "learning_rate": 1.510608856088561e-06,
      "loss": 1.3285,
      "step": 1770
    },
    {
      "epoch": 2.2028624766645923,
      "eval_loss": 1.6240391731262207,
      "eval_runtime": 47.9216,
      "eval_samples_per_second": 20.867,
      "eval_steps_per_second": 0.876,
      "step": 1770
    },
    {
      "epoch": 2.204107031736154,
      "grad_norm": 3.359754800796509,
      "learning_rate": 1.5083025830258305e-06,
      "loss": 1.3369,
      "step": 1771
    },
    {
      "epoch": 2.205351586807716,
      "grad_norm": 3.280728340148926,
      "learning_rate": 1.5059963099630997e-06,
      "loss": 1.3381,
      "step": 1772
    },
    {
      "epoch": 2.2065961418792783,
      "grad_norm": 3.4326274394989014,
      "learning_rate": 1.503690036900369e-06,
      "loss": 1.383,
      "step": 1773
    },
    {
      "epoch": 2.20784069695084,
      "grad_norm": 3.3060855865478516,
      "learning_rate": 1.5013837638376385e-06,
      "loss": 1.36,
      "step": 1774
    },
    {
      "epoch": 2.209085252022402,
      "grad_norm": 3.369748115539551,
      "learning_rate": 1.499077490774908e-06,
      "loss": 1.3699,
      "step": 1775
    },
    {
      "epoch": 2.210329807093964,
      "grad_norm": 3.203705072402954,
      "learning_rate": 1.4967712177121774e-06,
      "loss": 1.3314,
      "step": 1776
    },
    {
      "epoch": 2.2115743621655257,
      "grad_norm": 3.255790948867798,
      "learning_rate": 1.4944649446494466e-06,
      "loss": 1.3791,
      "step": 1777
    },
    {
      "epoch": 2.2128189172370876,
      "grad_norm": 3.2817792892456055,
      "learning_rate": 1.492158671586716e-06,
      "loss": 1.3321,
      "step": 1778
    },
    {
      "epoch": 2.21406347230865,
      "grad_norm": 3.3101930618286133,
      "learning_rate": 1.4898523985239854e-06,
      "loss": 1.4065,
      "step": 1779
    },
    {
      "epoch": 2.2153080273802117,
      "grad_norm": 3.360100030899048,
      "learning_rate": 1.4875461254612546e-06,
      "loss": 1.3083,
      "step": 1780
    },
    {
      "epoch": 2.2153080273802117,
      "eval_loss": 1.625471591949463,
      "eval_runtime": 46.2605,
      "eval_samples_per_second": 21.617,
      "eval_steps_per_second": 0.908,
      "step": 1780
    },
    {
      "epoch": 2.2165525824517736,
      "grad_norm": 3.3503854274749756,
      "learning_rate": 1.485239852398524e-06,
      "loss": 1.374,
      "step": 1781
    },
    {
      "epoch": 2.2177971375233354,
      "grad_norm": 3.196617603302002,
      "learning_rate": 1.4829335793357935e-06,
      "loss": 1.3226,
      "step": 1782
    },
    {
      "epoch": 2.2190416925948973,
      "grad_norm": 3.3946502208709717,
      "learning_rate": 1.4806273062730627e-06,
      "loss": 1.3445,
      "step": 1783
    },
    {
      "epoch": 2.220286247666459,
      "grad_norm": 3.252145528793335,
      "learning_rate": 1.4783210332103323e-06,
      "loss": 1.3514,
      "step": 1784
    },
    {
      "epoch": 2.2215308027380214,
      "grad_norm": 3.452378511428833,
      "learning_rate": 1.4760147601476015e-06,
      "loss": 1.3386,
      "step": 1785
    },
    {
      "epoch": 2.2227753578095832,
      "grad_norm": 3.3612773418426514,
      "learning_rate": 1.473708487084871e-06,
      "loss": 1.3572,
      "step": 1786
    },
    {
      "epoch": 2.224019912881145,
      "grad_norm": 3.295044422149658,
      "learning_rate": 1.4714022140221404e-06,
      "loss": 1.328,
      "step": 1787
    },
    {
      "epoch": 2.225264467952707,
      "grad_norm": 3.2037711143493652,
      "learning_rate": 1.4690959409594096e-06,
      "loss": 1.2999,
      "step": 1788
    },
    {
      "epoch": 2.226509023024269,
      "grad_norm": 3.3932483196258545,
      "learning_rate": 1.4667896678966792e-06,
      "loss": 1.3606,
      "step": 1789
    },
    {
      "epoch": 2.2277535780958306,
      "grad_norm": 3.4563357830047607,
      "learning_rate": 1.4644833948339484e-06,
      "loss": 1.3875,
      "step": 1790
    },
    {
      "epoch": 2.2277535780958306,
      "eval_loss": 1.622439980506897,
      "eval_runtime": 47.9981,
      "eval_samples_per_second": 20.834,
      "eval_steps_per_second": 0.875,
      "step": 1790
    },
    {
      "epoch": 2.2289981331673925,
      "grad_norm": 3.3003604412078857,
      "learning_rate": 1.4621771217712176e-06,
      "loss": 1.3373,
      "step": 1791
    },
    {
      "epoch": 2.2302426882389548,
      "grad_norm": 3.398839235305786,
      "learning_rate": 1.4598708487084873e-06,
      "loss": 1.3432,
      "step": 1792
    },
    {
      "epoch": 2.2314872433105166,
      "grad_norm": 3.3929128646850586,
      "learning_rate": 1.4575645756457565e-06,
      "loss": 1.347,
      "step": 1793
    },
    {
      "epoch": 2.2327317983820785,
      "grad_norm": 3.21239972114563,
      "learning_rate": 1.4552583025830261e-06,
      "loss": 1.347,
      "step": 1794
    },
    {
      "epoch": 2.2339763534536403,
      "grad_norm": 3.242029905319214,
      "learning_rate": 1.4529520295202953e-06,
      "loss": 1.3256,
      "step": 1795
    },
    {
      "epoch": 2.235220908525202,
      "grad_norm": 3.3526346683502197,
      "learning_rate": 1.4506457564575645e-06,
      "loss": 1.315,
      "step": 1796
    },
    {
      "epoch": 2.236465463596764,
      "grad_norm": 3.351020574569702,
      "learning_rate": 1.4483394833948342e-06,
      "loss": 1.357,
      "step": 1797
    },
    {
      "epoch": 2.2377100186683263,
      "grad_norm": 3.402653694152832,
      "learning_rate": 1.4460332103321034e-06,
      "loss": 1.3695,
      "step": 1798
    },
    {
      "epoch": 2.238954573739888,
      "grad_norm": 3.2951724529266357,
