{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.9975440032746623,
  "eval_steps": 500,
  "global_step": 1220,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "grad_norm": 0.3300960063934326,
      "learning_rate": 2.9999999999999997e-05,
      "loss": 0.9966,
      "step": 1
    },
    {
      "epoch": 0.0,
      "grad_norm": 0.4113194942474365,
      "learning_rate": 5.9999999999999995e-05,
      "loss": 1.1253,
      "step": 2
    },
    {
      "epoch": 0.0,
      "grad_norm": 0.2486647665500641,
      "learning_rate": 8.999999999999999e-05,
      "loss": 1.0721,
      "step": 3
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.2249160259962082,
      "learning_rate": 0.00011999999999999999,
      "loss": 0.9033,
      "step": 4
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.3706735074520111,
      "learning_rate": 0.00015,
      "loss": 1.0498,
      "step": 5
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.28104931116104126,
      "learning_rate": 0.00017999999999999998,
      "loss": 0.9108,
      "step": 6
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.27497801184654236,
      "learning_rate": 0.00020999999999999998,
      "loss": 0.9038,
      "step": 7
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.30283215641975403,
      "learning_rate": 0.00023999999999999998,
      "loss": 0.8605,
      "step": 8
    },
    {
      "epoch": 0.01,
      "grad_norm": 0.33457252383232117,
      "learning_rate": 0.00027,
      "loss": 0.9049,
      "step": 9
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.37725692987442017,
      "learning_rate": 0.0003,
      "loss": 0.772,
      "step": 10
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.27986466884613037,
      "learning_rate": 0.00029975206611570246,
      "loss": 0.7666,
      "step": 11
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.30687034130096436,
      "learning_rate": 0.00029950413223140494,
      "loss": 0.8312,
      "step": 12
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.3321741819381714,
      "learning_rate": 0.0002992561983471074,
      "loss": 0.8308,
      "step": 13
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.29080134630203247,
      "learning_rate": 0.0002990082644628099,
      "loss": 0.7597,
      "step": 14
    },
    {
      "epoch": 0.02,
      "grad_norm": 0.33823856711387634,
      "learning_rate": 0.0002987603305785124,
      "loss": 0.8693,
      "step": 15
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.3461182117462158,
      "learning_rate": 0.0002985123966942149,
      "loss": 1.0571,
      "step": 16
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.22306275367736816,
      "learning_rate": 0.0002982644628099173,
      "loss": 0.7706,
      "step": 17
    },
    {
      "epoch": 0.03,
      "grad_norm": 154.4940643310547,
      "learning_rate": 0.0002980165289256198,
      "loss": 2.6519,
      "step": 18
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.22956405580043793,
      "learning_rate": 0.00029776859504132227,
      "loss": 0.6897,
      "step": 19
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.25711989402770996,
      "learning_rate": 0.00029752066115702476,
      "loss": 0.7338,
      "step": 20
    },
    {
      "epoch": 0.03,
      "grad_norm": 0.2565441131591797,
      "learning_rate": 0.00029727272727272724,
      "loss": 0.8211,
      "step": 21
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.2437434047460556,
      "learning_rate": 0.0002970247933884297,
      "loss": 0.8027,
      "step": 22
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.21284469962120056,
      "learning_rate": 0.0002967768595041322,
      "loss": 0.7944,
      "step": 23
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.23338356614112854,
      "learning_rate": 0.0002965289256198347,
      "loss": 0.7696,
      "step": 24
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.25512659549713135,
      "learning_rate": 0.0002962809917355372,
      "loss": 0.7693,
      "step": 25
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.19500921666622162,
      "learning_rate": 0.0002960330578512396,
      "loss": 0.7599,
      "step": 26
    },
    {
      "epoch": 0.04,
      "grad_norm": 0.2554054260253906,
      "learning_rate": 0.00029578512396694214,
      "loss": 0.966,
      "step": 27
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.17682747542858124,
      "learning_rate": 0.0002955371900826446,
      "loss": 0.676,
      "step": 28
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.20516635477542877,
      "learning_rate": 0.0002952892561983471,
      "loss": 0.8144,
      "step": 29
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.3275119662284851,
      "learning_rate": 0.0002950413223140496,
      "loss": 0.7704,
      "step": 30
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.22231778502464294,
      "learning_rate": 0.000294793388429752,
      "loss": 0.7614,
      "step": 31
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.17065812647342682,
      "learning_rate": 0.0002945454545454545,
      "loss": 0.5634,
      "step": 32
    },
    {
      "epoch": 0.05,
      "grad_norm": 0.1771956831216812,
      "learning_rate": 0.000294297520661157,
      "loss": 0.7607,
      "step": 33
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.26693442463874817,
      "learning_rate": 0.00029404958677685947,
      "loss": 0.8171,
      "step": 34
    },
    {
      "epoch": 0.06,
      "grad_norm": 1.409070611000061,
      "learning_rate": 0.00029380165289256196,
      "loss": 0.7791,
      "step": 35
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.20727217197418213,
      "learning_rate": 0.00029355371900826444,
      "loss": 0.7357,
      "step": 36