      "learning_rate": 1.4437269372693726e-06,
      "loss": 1.3366,
      "step": 1799
    },
    {
      "epoch": 2.24019912881145,
      "grad_norm": 3.2638256549835205,
      "learning_rate": 1.4414206642066422e-06,
      "loss": 1.3147,
      "step": 1800
    },
    {
      "epoch": 2.24019912881145,
      "eval_loss": 1.6255451440811157,
      "eval_runtime": 49.0934,
      "eval_samples_per_second": 20.369,
      "eval_steps_per_second": 0.856,
      "step": 1800
    },
    {
      "epoch": 2.241443683883012,
      "grad_norm": 3.536435604095459,
      "learning_rate": 1.4391143911439114e-06,
      "loss": 1.3609,
      "step": 1801
    },
    {
      "epoch": 2.2426882389545737,
      "grad_norm": 3.219020366668701,
      "learning_rate": 1.436808118081181e-06,
      "loss": 1.3291,
      "step": 1802
    },
    {
      "epoch": 2.2439327940261355,
      "grad_norm": 3.1892435550689697,
      "learning_rate": 1.4345018450184503e-06,
      "loss": 1.3384,
      "step": 1803
    },
    {
      "epoch": 2.2451773490976974,
      "grad_norm": 3.2276196479797363,
      "learning_rate": 1.4321955719557195e-06,
      "loss": 1.3181,
      "step": 1804
    },
    {
      "epoch": 2.2464219041692597,
      "grad_norm": 3.317469596862793,
      "learning_rate": 1.4298892988929891e-06,
      "loss": 1.3088,
      "step": 1805
    },
    {
      "epoch": 2.2476664592408215,
      "grad_norm": 3.4198920726776123,
      "learning_rate": 1.4275830258302584e-06,
      "loss": 1.2905,
      "step": 1806
    },
    {
      "epoch": 2.2489110143123834,
      "grad_norm": 3.4940414428710938,
      "learning_rate": 1.425276752767528e-06,
      "loss": 1.3419,
      "step": 1807
    },
    {
      "epoch": 2.2501555693839452,
      "grad_norm": 3.4554691314697266,
      "learning_rate": 1.4229704797047972e-06,
      "loss": 1.3558,
      "step": 1808
    },
    {
      "epoch": 2.251400124455507,
      "grad_norm": 3.490118980407715,
      "learning_rate": 1.4206642066420664e-06,
      "loss": 1.3639,
      "step": 1809
    },
    {
      "epoch": 2.252644679527069,
      "grad_norm": 3.452986717224121,
      "learning_rate": 1.418357933579336e-06,
      "loss": 1.343,
      "step": 1810
    },
    {
      "epoch": 2.252644679527069,
      "eval_loss": 1.6230783462524414,
      "eval_runtime": 43.9128,
      "eval_samples_per_second": 22.772,
      "eval_steps_per_second": 0.956,
      "step": 1810
    },
    {
      "epoch": 2.253889234598631,
      "grad_norm": 3.3851468563079834,
      "learning_rate": 1.4160516605166053e-06,
      "loss": 1.3067,
      "step": 1811
    },
    {
      "epoch": 2.255133789670193,
      "grad_norm": 3.373572826385498,
      "learning_rate": 1.4137453874538745e-06,
      "loss": 1.3826,
      "step": 1812
    },
    {
      "epoch": 2.256378344741755,
      "grad_norm": 3.5561487674713135,
      "learning_rate": 1.411439114391144e-06,
      "loss": 1.3662,
      "step": 1813
    },
    {
      "epoch": 2.2576228998133168,
      "grad_norm": 3.4163501262664795,
      "learning_rate": 1.4091328413284133e-06,
      "loss": 1.3664,
      "step": 1814
    },
    {
      "epoch": 2.2588674548848786,
      "grad_norm": 3.2873945236206055,
      "learning_rate": 1.406826568265683e-06,
      "loss": 1.3277,
      "step": 1815
    },
    {
      "epoch": 2.2601120099564405,
      "grad_norm": 3.2029595375061035,
      "learning_rate": 1.4045202952029522e-06,
      "loss": 1.3202,
      "step": 1816
    },
    {
      "epoch": 2.2613565650280023,
      "grad_norm": 3.376417398452759,
      "learning_rate": 1.4022140221402214e-06,
      "loss": 1.3264,
      "step": 1817
    },
    {
      "epoch": 2.2626011200995646,
      "grad_norm": 3.4385898113250732,
      "learning_rate": 1.399907749077491e-06,
      "loss": 1.3571,
      "step": 1818
    },
    {
      "epoch": 2.2638456751711264,
      "grad_norm": 3.342299461364746,
      "learning_rate": 1.3976014760147602e-06,
      "loss": 1.3385,
      "step": 1819
    },
    {
      "epoch": 2.2650902302426883,
      "grad_norm": 3.355975389480591,
      "learning_rate": 1.3952952029520298e-06,
      "loss": 1.346,
      "step": 1820
    },
    {
      "epoch": 2.2650902302426883,
      "eval_loss": 1.6243224143981934,
      "eval_runtime": 46.6848,
      "eval_samples_per_second": 21.42,
      "eval_steps_per_second": 0.9,
      "step": 1820
    },
    {
      "epoch": 2.26633478531425,
      "grad_norm": 3.4472744464874268,
      "learning_rate": 1.392988929889299e-06,
      "loss": 1.404,
      "step": 1821
    },
    {
      "epoch": 2.267579340385812,
      "grad_norm": 3.3384392261505127,
      "learning_rate": 1.3906826568265683e-06,
      "loss": 1.3492,
      "step": 1822
    },
    {
      "epoch": 2.268823895457374,
      "grad_norm": 3.3477087020874023,
      "learning_rate": 1.388376383763838e-06,
      "loss": 1.3565,
      "step": 1823
    },
    {
      "epoch": 2.270068450528936,
      "grad_norm": 3.3641061782836914,
      "learning_rate": 1.3860701107011071e-06,
      "loss": 1.344,
      "step": 1824
    },
    {
      "epoch": 2.271313005600498,
      "grad_norm": 3.435539960861206,
      "learning_rate": 1.3837638376383765e-06,
      "loss": 1.3709,
      "step": 1825
    },
    {
      "epoch": 2.27255756067206,
      "grad_norm": 3.349618434906006,
      "learning_rate": 1.381457564575646e-06,
      "loss": 1.3265,
      "step": 1826
    },
    {
      "epoch": 2.2738021157436217,
      "grad_norm": 3.402650833129883,
      "learning_rate": 1.3791512915129152e-06,
      "loss": 1.3537,
      "step": 1827
    },
    {
      "epoch": 2.2750466708151835,
      "grad_norm": 3.389064311981201,
      "learning_rate": 1.3768450184501846e-06,
      "loss": 1.3843,
      "step": 1828
    },
    {
      "epoch": 2.2762912258867454,
      "grad_norm": 3.3653721809387207,
      "learning_rate": 1.374538745387454e-06,
      "loss": 1.376,
      "step": 1829
    },
    {
      "epoch": 2.277535780958307,
      "grad_norm": 3.254830837249756,
      "learning_rate": 1.3722324723247232e-06,
      "loss": 1.344,
      "step": 1830
    },
    {
      "epoch": 2.277535780958307,
      "eval_loss": 1.621465802192688,
      "eval_runtime": 52.2643,
      "eval_samples_per_second": 19.134,
      "eval_steps_per_second": 0.804,
      "step": 1830
    },
    {
      "epoch": 2.2787803360298695,
      "grad_norm": 3.4339189529418945,
      "learning_rate": 1.3699261992619929e-06,
      "loss": 1.3539,
      "step": 1831
    },
    {
      "epoch": 2.2800248911014314,
      "grad_norm": 3.5471079349517822,
      "learning_rate": 1.367619926199262e-06,
      "loss": 1.4031,
      "step": 1832
    },
    {
      "epoch": 2.281269446172993,
      "grad_norm": 3.4783706665039062,
      "learning_rate": 1.3653136531365315e-06,
      "loss": 1.3504,
      "step": 1833
    },
    {
      "epoch": 2.282514001244555,
      "grad_norm": 3.360304355621338,
      "learning_rate": 1.363007380073801e-06,
      "loss": 1.3215,
      "step": 1834
    },
    {
      "epoch": 2.283758556316117,
      "grad_norm": 3.447568416595459,
      "learning_rate": 1.3607011070110701e-06,
      "loss": 1.3286,
      "step": 1835
    },
    {
      "epoch": 2.2850031113876788,
      "grad_norm": 3.311884641647339,
      "learning_rate": 1.3583948339483396e-06,
      "loss": 1.3288,
      "step": 1836
    },
    {
      "epoch": 2.286247666459241,
      "grad_norm": 3.3999180793762207,
      "learning_rate": 1.356088560885609e-06,
      "loss": 1.2837,
      "step": 1837
    },
    {
      "epoch": 2.287492221530803,
      "grad_norm": 3.360581636428833,
      "learning_rate": 1.3537822878228784e-06,
      "loss": 1.3233,
      "step": 1838
    },
    {
      "epoch": 2.2887367766023647,
      "grad_norm": 3.3952338695526123,
      "learning_rate": 1.3514760147601478e-06,
      "loss": 1.3702,
      "step": 1839
    },
    {
      "epoch": 2.2899813316739266,
      "grad_norm": 3.5232338905334473,
      "learning_rate": 1.349169741697417e-06,
      "loss": 1.4087,
      "step": 1840
    },
    {
      "epoch": 2.2899813316739266,
      "eval_loss": 1.6189563274383545,
      "eval_runtime": 51.0863,
      "eval_samples_per_second": 19.575,
      "eval_steps_per_second": 0.822,
      "step": 1840
    },
    {
      "epoch": 2.2912258867454884,
      "grad_norm": 3.4118707180023193,
      "learning_rate": 1.3468634686346865e-06,
      "loss": 1.3838,
      "step": 1841
    },
    {
      "epoch": 2.2924704418170503,
      "grad_norm": 3.511007070541382,
      "learning_rate": 1.3445571955719559e-06,
      "loss": 1.3846,
      "step": 1842
    },
    {
      "epoch": 2.293714996888612,
      "grad_norm": 3.44270920753479,
      "learning_rate": 1.3422509225092253e-06,
      "loss": 1.339,
      "step": 1843
    },
    {
      "epoch": 2.2949595519601744,
      "grad_norm": 3.322871446609497,
      "learning_rate": 1.3399446494464945e-06,
      "loss": 1.289,
      "step": 1844
    },
    {
      "epoch": 2.2962041070317363,
      "grad_norm": 3.3219494819641113,
      "learning_rate": 1.337638376383764e-06,
      "loss": 1.3613,
      "step": 1845
    },
    {
      "epoch": 2.297448662103298,
      "grad_norm": 3.3900601863861084,
      "learning_rate": 1.3353321033210334e-06,
      "loss": 1.3516,
      "step": 1846
    },
    {
      "epoch": 2.29869321717486,
      "grad_norm": 3.4780476093292236,
      "learning_rate": 1.3330258302583028e-06,
      "loss": 1.3779,
      "step": 1847
    },
    {
      "epoch": 2.299937772246422,
      "grad_norm": 3.485691547393799,
      "learning_rate": 1.330719557195572e-06,
      "loss": 1.3,
      "step": 1848
    },
    {
      "epoch": 2.3011823273179837,
      "grad_norm": 3.456273317337036,
      "learning_rate": 1.3284132841328414e-06,
      "loss": 1.3946,
      "step": 1849
    },
    {
      "epoch": 2.302426882389546,
      "grad_norm": 3.2921931743621826,
      "learning_rate": 1.3261070110701108e-06,
      "loss": 1.3338,
      "step": 1850
    },
    {
      "epoch": 2.302426882389546,
      "eval_loss": 1.6193547248840332,
      "eval_runtime": 51.2837,
      "eval_samples_per_second": 19.499,
      "eval_steps_per_second": 0.819,
      "step": 1850
    },
    {
      "epoch": 2.303671437461108,
      "grad_norm": 3.311541795730591,
      "learning_rate": 1.3238007380073803e-06,
      "loss": 1.344,
      "step": 1851
    },
    {
      "epoch": 2.3049159925326697,
      "grad_norm": 3.2640180587768555,
      "learning_rate": 1.3214944649446495e-06,
      "loss": 1.3345,
      "step": 1852
    },
    {
      "epoch": 2.3061605476042315,
      "grad_norm": 3.3187389373779297,
      "learning_rate": 1.319188191881919e-06,
      "loss": 1.3325,
      "step": 1853
    },
    {
      "epoch": 2.3074051026757934,
      "grad_norm": 3.478851795196533,
      "learning_rate": 1.3168819188191883e-06,
      "loss": 1.3748,
      "step": 1854
    },
    {
      "epoch": 2.308649657747355,
      "grad_norm": 3.32222580909729,
      "learning_rate": 1.3145756457564575e-06,
      "loss": 1.3538,
      "step": 1855
    },
    {
      "epoch": 2.309894212818917,
      "grad_norm": 3.2635533809661865,
      "learning_rate": 1.3122693726937272e-06,
      "loss": 1.3459,
      "step": 1856
    },
    {
      "epoch": 2.3111387678904793,
      "grad_norm": 3.335123300552368,