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.2145707905292511,
      "learning_rate": 0.0002933057851239669,
      "loss": 0.8458,
      "step": 37
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.2068527340888977,
      "learning_rate": 0.0002930578512396694,
      "loss": 0.78,
      "step": 38
    },
    {
      "epoch": 0.06,
      "grad_norm": 0.22432388365268707,
      "learning_rate": 0.00029280991735537184,
      "loss": 0.8523,
      "step": 39
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.19982610642910004,
      "learning_rate": 0.0002925619834710743,
      "loss": 0.7372,
      "step": 40
    },
    {
      "epoch": 0.07,
      "grad_norm": 6.248472213745117,
      "learning_rate": 0.00029231404958677686,
      "loss": 0.7399,
      "step": 41
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.2269737422466278,
      "learning_rate": 0.00029206611570247934,
      "loss": 0.7842,
      "step": 42
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.23117898404598236,
      "learning_rate": 0.0002918181818181818,
      "loss": 0.7111,
      "step": 43
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.22466522455215454,
      "learning_rate": 0.00029157024793388425,
      "loss": 0.8979,
      "step": 44
    },
    {
      "epoch": 0.07,
      "grad_norm": 0.20770332217216492,
      "learning_rate": 0.00029132231404958674,
      "loss": 0.774,
      "step": 45
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.2376495748758316,
      "learning_rate": 0.0002910743801652892,
      "loss": 0.7216,
      "step": 46
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.2470778226852417,
      "learning_rate": 0.0002908264462809917,
      "loss": 0.7369,
      "step": 47
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.23465900123119354,
      "learning_rate": 0.0002905785123966942,
      "loss": 0.7528,
      "step": 48
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.5718627572059631,
      "learning_rate": 0.00029033057851239667,
      "loss": 0.7535,
      "step": 49
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.21493370831012726,
      "learning_rate": 0.00029008264462809916,
      "loss": 0.8593,
      "step": 50
    },
    {
      "epoch": 0.08,
      "grad_norm": 0.21197210252285004,
      "learning_rate": 0.00028983471074380164,
      "loss": 0.8013,
      "step": 51
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.20836398005485535,
      "learning_rate": 0.0002895867768595041,
      "loss": 0.7905,
      "step": 52
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.2096678912639618,
      "learning_rate": 0.00028933884297520655,
      "loss": 0.6754,
      "step": 53
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.25898435711860657,
      "learning_rate": 0.00028909090909090904,
      "loss": 0.7725,
      "step": 54
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.23370735347270966,
      "learning_rate": 0.0002888429752066116,
      "loss": 0.7007,
      "step": 55
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.23006942868232727,
      "learning_rate": 0.00028859504132231406,
      "loss": 0.7534,
      "step": 56
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.20855402946472168,
      "learning_rate": 0.0002883471074380165,
      "loss": 0.9491,
      "step": 57
    },
    {
      "epoch": 0.09,
      "grad_norm": 0.24340493977069855,
      "learning_rate": 0.00028809917355371897,
      "loss": 0.8089,
      "step": 58
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.20169466733932495,
      "learning_rate": 0.00028785123966942145,
      "loss": 0.64,
      "step": 59
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.23272906243801117,
      "learning_rate": 0.00028760330578512394,
      "loss": 0.8456,
      "step": 60
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.1767100691795349,
      "learning_rate": 0.0002873553719008264,
      "loss": 0.6686,
      "step": 61
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.24511106312274933,
      "learning_rate": 0.0002871074380165289,
      "loss": 0.6998,
      "step": 62
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.22284479439258575,
      "learning_rate": 0.0002868595041322314,
      "loss": 0.6699,
      "step": 63
    },
    {
      "epoch": 0.1,
      "grad_norm": 0.21842750906944275,
      "learning_rate": 0.00028661157024793387,
      "loss": 0.7413,
      "step": 64
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.2669163644313812,
      "learning_rate": 0.00028636363636363636,
      "loss": 0.931,
      "step": 65
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.1864808052778244,
      "learning_rate": 0.0002861157024793388,
      "loss": 0.5652,
      "step": 66
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.18369853496551514,
      "learning_rate": 0.00028586776859504127,
      "loss": 0.6847,
      "step": 67
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.22353056073188782,
      "learning_rate": 0.00028561983471074375,
      "loss": 0.598,
      "step": 68
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.20269523561000824,
      "learning_rate": 0.0002853719008264463,
      "loss": 0.8688,
      "step": 69
    },
    {
      "epoch": 0.11,
      "grad_norm": 0.2291198968887329,
      "learning_rate": 0.0002851239669421488,
      "loss": 0.7535,
      "step": 70
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.22033120691776276,
      "learning_rate": 0.0002848760330578512,
      "loss": 0.8377,
      "step": 71
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.2687983214855194,
      "learning_rate": 0.0002846280991735537,
      "loss": 0.6926,
      "step": 72
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.1933681070804596,
      "learning_rate": 0.00028438016528925617,