      "learning_rate": 1.3099630996309964e-06,
      "loss": 1.3621,
      "step": 1857
    },
    {
      "epoch": 2.312383322962041,
      "grad_norm": 3.4098165035247803,
      "learning_rate": 1.3076568265682658e-06,
      "loss": 1.3431,
      "step": 1858
    },
    {
      "epoch": 2.313627878033603,
      "grad_norm": 3.2582387924194336,
      "learning_rate": 1.3053505535055352e-06,
      "loss": 1.3591,
      "step": 1859
    },
    {
      "epoch": 2.314872433105165,
      "grad_norm": 3.2456133365631104,
      "learning_rate": 1.3030442804428044e-06,
      "loss": 1.3367,
      "step": 1860
    },
    {
      "epoch": 2.314872433105165,
      "eval_loss": 1.6218940019607544,
      "eval_runtime": 54.4652,
      "eval_samples_per_second": 18.36,
      "eval_steps_per_second": 0.771,
      "step": 1860
    },
    {
      "epoch": 2.3161169881767267,
      "grad_norm": 3.220767021179199,
      "learning_rate": 1.3007380073800739e-06,
      "loss": 1.2744,
      "step": 1861
    },
    {
      "epoch": 2.3173615432482886,
      "grad_norm": 3.4657704830169678,
      "learning_rate": 1.2984317343173433e-06,
      "loss": 1.366,
      "step": 1862
    },
    {
      "epoch": 2.318606098319851,
      "grad_norm": 3.399174928665161,
      "learning_rate": 1.2961254612546125e-06,
      "loss": 1.3775,
      "step": 1863
    },
    {
      "epoch": 2.3198506533914127,
      "grad_norm": 3.4700374603271484,
      "learning_rate": 1.2938191881918821e-06,
      "loss": 1.3456,
      "step": 1864
    },
    {
      "epoch": 2.3210952084629746,
      "grad_norm": 3.2533438205718994,
      "learning_rate": 1.2915129151291513e-06,
      "loss": 1.2934,
      "step": 1865
    },
    {
      "epoch": 2.3223397635345364,
      "grad_norm": 3.4083049297332764,
      "learning_rate": 1.2892066420664208e-06,
      "loss": 1.3759,
      "step": 1866
    },
    {
      "epoch": 2.3235843186060983,
      "grad_norm": 3.3690836429595947,
      "learning_rate": 1.2869003690036902e-06,
      "loss": 1.313,
      "step": 1867
    },
    {
      "epoch": 2.32482887367766,
      "grad_norm": 3.4750704765319824,
      "learning_rate": 1.2845940959409594e-06,
      "loss": 1.3658,
      "step": 1868
    },
    {
      "epoch": 2.326073428749222,
      "grad_norm": 3.458674192428589,
      "learning_rate": 1.282287822878229e-06,
      "loss": 1.4026,
      "step": 1869
    },
    {
      "epoch": 2.3273179838207843,
      "grad_norm": 3.5049021244049072,
      "learning_rate": 1.2799815498154982e-06,
      "loss": 1.3638,
      "step": 1870
    },
    {
      "epoch": 2.3273179838207843,
      "eval_loss": 1.622696876525879,
      "eval_runtime": 49.0313,
      "eval_samples_per_second": 20.395,
      "eval_steps_per_second": 0.857,
      "step": 1870
    },
    {
      "epoch": 2.328562538892346,
      "grad_norm": 3.3939497470855713,
      "learning_rate": 1.2776752767527675e-06,
      "loss": 1.3198,
      "step": 1871
    },
    {
      "epoch": 2.329807093963908,
      "grad_norm": 3.3821396827697754,
      "learning_rate": 1.275369003690037e-06,
      "loss": 1.3682,
      "step": 1872
    },
    {
      "epoch": 2.33105164903547,
      "grad_norm": 3.6034491062164307,
      "learning_rate": 1.2730627306273063e-06,
      "loss": 1.3476,
      "step": 1873
    },
    {
      "epoch": 2.3322962041070316,
      "grad_norm": 3.3996806144714355,
      "learning_rate": 1.270756457564576e-06,
      "loss": 1.3239,
      "step": 1874
    },
    {
      "epoch": 2.3335407591785935,
      "grad_norm": 3.2999489307403564,
      "learning_rate": 1.2684501845018451e-06,
      "loss": 1.311,
      "step": 1875
    },
    {
      "epoch": 2.334785314250156,
      "grad_norm": 3.292593002319336,
      "learning_rate": 1.2661439114391144e-06,
      "loss": 1.3333,
      "step": 1876
    },
    {
      "epoch": 2.3360298693217176,
      "grad_norm": 3.323923349380493,
      "learning_rate": 1.263837638376384e-06,
      "loss": 1.3154,
      "step": 1877
    },
    {
      "epoch": 2.3372744243932795,
      "grad_norm": 3.2100625038146973,
      "learning_rate": 1.2615313653136532e-06,
      "loss": 1.3248,
      "step": 1878
    },
    {
      "epoch": 2.3385189794648413,
      "grad_norm": 3.3764309883117676,
      "learning_rate": 1.2592250922509224e-06,
      "loss": 1.3574,
      "step": 1879
    },
    {
      "epoch": 2.339763534536403,
      "grad_norm": 3.3943352699279785,
      "learning_rate": 1.256918819188192e-06,
      "loss": 1.3958,
      "step": 1880
    },
    {
      "epoch": 2.339763534536403,
      "eval_loss": 1.6206164360046387,
      "eval_runtime": 47.5352,
      "eval_samples_per_second": 21.037,
      "eval_steps_per_second": 0.884,
      "step": 1880
    },
    {
      "epoch": 2.341008089607965,
      "grad_norm": 3.475947141647339,
      "learning_rate": 1.2546125461254613e-06,
      "loss": 1.3491,
      "step": 1881
    },
    {
      "epoch": 2.342252644679527,
      "grad_norm": 3.3503031730651855,
      "learning_rate": 1.2523062730627309e-06,
      "loss": 1.3198,
      "step": 1882
    },
    {
      "epoch": 2.343497199751089,
      "grad_norm": 3.290644884109497,
      "learning_rate": 1.25e-06,
      "loss": 1.3237,
      "step": 1883
    },
    {
      "epoch": 2.344741754822651,
      "grad_norm": 3.2442216873168945,
      "learning_rate": 1.2476937269372695e-06,
      "loss": 1.3149,
      "step": 1884
    },
    {
      "epoch": 2.345986309894213,
      "grad_norm": 3.2475531101226807,
      "learning_rate": 1.245387453874539e-06,
      "loss": 1.3265,
      "step": 1885
    },
    {
      "epoch": 2.3472308649657747,
      "grad_norm": 3.16241192817688,
      "learning_rate": 1.2430811808118082e-06,
      "loss": 1.3237,
      "step": 1886
    },
    {
      "epoch": 2.3484754200373366,
      "grad_norm": 3.6679129600524902,
      "learning_rate": 1.2407749077490776e-06,
      "loss": 1.3583,
      "step": 1887
    },
    {
      "epoch": 2.3497199751088984,
      "grad_norm": 3.5134646892547607,
      "learning_rate": 1.238468634686347e-06,
      "loss": 1.3543,
      "step": 1888
    },
    {
      "epoch": 2.3509645301804607,
      "grad_norm": 3.2337441444396973,
      "learning_rate": 1.2361623616236164e-06,
      "loss": 1.3,
      "step": 1889
    },
    {
      "epoch": 2.3522090852520225,
      "grad_norm": 3.4864821434020996,
      "learning_rate": 1.2338560885608856e-06,
      "loss": 1.3786,
      "step": 1890
    },
    {
      "epoch": 2.3522090852520225,
      "eval_loss": 1.6194350719451904,
      "eval_runtime": 53.3692,
      "eval_samples_per_second": 18.737,
      "eval_steps_per_second": 0.787,
      "step": 1890
    },
    {
      "epoch": 2.3534536403235844,
      "grad_norm": 3.4141929149627686,
      "learning_rate": 1.231549815498155e-06,
      "loss": 1.3584,
      "step": 1891
    },
    {
      "epoch": 2.3546981953951462,
      "grad_norm": 3.3861517906188965,
      "learning_rate": 1.2292435424354245e-06,
      "loss": 1.371,
      "step": 1892
    },
    {
      "epoch": 2.355942750466708,
      "grad_norm": 3.345811128616333,
      "learning_rate": 1.226937269372694e-06,
      "loss": 1.326,
      "step": 1893
    },
    {
      "epoch": 2.35718730553827,
      "grad_norm": 3.3748316764831543,
      "learning_rate": 1.2246309963099633e-06,
      "loss": 1.3422,
      "step": 1894
    },
    {
      "epoch": 2.358431860609832,
      "grad_norm": 3.4072210788726807,
      "learning_rate": 1.2223247232472325e-06,
      "loss": 1.3706,
      "step": 1895
    },
    {
      "epoch": 2.359676415681394,
      "grad_norm": 3.3167407512664795,
      "learning_rate": 1.220018450184502e-06,
      "loss": 1.3553,
      "step": 1896
    },
    {
      "epoch": 2.360920970752956,
      "grad_norm": 3.3539159297943115,
      "learning_rate": 1.2177121771217714e-06,
      "loss": 1.3306,
      "step": 1897
    },
    {
      "epoch": 2.3621655258245178,
      "grad_norm": 3.4795944690704346,
      "learning_rate": 1.2154059040590408e-06,
      "loss": 1.3536,
      "step": 1898
    },
    {
      "epoch": 2.3634100808960796,
      "grad_norm": 3.425493001937866,
      "learning_rate": 1.21309963099631e-06,
      "loss": 1.321,
      "step": 1899
    },
    {
      "epoch": 2.3646546359676415,
      "grad_norm": 3.415886402130127,
      "learning_rate": 1.2107933579335794e-06,
      "loss": 1.3477,
      "step": 1900
    },
    {
      "epoch": 2.3646546359676415,
      "eval_loss": 1.619869589805603,
      "eval_runtime": 54.2118,
      "eval_samples_per_second": 18.446,
      "eval_steps_per_second": 0.775,
      "step": 1900
    },
    {
      "epoch": 2.3658991910392033,
      "grad_norm": 3.6500039100646973,
      "learning_rate": 1.2084870848708489e-06,
      "loss": 1.406,
      "step": 1901
    },
    {
      "epoch": 2.3671437461107656,
      "grad_norm": 3.332648992538452,
      "learning_rate": 1.2061808118081183e-06,
      "loss": 1.3288,
      "step": 1902
    },
    {
      "epoch": 2.3683883011823275,
      "grad_norm": 3.2978408336639404,
      "learning_rate": 1.2038745387453875e-06,
      "loss": 1.3087,
      "step": 1903
    },
    {
      "epoch": 2.3696328562538893,
      "grad_norm": 3.459388494491577,
      "learning_rate": 1.201568265682657e-06,
      "loss": 1.3754,
      "step": 1904
    },
    {
      "epoch": 2.370877411325451,
      "grad_norm": 3.4902889728546143,
      "learning_rate": 1.1992619926199263e-06,
      "loss": 1.3529,
      "step": 1905
    },
    {
      "epoch": 2.372121966397013,
      "grad_norm": 3.393232822418213,
      "learning_rate": 1.1969557195571958e-06,
      "loss": 1.3517,
      "step": 1906
    },
    {
      "epoch": 2.373366521468575,
      "grad_norm": 3.4551451206207275,
      "learning_rate": 1.1946494464944652e-06,
      "loss": 1.3372,
      "step": 1907
    },
    {
      "epoch": 2.3746110765401367,
      "grad_norm": 3.3197243213653564,
      "learning_rate": 1.1923431734317344e-06,
      "loss": 1.3169,
      "step": 1908
    },
    {
      "epoch": 2.375855631611699,
      "grad_norm": 3.3255650997161865,
      "learning_rate": 1.1900369003690038e-06,
      "loss": 1.3462,
      "step": 1909
    },
    {
      "epoch": 2.377100186683261,
      "grad_norm": 3.4901344776153564,
      "learning_rate": 1.1877306273062733e-06,
      "loss": 1.3296,
      "step": 1910
    },
    {
      "epoch": 2.377100186683261,
      "eval_loss": 1.6212565898895264,
      "eval_runtime": 43.2416,
      "eval_samples_per_second": 23.126,
      "eval_steps_per_second": 0.971,
      "step": 1910
    },
    {
      "epoch": 2.3783447417548227,
      "grad_norm": 3.489834785461426,
      "learning_rate": 1.1854243542435427e-06,
      "loss": 1.358,
      "step": 1911
    },
    {
      "epoch": 2.3795892968263845,
      "grad_norm": 3.513260841369629,
      "learning_rate": 1.1831180811808119e-06,
      "loss": 1.3688,
      "step": 1912
    },
    {
      "epoch": 2.3808338518979464,
      "grad_norm": 3.464785099029541,
      "learning_rate": 1.1808118081180813e-06,
      "loss": 1.3291,
      "step": 1913
    },
    {
      "epoch": 2.3820784069695082,
      "grad_norm": 3.2861804962158203,
      "learning_rate": 1.1785055350553507e-06,
      "loss": 1.3056,
      "step": 1914
    },
    {
      "epoch": 2.3833229620410705,
      "grad_norm": 3.435983419418335,