      "loss": 0.6276,
      "step": 73
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.2820705473423004,
      "learning_rate": 0.00028413223140495865,
      "loss": 0.848,
      "step": 74
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.19532324373722076,
      "learning_rate": 0.00028388429752066114,
      "loss": 0.6198,
      "step": 75
    },
    {
      "epoch": 0.12,
      "grad_norm": 0.25057846307754517,
      "learning_rate": 0.0002836363636363636,
      "loss": 0.6838,
      "step": 76
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.2168462574481964,
      "learning_rate": 0.0002833884297520661,
      "loss": 0.7885,
      "step": 77
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.2106674313545227,
      "learning_rate": 0.0002831404958677686,
      "loss": 0.6757,
      "step": 78
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.24460363388061523,
      "learning_rate": 0.000282892561983471,
      "loss": 0.7414,
      "step": 79
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.3706071078777313,
      "learning_rate": 0.0002826446280991735,
      "loss": 0.621,
      "step": 80
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.2251998782157898,
      "learning_rate": 0.000282396694214876,
      "loss": 0.7453,
      "step": 81
    },
    {
      "epoch": 0.13,
      "grad_norm": 0.24521738290786743,
      "learning_rate": 0.00028214876033057847,
      "loss": 0.6985,
      "step": 82
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.2262742966413498,
      "learning_rate": 0.000281900826446281,
      "loss": 0.6316,
      "step": 83
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.19723354279994965,
      "learning_rate": 0.00028165289256198344,
      "loss": 0.4798,
      "step": 84
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.20684833824634552,
      "learning_rate": 0.0002814049586776859,
      "loss": 0.7993,
      "step": 85
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.19534814357757568,
      "learning_rate": 0.0002811570247933884,
      "loss": 0.7735,
      "step": 86
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.2585545480251312,
      "learning_rate": 0.0002809090909090909,
      "loss": 0.8126,
      "step": 87
    },
    {
      "epoch": 0.14,
      "grad_norm": 0.2510583996772766,
      "learning_rate": 0.00028066115702479337,
      "loss": 0.6973,
      "step": 88
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.1884051263332367,
      "learning_rate": 0.00028041322314049585,
      "loss": 0.701,
      "step": 89
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.2526257038116455,
      "learning_rate": 0.00028016528925619834,
      "loss": 0.7132,
      "step": 90
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.200734481215477,
      "learning_rate": 0.0002799173553719008,
      "loss": 0.7024,
      "step": 91
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.2404022514820099,
      "learning_rate": 0.0002796694214876033,
      "loss": 0.704,
      "step": 92
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.23063871264457703,
      "learning_rate": 0.00027942148760330573,
      "loss": 0.6312,
      "step": 93
    },
    {
      "epoch": 0.15,
      "grad_norm": 0.1759747564792633,
      "learning_rate": 0.0002791735537190082,
      "loss": 0.6577,
      "step": 94
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.2009582370519638,
      "learning_rate": 0.0002789256198347107,
      "loss": 0.8036,
      "step": 95
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.2200164943933487,
      "learning_rate": 0.0002786776859504132,
      "loss": 0.7101,
      "step": 96
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.19693537056446075,
      "learning_rate": 0.00027842975206611567,
      "loss": 0.6221,
      "step": 97
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.23269779980182648,
      "learning_rate": 0.00027818181818181815,
      "loss": 0.8264,
      "step": 98
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.2440226823091507,
      "learning_rate": 0.00027793388429752064,
      "loss": 0.8051,
      "step": 99
    },
    {
      "epoch": 0.16,
      "grad_norm": 0.2307034134864807,
      "learning_rate": 0.0002776859504132231,
      "loss": 0.631,
      "step": 100
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.2530567944049835,
      "learning_rate": 0.0002774380165289256,
      "loss": 0.8616,
      "step": 101
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.2808806300163269,
      "learning_rate": 0.0002771900826446281,
      "loss": 0.8333,
      "step": 102
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.20667941868305206,
      "learning_rate": 0.00027694214876033057,
      "loss": 0.7212,
      "step": 103
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.17540781199932098,
      "learning_rate": 0.00027669421487603305,
      "loss": 0.5964,
      "step": 104
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.2526637613773346,
      "learning_rate": 0.00027644628099173554,
      "loss": 0.6868,
      "step": 105
    },
    {
      "epoch": 0.17,
      "grad_norm": 0.2137339860200882,
      "learning_rate": 0.00027619834710743797,
      "loss": 0.6155,
      "step": 106
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.21061092615127563,
      "learning_rate": 0.00027595041322314045,
      "loss": 0.813,
      "step": 107
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.21619191765785217,
      "learning_rate": 0.00027570247933884293,
      "loss": 0.8046,
      "step": 108
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.2212170660495758,
      "learning_rate": 0.0002754545454545454,
      "loss": 0.6706,
      "step": 109
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.23427413403987885,
      "learning_rate": 0.0002752066115702479,
      "loss": 0.7152,
      "step": 110
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.20566123723983765,
      "learning_rate": 0.0002749586776859504,
      "loss": 0.6568,
      "step": 111
    },
    {
      "epoch": 0.18,
      "grad_norm": 0.22977930307388306,
      "learning_rate": 0.00027471074380165287,
      "loss": 0.7832,
      "step": 112
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.28307485580444336,
      "learning_rate": 0.00027446280991735535,
      "loss": 0.7446,
      "step": 113
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.19567596912384033,
      "learning_rate": 0.00027421487603305784,
      "loss": 0.6394,
      "step": 114
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.24577689170837402,
      "learning_rate": 0.0002739669421487603,
      "loss": 0.6389,
      "step": 115
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.2180463820695877,
      "learning_rate": 0.0002737190082644628,
      "loss": 0.7814,
      "step": 116
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.19546380639076233,
      "learning_rate": 0.0002734710743801653,
      "loss": 0.8312,
      "step": 117
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.22698360681533813,
      "learning_rate": 0.00027322314049586777,
      "loss": 0.7443,
      "step": 118
    },
    {
      "epoch": 0.19,
      "grad_norm": 0.22987066209316254,
      "learning_rate": 0.0002729752066115702,
      "loss": 0.7839,
      "step": 119
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.20548178255558014,
      "learning_rate": 0.0002727272727272727,
      "loss": 0.7805,
      "step": 120
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.2477702796459198,
      "learning_rate": 0.00027247933884297517,
      "loss": 0.5694,
      "step": 121
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.20593340694904327,
      "learning_rate": 0.00027223140495867765,
      "loss": 0.6479,
      "step": 122
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.23635917901992798,
      "learning_rate": 0.00027198347107438013,
      "loss": 0.8107,
      "step": 123
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.25808119773864746,
      "learning_rate": 0.0002717355371900826,
      "loss": 0.7876,
      "step": 124
    },
    {
      "epoch": 0.2,
      "grad_norm": 0.22156469523906708,
      "learning_rate": 0.0002714876033057851,
      "loss": 0.7261,
      "step": 125
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.19892215728759766,
      "learning_rate": 0.0002712396694214876,
      "loss": 0.6874,
      "step": 126
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.24936752021312714,
      "learning_rate": 0.00027099173553719007,
      "loss": 0.6155,
      "step": 127
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.23287539184093475,
      "learning_rate": 0.0002707438016528925,
      "loss": 0.602,
      "step": 128
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.2086639404296875,
      "learning_rate": 0.00027049586776859504,
      "loss": 0.7198,
      "step": 129
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.24974922835826874,
      "learning_rate": 0.0002702479338842975,
      "loss": 0.6873,
      "step": 130
    },
    {
      "epoch": 0.21,
      "grad_norm": 0.2066827118396759,
      "learning_rate": 0.00027,
      "loss": 0.5821,
      "step": 131
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.28004395961761475,
      "learning_rate": 0.0002697520661157025,
      "loss": 0.7864,
      "step": 132
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.22391608357429504,
      "learning_rate": 0.0002695041322314049,
      "loss": 0.6773,
      "step": 133
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.2821199297904968,
      "learning_rate": 0.0002692561983471074,
      "loss": 0.6806,
      "step": 134
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.21736428141593933,
      "learning_rate": 0.0002690082644628099,
      "loss": 0.6662,
      "step": 135
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.23889939486980438,
      "learning_rate": 0.00026876033057851237,
      "loss": 0.6356,
      "step": 136
    },
    {
      "epoch": 0.22,
      "grad_norm": 0.21096719801425934,
      "learning_rate": 0.00026851239669421485,
      "loss": 0.6762,
      "step": 137
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.22622421383857727,
      "learning_rate": 0.00026826446280991733,
      "loss": 0.8085,
      "step": 138
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.19824957847595215,
      "learning_rate": 0.0002680165289256198,
      "loss": 0.6031,
      "step": 139
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.24482691287994385,
      "learning_rate": 0.0002677685950413223,
      "loss": 0.6649,
      "step": 140
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.21291929483413696,
      "learning_rate": 0.0002675206611570248,
      "loss": 0.6671,
      "step": 141
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.2202674299478531,
      "learning_rate": 0.0002672727272727272,
      "loss": 0.6469,
      "step": 142
    },
    {
      "epoch": 0.23,
      "grad_norm": 0.23572632670402527,
      "learning_rate": 0.0002670247933884297,
      "loss": 0.7377,
      "step": 143
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.2051907777786255,
      "learning_rate": 0.00026677685950413224,
      "loss": 0.6217,
      "step": 144
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.23270072042942047,
      "learning_rate": 0.0002665289256198347,
      "loss": 0.7933,
      "step": 145
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.20652809739112854,
      "learning_rate": 0.00026628099173553715,
      "loss": 0.6007,