      "learning_rate": 1.1761992619926202e-06,
      "loss": 1.3551,
      "step": 1915
    },
    {
      "epoch": 2.3845675171126324,
      "grad_norm": 3.330296516418457,
      "learning_rate": 1.1738929889298894e-06,
      "loss": 1.3152,
      "step": 1916
    },
    {
      "epoch": 2.385812072184194,
      "grad_norm": 3.300424337387085,
      "learning_rate": 1.1715867158671588e-06,
      "loss": 1.3446,
      "step": 1917
    },
    {
      "epoch": 2.387056627255756,
      "grad_norm": 3.2503697872161865,
      "learning_rate": 1.1692804428044282e-06,
      "loss": 1.3378,
      "step": 1918
    },
    {
      "epoch": 2.388301182327318,
      "grad_norm": 3.330183982849121,
      "learning_rate": 1.1669741697416974e-06,
      "loss": 1.3308,
      "step": 1919
    },
    {
      "epoch": 2.3895457373988798,
      "grad_norm": 3.4580652713775635,
      "learning_rate": 1.1646678966789668e-06,
      "loss": 1.3309,
      "step": 1920
    },
    {
      "epoch": 2.3895457373988798,
      "eval_loss": 1.6166638135910034,
      "eval_runtime": 44.9035,
      "eval_samples_per_second": 22.27,
      "eval_steps_per_second": 0.935,
      "step": 1920
    },
    {
      "epoch": 2.3907902924704416,
      "grad_norm": 3.2584407329559326,
      "learning_rate": 1.1623616236162363e-06,
      "loss": 1.3084,
      "step": 1921
    },
    {
      "epoch": 2.392034847542004,
      "grad_norm": 3.272989511489868,
      "learning_rate": 1.1600553505535057e-06,
      "loss": 1.3238,
      "step": 1922
    },
    {
      "epoch": 2.3932794026135658,
      "grad_norm": 3.3732693195343018,
      "learning_rate": 1.157749077490775e-06,
      "loss": 1.3391,
      "step": 1923
    },
    {
      "epoch": 2.3945239576851276,
      "grad_norm": 3.3817999362945557,
      "learning_rate": 1.1554428044280443e-06,
      "loss": 1.3568,
      "step": 1924
    },
    {
      "epoch": 2.3957685127566895,
      "grad_norm": 3.285632610321045,
      "learning_rate": 1.1531365313653137e-06,
      "loss": 1.3158,
      "step": 1925
    },
    {
      "epoch": 2.3970130678282513,
      "grad_norm": 3.267397880554199,
      "learning_rate": 1.1508302583025832e-06,
      "loss": 1.2795,
      "step": 1926
    },
    {
      "epoch": 2.398257622899813,
      "grad_norm": 3.555927276611328,
      "learning_rate": 1.1485239852398524e-06,
      "loss": 1.3788,
      "step": 1927
    },
    {
      "epoch": 2.3995021779713754,
      "grad_norm": 3.4175591468811035,
      "learning_rate": 1.1462177121771218e-06,
      "loss": 1.3019,
      "step": 1928
    },
    {
      "epoch": 2.4007467330429373,
      "grad_norm": 3.4269983768463135,
      "learning_rate": 1.1439114391143912e-06,
      "loss": 1.3145,
      "step": 1929
    },
    {
      "epoch": 2.401991288114499,
      "grad_norm": 3.3499817848205566,
      "learning_rate": 1.1416051660516606e-06,
      "loss": 1.3206,
      "step": 1930
    },
    {
      "epoch": 2.401991288114499,
      "eval_loss": 1.6193190813064575,
      "eval_runtime": 44.8302,
      "eval_samples_per_second": 22.306,
      "eval_steps_per_second": 0.937,
      "step": 1930
    },
    {
      "epoch": 2.403235843186061,
      "grad_norm": 3.3621203899383545,
      "learning_rate": 1.1392988929889299e-06,
      "loss": 1.3276,
      "step": 1931
    },
    {
      "epoch": 2.404480398257623,
      "grad_norm": 3.2818424701690674,
      "learning_rate": 1.1369926199261993e-06,
      "loss": 1.3234,
      "step": 1932
    },
    {
      "epoch": 2.4057249533291847,
      "grad_norm": 3.2797811031341553,
      "learning_rate": 1.1346863468634687e-06,
      "loss": 1.3069,
      "step": 1933
    },
    {
      "epoch": 2.4069695084007465,
      "grad_norm": 3.3689663410186768,
      "learning_rate": 1.1323800738007381e-06,
      "loss": 1.314,
      "step": 1934
    },
    {
      "epoch": 2.408214063472309,
      "grad_norm": 3.4538166522979736,
      "learning_rate": 1.1300738007380073e-06,
      "loss": 1.3236,
      "step": 1935
    },
    {
      "epoch": 2.4094586185438707,
      "grad_norm": 3.3834025859832764,
      "learning_rate": 1.1277675276752768e-06,
      "loss": 1.3587,
      "step": 1936
    },
    {
      "epoch": 2.4107031736154325,
      "grad_norm": 3.3694424629211426,
      "learning_rate": 1.1254612546125462e-06,
      "loss": 1.3452,
      "step": 1937
    },
    {
      "epoch": 2.4119477286869944,
      "grad_norm": 3.471663236618042,
      "learning_rate": 1.1231549815498156e-06,
      "loss": 1.3622,
      "step": 1938
    },
    {
      "epoch": 2.413192283758556,
      "grad_norm": 3.505692958831787,
      "learning_rate": 1.1208487084870848e-06,
      "loss": 1.326,
      "step": 1939
    },
    {
      "epoch": 2.414436838830118,
      "grad_norm": 3.4514808654785156,
      "learning_rate": 1.1185424354243542e-06,
      "loss": 1.3603,
      "step": 1940
    },
    {
      "epoch": 2.414436838830118,
      "eval_loss": 1.6171061992645264,
      "eval_runtime": 44.7206,
      "eval_samples_per_second": 22.361,
      "eval_steps_per_second": 0.939,
      "step": 1940
    },
    {
      "epoch": 2.4156813939016804,
      "grad_norm": 3.5669631958007812,
      "learning_rate": 1.1162361623616237e-06,
      "loss": 1.3357,
      "step": 1941
    },
    {
      "epoch": 2.416925948973242,
      "grad_norm": 3.396930694580078,
      "learning_rate": 1.113929889298893e-06,
      "loss": 1.337,
      "step": 1942
    },
    {
      "epoch": 2.418170504044804,
      "grad_norm": 3.389120101928711,
      "learning_rate": 1.1116236162361623e-06,
      "loss": 1.319,
      "step": 1943
    },
    {
      "epoch": 2.419415059116366,
      "grad_norm": 3.613391399383545,