      "step": 146
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.23084674775600433,
      "learning_rate": 0.00026603305785123963,
      "loss": 0.701,
      "step": 147
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.25663891434669495,
      "learning_rate": 0.0002657851239669421,
      "loss": 0.7271,
      "step": 148
    },
    {
      "epoch": 0.24,
      "grad_norm": 0.25880497694015503,
      "learning_rate": 0.0002655371900826446,
      "loss": 0.6562,
      "step": 149
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.19349205493927002,
      "learning_rate": 0.0002652892561983471,
      "loss": 0.5016,
      "step": 150
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.2401740401983261,
      "learning_rate": 0.00026504132231404957,
      "loss": 0.6978,
      "step": 151
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.19495394825935364,
      "learning_rate": 0.00026479338842975205,
      "loss": 0.5562,
      "step": 152
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.21485286951065063,
      "learning_rate": 0.00026454545454545453,
      "loss": 0.7847,
      "step": 153
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.241348534822464,
      "learning_rate": 0.000264297520661157,
      "loss": 0.7513,
      "step": 154
    },
    {
      "epoch": 0.25,
      "grad_norm": 0.3316986858844757,
      "learning_rate": 0.00026404958677685945,
      "loss": 0.664,
      "step": 155
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.2419958859682083,
      "learning_rate": 0.00026380165289256193,
      "loss": 0.7322,
      "step": 156
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.2868640124797821,
      "learning_rate": 0.0002635537190082644,
      "loss": 0.7004,
      "step": 157
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.24806949496269226,
      "learning_rate": 0.00026330578512396695,
      "loss": 0.6497,
      "step": 158
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.23873400688171387,
      "learning_rate": 0.00026305785123966944,
      "loss": 0.7543,
      "step": 159
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.2480355203151703,
      "learning_rate": 0.00026280991735537187,
      "loss": 0.6048,
      "step": 160
    },
    {
      "epoch": 0.26,
      "grad_norm": 0.2619112730026245,
      "learning_rate": 0.00026256198347107435,
      "loss": 0.762,
      "step": 161
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.22763262689113617,
      "learning_rate": 0.00026231404958677683,
      "loss": 0.6557,
      "step": 162
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.3291528522968292,
      "learning_rate": 0.0002620661157024793,
      "loss": 0.7059,
      "step": 163
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.2959338426589966,
      "learning_rate": 0.0002618181818181818,
      "loss": 0.6622,
      "step": 164
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.23001112043857574,
      "learning_rate": 0.0002615702479338843,
      "loss": 0.6465,
      "step": 165
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.1998877376317978,
      "learning_rate": 0.00026132231404958677,
      "loss": 0.666,
      "step": 166
    },
    {
      "epoch": 0.27,
      "grad_norm": 0.23009613156318665,
      "learning_rate": 0.00026107438016528925,
      "loss": 0.8793,
      "step": 167
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.24525685608386993,
      "learning_rate": 0.0002608264462809917,
      "loss": 0.8009,
      "step": 168
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.21605077385902405,
      "learning_rate": 0.00026057851239669416,
      "loss": 0.5459,
      "step": 169
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.2576725482940674,
      "learning_rate": 0.00026033057851239665,
      "loss": 0.6818,
      "step": 170
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.23385170102119446,
      "learning_rate": 0.00026008264462809913,
      "loss": 0.7559,
      "step": 171
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.1973017454147339,
      "learning_rate": 0.00025983471074380167,
      "loss": 0.6798,
      "step": 172
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.22262559831142426,
      "learning_rate": 0.0002595867768595041,
      "loss": 0.5566,
      "step": 173
    },
    {
      "epoch": 0.28,
      "grad_norm": 0.23010462522506714,
      "learning_rate": 0.0002593388429752066,
      "loss": 0.7101,
      "step": 174
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.21676452457904816,
      "learning_rate": 0.00025909090909090907,
      "loss": 0.7038,
      "step": 175
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.22475261986255646,
      "learning_rate": 0.00025884297520661155,
      "loss": 0.7812,
      "step": 176
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.28893202543258667,
      "learning_rate": 0.00025859504132231403,
      "loss": 0.5925,
      "step": 177
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.22777552902698517,
      "learning_rate": 0.0002583471074380165,
      "loss": 0.7319,
      "step": 178
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.2287953644990921,
      "learning_rate": 0.000258099173553719,
      "loss": 0.7775,
      "step": 179
    },
    {
      "epoch": 0.29,
      "grad_norm": 0.2049843668937683,
      "learning_rate": 0.0002578512396694215,
      "loss": 0.7448,
      "step": 180
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.22585280239582062,
      "learning_rate": 0.00025760330578512397,
      "loss": 0.59,
      "step": 181
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.23159150779247284,
      "learning_rate": 0.0002573553719008264,
      "loss": 0.737,
      "step": 182
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.3393082320690155,
      "learning_rate": 0.0002571074380165289,
      "loss": 0.6948,
      "step": 183
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.2345617413520813,
      "learning_rate": 0.00025685950413223136,
      "loss": 0.6351,
      "step": 184
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.23474591970443726,
      "learning_rate": 0.00025661157024793385,
      "loss": 0.6643,
      "step": 185
    },
    {
      "epoch": 0.3,
      "grad_norm": 0.2473030984401703,
      "learning_rate": 0.00025636363636363633,
      "loss": 0.7663,
      "step": 186
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.2971685230731964,
      "learning_rate": 0.0002561157024793388,
      "loss": 0.7449,
      "step": 187
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.2745087742805481,
      "learning_rate": 0.0002558677685950413,
      "loss": 0.6125,
      "step": 188
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.23520545661449432,
      "learning_rate": 0.0002556198347107438,
      "loss": 0.573,
      "step": 189
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.2955464720726013,
      "learning_rate": 0.00025537190082644627,
      "loss": 0.5315,
      "step": 190
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.23987281322479248,
      "learning_rate": 0.00025512396694214875,
      "loss": 0.5636,
      "step": 191
    },
    {
      "epoch": 0.31,
      "grad_norm": 0.24263744056224823,
      "learning_rate": 0.00025487603305785123,
      "loss": 0.6047,
      "step": 192
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.26061922311782837,
      "learning_rate": 0.0002546280991735537,
      "loss": 0.7812,
      "step": 193
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.2458687126636505,
      "learning_rate": 0.0002543801652892562,
      "loss": 0.58,
      "step": 194
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.24598994851112366,
      "learning_rate": 0.00025413223140495863,
      "loss": 0.7432,
      "step": 195
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.248992919921875,
      "learning_rate": 0.0002538842975206611,
      "loss": 0.6953,
      "step": 196
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.2518531382083893,
      "learning_rate": 0.0002536363636363636,
      "loss": 0.6707,
      "step": 197
    },
    {
      "epoch": 0.32,
      "grad_norm": 0.23844210803508759,
      "learning_rate": 0.0002533884297520661,
      "loss": 0.6285,
      "step": 198
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.21948237717151642,
      "learning_rate": 0.00025314049586776856,
      "loss": 0.6859,
      "step": 199
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.2003835141658783,
      "learning_rate": 0.00025289256198347105,
      "loss": 0.6305,
      "step": 200
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.23421582579612732,
      "learning_rate": 0.00025264462809917353,
      "loss": 0.7164,
      "step": 201
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.22344104945659637,
      "learning_rate": 0.000252396694214876,
      "loss": 0.6498,
      "step": 202
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.17792212963104248,
      "learning_rate": 0.0002521487603305785,
      "loss": 0.614,
      "step": 203
    },
    {
      "epoch": 0.33,
      "grad_norm": 0.217886820435524,
      "learning_rate": 0.000251900826446281,
      "loss": 0.7033,
      "step": 204
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.209726020693779,
      "learning_rate": 0.00025165289256198347,
      "loss": 0.5913,
      "step": 205
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.2401910424232483,
      "learning_rate": 0.00025140495867768595,
      "loss": 0.6405,
      "step": 206
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.21315626800060272,
      "learning_rate": 0.00025115702479338843,
      "loss": 0.7369,
      "step": 207
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.20102320611476898,
      "learning_rate": 0.00025090909090909086,
      "loss": 0.6245,
      "step": 208
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.20447981357574463,
      "learning_rate": 0.00025066115702479335,
      "loss": 0.5423,
      "step": 209
    },
    {
      "epoch": 0.34,
      "grad_norm": 0.24979281425476074,
      "learning_rate": 0.00025041322314049583,
      "loss": 0.8078,
      "step": 210
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.20141547918319702,
      "learning_rate": 0.0002501652892561983,
      "loss": 0.7386,
      "step": 211
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.2538990378379822,
      "learning_rate": 0.0002499173553719008,
      "loss": 0.7219,
      "step": 212
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.2613961100578308,
      "learning_rate": 0.0002496694214876033,
      "loss": 0.7903,
      "step": 213
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.24777857959270477,
      "learning_rate": 0.00024942148760330576,
      "loss": 0.664,
      "step": 214
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.21958425641059875,
      "learning_rate": 0.00024917355371900825,
      "loss": 0.6755,
      "step": 215
    },
    {
      "epoch": 0.35,
      "grad_norm": 0.2218528538942337,
      "learning_rate": 0.00024892561983471073,
      "loss": 0.5568,
      "step": 216
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.23632755875587463,
      "learning_rate": 0.00024867768595041316,
      "loss": 0.6858,
      "step": 217
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.2641279697418213,
      "learning_rate": 0.0002484297520661157,
      "loss": 0.7783,
      "step": 218
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.3147680163383484,
      "learning_rate": 0.0002481818181818182,