      "learning_rate": 1.1093173431734317e-06,
      "loss": 1.371,
      "step": 1944
    },
    {
      "epoch": 2.4206596141879277,
      "grad_norm": 3.441382646560669,
      "learning_rate": 1.1070110701107011e-06,
      "loss": 1.3522,
      "step": 1945
    },
    {
      "epoch": 2.4219041692594896,
      "grad_norm": 3.3786375522613525,
      "learning_rate": 1.1047047970479706e-06,
      "loss": 1.3468,
      "step": 1946
    },
    {
      "epoch": 2.4231487243310514,
      "grad_norm": 3.432734489440918,
      "learning_rate": 1.10239852398524e-06,
      "loss": 1.3571,
      "step": 1947
    },
    {
      "epoch": 2.4243932794026137,
      "grad_norm": 3.621337413787842,
      "learning_rate": 1.1000922509225092e-06,
      "loss": 1.3591,
      "step": 1948
    },
    {
      "epoch": 2.4256378344741756,
      "grad_norm": 3.4495420455932617,
      "learning_rate": 1.0977859778597786e-06,
      "loss": 1.375,
      "step": 1949
    },
    {
      "epoch": 2.4268823895457374,
      "grad_norm": 3.4852328300476074,
      "learning_rate": 1.095479704797048e-06,
      "loss": 1.3391,
      "step": 1950
    },
    {
      "epoch": 2.4268823895457374,
      "eval_loss": 1.6171005964279175,
      "eval_runtime": 51.922,
      "eval_samples_per_second": 19.26,
      "eval_steps_per_second": 0.809,
      "step": 1950
    },
    {
      "epoch": 2.4281269446172993,
      "grad_norm": 3.455693006515503,
      "learning_rate": 1.0931734317343175e-06,
      "loss": 1.3703,
      "step": 1951
    },
    {
      "epoch": 2.429371499688861,
      "grad_norm": 3.2051644325256348,
      "learning_rate": 1.0908671586715867e-06,
      "loss": 1.2962,
      "step": 1952
    },
    {
      "epoch": 2.430616054760423,
      "grad_norm": 3.384723424911499,
      "learning_rate": 1.088560885608856e-06,
      "loss": 1.3794,
      "step": 1953
    },
    {
      "epoch": 2.4318606098319853,
      "grad_norm": 3.3716330528259277,
      "learning_rate": 1.0862546125461255e-06,
      "loss": 1.3045,
      "step": 1954
    },
    {
      "epoch": 2.433105164903547,
      "grad_norm": 3.48067569732666,
      "learning_rate": 1.083948339483395e-06,
      "loss": 1.2728,
      "step": 1955
    },
    {
      "epoch": 2.434349719975109,
      "grad_norm": 3.200029134750366,
      "learning_rate": 1.0816420664206644e-06,
      "loss": 1.353,
      "step": 1956
    },
    {
      "epoch": 2.435594275046671,
      "grad_norm": 3.413900375366211,
      "learning_rate": 1.0793357933579336e-06,
      "loss": 1.3946,
      "step": 1957
    },
    {
      "epoch": 2.4368388301182327,
      "grad_norm": 3.2179441452026367,
      "learning_rate": 1.077029520295203e-06,
      "loss": 1.3164,
      "step": 1958
    },
    {
      "epoch": 2.4380833851897945,
      "grad_norm": 3.321248769760132,
      "learning_rate": 1.0747232472324724e-06,
      "loss": 1.2627,
      "step": 1959
    },
    {
      "epoch": 2.4393279402613564,
      "grad_norm": 3.2880983352661133,
      "learning_rate": 1.0724169741697419e-06,
      "loss": 1.3106,
      "step": 1960
    },
    {
      "epoch": 2.4393279402613564,
      "eval_loss": 1.6168919801712036,
      "eval_runtime": 45.4429,
      "eval_samples_per_second": 22.006,
      "eval_steps_per_second": 0.924,
      "step": 1960
    },
    {
      "epoch": 2.4405724953329186,
      "grad_norm": 3.445110559463501,
      "learning_rate": 1.070110701107011e-06,
      "loss": 1.4013,
      "step": 1961
    },
    {
      "epoch": 2.4418170504044805,
      "grad_norm": 3.694018602371216,
      "learning_rate": 1.0678044280442805e-06,
      "loss": 1.3347,
      "step": 1962
    },
    {
      "epoch": 2.4430616054760423,
      "grad_norm": 3.2435543537139893,
      "learning_rate": 1.06549815498155e-06,
      "loss": 1.3277,
      "step": 1963
    },
    {
      "epoch": 2.444306160547604,
      "grad_norm": 3.271838903427124,
      "learning_rate": 1.0631918819188193e-06,
      "loss": 1.3149,
      "step": 1964
    },
    {
      "epoch": 2.445550715619166,
      "grad_norm": 3.4159295558929443,
      "learning_rate": 1.0608856088560888e-06,
      "loss": 1.3658,
      "step": 1965
    },
    {
      "epoch": 2.446795270690728,
      "grad_norm": 3.3939833641052246,
      "learning_rate": 1.058579335793358e-06,
      "loss": 1.3466,
      "step": 1966
    },
    {
      "epoch": 2.44803982576229,
      "grad_norm": 3.384528636932373,
      "learning_rate": 1.0562730627306274e-06,
      "loss": 1.3756,
      "step": 1967
    },
    {
      "epoch": 2.449284380833852,
      "grad_norm": 3.414613962173462,
      "learning_rate": 1.0539667896678968e-06,
      "loss": 1.3655,
      "step": 1968
    },
    {
      "epoch": 2.450528935905414,
      "grad_norm": 3.403575897216797,
      "learning_rate": 1.0516605166051662e-06,
      "loss": 1.3402,
      "step": 1969
    },
    {
      "epoch": 2.4517734909769757,
      "grad_norm": 3.4550907611846924,
      "learning_rate": 1.0493542435424354e-06,
      "loss": 1.3727,
      "step": 1970
    },
    {
      "epoch": 2.4517734909769757,
      "eval_loss": 1.6175874471664429,
      "eval_runtime": 46.6225,
      "eval_samples_per_second": 21.449,
      "eval_steps_per_second": 0.901,
      "step": 1970
    },
    {
      "epoch": 2.4530180460485376,
      "grad_norm": 3.4611523151397705,
      "learning_rate": 1.0470479704797049e-06,
      "loss": 1.3432,
      "step": 1971
    },
    {
      "epoch": 2.4542626011200994,
      "grad_norm": 3.4850761890411377,
      "learning_rate": 1.0447416974169743e-06,
      "loss": 1.3441,
      "step": 1972
    },
    {
      "epoch": 2.4555071561916613,
      "grad_norm": 3.49817156791687,