      "loss": 0.662,
      "step": 219
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.27947697043418884,
      "learning_rate": 0.00024793388429752067,
      "loss": 0.6477,
      "step": 220
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.2297278195619583,
      "learning_rate": 0.00024768595041322315,
      "loss": 0.5895,
      "step": 221
    },
    {
      "epoch": 0.36,
      "grad_norm": 0.23085851967334747,
      "learning_rate": 0.0002474380165289256,
      "loss": 0.5806,
      "step": 222
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.19654251635074615,
      "learning_rate": 0.00024719008264462806,
      "loss": 0.5942,
      "step": 223
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.2467166632413864,
      "learning_rate": 0.00024694214876033055,
      "loss": 0.5059,
      "step": 224
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.22614917159080505,
      "learning_rate": 0.00024669421487603303,
      "loss": 0.643,
      "step": 225
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.2622920274734497,
      "learning_rate": 0.0002464462809917355,
      "loss": 0.6257,
      "step": 226
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.21843163669109344,
      "learning_rate": 0.000246198347107438,
      "loss": 0.6057,
      "step": 227
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.2294640988111496,
      "learning_rate": 0.0002459504132231405,
      "loss": 0.6876,
      "step": 228
    },
    {
      "epoch": 0.37,
      "grad_norm": 0.1791463941335678,
      "learning_rate": 0.00024570247933884296,
      "loss": 0.5348,
      "step": 229
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.17243699729442596,
      "learning_rate": 0.00024545454545454545,
      "loss": 0.5966,
      "step": 230
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.22769273817539215,
      "learning_rate": 0.0002452066115702479,
      "loss": 0.7912,
      "step": 231
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.2325255423784256,
      "learning_rate": 0.0002449586776859504,
      "loss": 0.7441,
      "step": 232
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.24277740716934204,
      "learning_rate": 0.0002447107438016529,
      "loss": 0.6653,
      "step": 233
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.21596141159534454,
      "learning_rate": 0.0002444628099173554,
      "loss": 0.6668,
      "step": 234
    },
    {
      "epoch": 0.38,
      "grad_norm": 0.20814135670661926,
      "learning_rate": 0.0002442148760330578,
      "loss": 0.6306,
      "step": 235
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.25570017099380493,
      "learning_rate": 0.0002439669421487603,
      "loss": 0.6524,
      "step": 236
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.2502390146255493,
      "learning_rate": 0.00024371900826446278,
      "loss": 0.6048,
      "step": 237
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.23688243329524994,
      "learning_rate": 0.0002434710743801653,
      "loss": 0.568,
      "step": 238
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.21041709184646606,
      "learning_rate": 0.00024322314049586777,
      "loss": 0.6908,
      "step": 239
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.21656759083271027,
      "learning_rate": 0.00024297520661157023,
      "loss": 0.4993,
      "step": 240
    },
    {
      "epoch": 0.39,
      "grad_norm": 0.25133028626441956,
      "learning_rate": 0.0002427272727272727,
      "loss": 0.718,
      "step": 241
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.22228790819644928,
      "learning_rate": 0.0002424793388429752,
      "loss": 0.6146,
      "step": 242
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.26273205876350403,
      "learning_rate": 0.00024223140495867768,
      "loss": 0.7459,
      "step": 243
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.2156606763601303,
      "learning_rate": 0.00024198347107438014,
      "loss": 0.6692,
      "step": 244
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.2075020670890808,
      "learning_rate": 0.00024173553719008262,
      "loss": 0.6427,
      "step": 245
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.25821176171302795,
      "learning_rate": 0.0002414876033057851,
      "loss": 0.7964,
      "step": 246
    },
    {
      "epoch": 0.4,
      "grad_norm": 0.23016126453876495,
      "learning_rate": 0.0002412396694214876,
      "loss": 0.536,
      "step": 247
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.23115016520023346,
      "learning_rate": 0.00024099173553719004,
      "loss": 0.6053,
      "step": 248
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.18249157071113586,
      "learning_rate": 0.00024074380165289253,
      "loss": 0.6574,
      "step": 249
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.28391778469085693,
      "learning_rate": 0.000240495867768595,
      "loss": 0.7152,
      "step": 250
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.2581539452075958,
      "learning_rate": 0.0002402479338842975,
      "loss": 0.8476,
      "step": 251
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.2304867058992386,
      "learning_rate": 0.00023999999999999998,
      "loss": 0.5781,
      "step": 252
    },
    {
      "epoch": 0.41,
      "grad_norm": 0.239717036485672,
      "learning_rate": 0.00023975206611570244,
      "loss": 0.6543,
      "step": 253
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.22493794560432434,
      "learning_rate": 0.00023950413223140495,
      "loss": 0.7048,
      "step": 254
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.22085991501808167,
      "learning_rate": 0.00023925619834710743,
      "loss": 0.5572,
      "step": 255