      "learning_rate": 1.0424354243542437e-06,
      "loss": 1.3137,
      "step": 1973
    },
    {
      "epoch": 2.4567517112632236,
      "grad_norm": 3.390441656112671,
      "learning_rate": 1.0401291512915131e-06,
      "loss": 1.2719,
      "step": 1974
    },
    {
      "epoch": 2.4579962663347854,
      "grad_norm": 3.43735933303833,
      "learning_rate": 1.0378228782287824e-06,
      "loss": 1.3539,
      "step": 1975
    },
    {
      "epoch": 2.4592408214063473,
      "grad_norm": 3.173717737197876,
      "learning_rate": 1.0355166051660518e-06,
      "loss": 1.2953,
      "step": 1976
    },
    {
      "epoch": 2.460485376477909,
      "grad_norm": 3.313398599624634,
      "learning_rate": 1.0332103321033212e-06,
      "loss": 1.2967,
      "step": 1977
    },
    {
      "epoch": 2.461729931549471,
      "grad_norm": 3.2390615940093994,
      "learning_rate": 1.0309040590405906e-06,
      "loss": 1.3208,
      "step": 1978
    },
    {
      "epoch": 2.462974486621033,
      "grad_norm": 3.2934212684631348,
      "learning_rate": 1.0285977859778598e-06,
      "loss": 1.3129,
      "step": 1979
    },
    {
      "epoch": 2.464219041692595,
      "grad_norm": 3.342038631439209,
      "learning_rate": 1.0262915129151293e-06,
      "loss": 1.3438,
      "step": 1980
    },
    {
      "epoch": 2.464219041692595,
      "eval_loss": 1.616586685180664,
      "eval_runtime": 44.6397,
      "eval_samples_per_second": 22.402,
      "eval_steps_per_second": 0.941,
      "step": 1980
    },
    {
      "epoch": 2.465463596764157,
      "grad_norm": 3.3280296325683594,
      "learning_rate": 1.0239852398523987e-06,
      "loss": 1.3251,
      "step": 1981
    },
    {
      "epoch": 2.466708151835719,
      "grad_norm": 3.2722764015197754,
      "learning_rate": 1.021678966789668e-06,
      "loss": 1.3422,
      "step": 1982
    },
    {
      "epoch": 2.4679527069072806,
      "grad_norm": 3.3998475074768066,
      "learning_rate": 1.0193726937269373e-06,
      "loss": 1.3285,
      "step": 1983
    },
    {
      "epoch": 2.4691972619788425,
      "grad_norm": 3.4187684059143066,
      "learning_rate": 1.0170664206642067e-06,
      "loss": 1.398,
      "step": 1984
    },
    {
      "epoch": 2.4704418170504043,
      "grad_norm": 3.324049472808838,
      "learning_rate": 1.0147601476014762e-06,
      "loss": 1.3178,
      "step": 1985
    },
    {
      "epoch": 2.471686372121966,
      "grad_norm": 3.333989143371582,
      "learning_rate": 1.0124538745387456e-06,
      "loss": 1.265,
      "step": 1986
    },
    {
      "epoch": 2.4729309271935285,
      "grad_norm": 3.5451438426971436,
      "learning_rate": 1.010147601476015e-06,
      "loss": 1.3259,
      "step": 1987
    },
    {
      "epoch": 2.4741754822650903,
      "grad_norm": 3.5002248287200928,
      "learning_rate": 1.0078413284132842e-06,
      "loss": 1.3452,
      "step": 1988
    },
    {
      "epoch": 2.475420037336652,
      "grad_norm": 3.3579916954040527,
      "learning_rate": 1.0055350553505536e-06,
      "loss": 1.2945,
      "step": 1989
    },
    {
      "epoch": 2.476664592408214,
      "grad_norm": 3.612515687942505,
      "learning_rate": 1.003228782287823e-06,
      "loss": 1.3483,
      "step": 1990
    },
    {
      "epoch": 2.476664592408214,
      "eval_loss": 1.6154413223266602,
      "eval_runtime": 45.8894,
      "eval_samples_per_second": 21.792,
      "eval_steps_per_second": 0.915,
      "step": 1990
    },
    {
      "epoch": 2.477909147479776,
      "grad_norm": 3.379666805267334,
      "learning_rate": 1.0009225092250923e-06,
      "loss": 1.329,
      "step": 1991
    },
    {
      "epoch": 2.4791537025513377,
      "grad_norm": 3.4102606773376465,
      "learning_rate": 9.986162361623617e-07,
      "loss": 1.3485,
      "step": 1992
    },
    {
      "epoch": 2.4803982576229,
      "grad_norm": 3.489799737930298,
      "learning_rate": 9.963099630996311e-07,
      "loss": 1.3735,
      "step": 1993
    },
    {
      "epoch": 2.481642812694462,
      "grad_norm": 3.348045825958252,
      "learning_rate": 9.940036900369005e-07,
      "loss": 1.3586,
      "step": 1994
    },
    {
      "epoch": 2.4828873677660237,
      "grad_norm": 3.4739747047424316,
      "learning_rate": 9.916974169741698e-07,
      "loss": 1.3584,
      "step": 1995
    },
    {
      "epoch": 2.4841319228375855,
      "grad_norm": 3.338620185852051,
      "learning_rate": 9.893911439114392e-07,
      "loss": 1.3364,
      "step": 1996
    },
    {
      "epoch": 2.4853764779091474,
      "grad_norm": 3.3837015628814697,
      "learning_rate": 9.870848708487086e-07,
      "loss": 1.3663,
      "step": 1997
    },
    {
      "epoch": 2.4866210329807092,
      "grad_norm": 3.451869010925293,
      "learning_rate": 9.84778597785978e-07,
      "loss": 1.349,
      "step": 1998
    },
    {
      "epoch": 2.487865588052271,
      "grad_norm": 3.2855687141418457,
      "learning_rate": 9.824723247232472e-07,
      "loss": 1.3403,
      "step": 1999
    },
    {
      "epoch": 2.4891101431238334,
      "grad_norm": 3.3178985118865967,
      "learning_rate": 9.801660516605167e-07,
      "loss": 1.3468,
      "step": 2000
    },
    {
      "epoch": 2.4891101431238334,
      "eval_loss": 1.6164495944976807,
      "eval_runtime": 45.108,
      "eval_samples_per_second": 22.169,
      "eval_steps_per_second": 0.931,
      "step": 2000
    }
  ],
  "logging_steps": 1,
  "max_steps": 2409,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "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": 6.23461881872384e+17,
  "train_batch_size": 3,
  "trial_name": null,
  "trial_params": null
}