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.35917988419532776,
      "learning_rate": 0.0002390082644628099,
      "loss": 0.8485,
      "step": 256
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.28269943594932556,
      "learning_rate": 0.00023876033057851237,
      "loss": 0.5732,
      "step": 257
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.26313093304634094,
      "learning_rate": 0.00023851239669421485,
      "loss": 0.8212,
      "step": 258
    },
    {
      "epoch": 0.42,
      "grad_norm": 0.30286532640457153,
      "learning_rate": 0.00023826446280991734,
      "loss": 0.5878,
      "step": 259
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.22270837426185608,
      "learning_rate": 0.00023801652892561982,
      "loss": 0.6933,
      "step": 260
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.29011014103889465,
      "learning_rate": 0.0002377685950413223,
      "loss": 0.6188,
      "step": 261
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.2390982061624527,
      "learning_rate": 0.00023752066115702476,
      "loss": 0.6426,
      "step": 262
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.3416346609592438,
      "learning_rate": 0.00023727272727272724,
      "loss": 0.8845,
      "step": 263
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.25051388144493103,
      "learning_rate": 0.00023702479338842973,
      "loss": 0.7286,
      "step": 264
    },
    {
      "epoch": 0.43,
      "grad_norm": 0.2497546523809433,
      "learning_rate": 0.0002367768595041322,
      "loss": 0.6027,
      "step": 265
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.23835037648677826,
      "learning_rate": 0.00023652892561983467,
      "loss": 0.7052,
      "step": 266
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.22467398643493652,
      "learning_rate": 0.00023628099173553715,
      "loss": 0.5806,
      "step": 267
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.2663390338420868,
      "learning_rate": 0.00023603305785123964,
      "loss": 0.6943,
      "step": 268
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.22997191548347473,
      "learning_rate": 0.00023578512396694215,
      "loss": 0.6411,
      "step": 269
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.23266558349132538,
      "learning_rate": 0.00023553719008264463,
      "loss": 0.6068,
      "step": 270
    },
    {
      "epoch": 0.44,
      "grad_norm": 0.2304474264383316,
      "learning_rate": 0.00023528925619834709,
      "loss": 0.6427,
      "step": 271
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.28231826424598694,
      "learning_rate": 0.00023504132231404957,
      "loss": 0.8011,
      "step": 272
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.28013259172439575,
      "learning_rate": 0.00023479338842975205,
      "loss": 0.5988,
      "step": 273
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.22702372074127197,
      "learning_rate": 0.00023454545454545454,
      "loss": 0.6737,
      "step": 274
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.27958643436431885,
      "learning_rate": 0.000234297520661157,
      "loss": 0.6621,
      "step": 275
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.23902451992034912,
      "learning_rate": 0.00023404958677685948,
      "loss": 0.6525,
      "step": 276
    },
    {
      "epoch": 0.45,
      "grad_norm": 0.2778523564338684,
      "learning_rate": 0.00023380165289256196,
      "loss": 0.6697,
      "step": 277
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.2382276952266693,
      "learning_rate": 0.00023355371900826444,
      "loss": 0.6281,
      "step": 278
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.24487091600894928,
      "learning_rate": 0.00023330578512396693,
      "loss": 0.6842,
      "step": 279
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.2063397765159607,
      "learning_rate": 0.00023305785123966938,
      "loss": 0.6554,
      "step": 280
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.21523278951644897,
      "learning_rate": 0.00023280991735537187,
      "loss": 0.632,
      "step": 281
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.2420080006122589,
      "learning_rate": 0.00023256198347107435,
      "loss": 0.6001,
      "step": 282
    },
    {
      "epoch": 0.46,
      "grad_norm": 0.2390110194683075,
      "learning_rate": 0.00023231404958677686,
      "loss": 0.5648,
      "step": 283
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.24080687761306763,
      "learning_rate": 0.0002320661157024793,
      "loss": 0.86,
      "step": 284
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.29456445574760437,
      "learning_rate": 0.0002318181818181818,
      "loss": 0.7418,
      "step": 285
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.23326683044433594,
      "learning_rate": 0.00023157024793388429,
      "loss": 0.6967,
      "step": 286
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.20866093039512634,
      "learning_rate": 0.00023132231404958677,
      "loss": 0.5205,
      "step": 287
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.3158474266529083,
      "learning_rate": 0.00023107438016528925,
      "loss": 0.7879,
      "step": 288
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.2730140686035156,
      "learning_rate": 0.0002308264462809917,
      "loss": 0.7292,
      "step": 289
    },
    {
      "epoch": 0.47,
      "grad_norm": 0.25384965538978577,
      "learning_rate": 0.0002305785123966942,
      "loss": 0.7258,
      "step": 290
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.20765069127082825,
      "learning_rate": 0.00023033057851239668,
      "loss": 0.7108,
      "step": 291
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.25662195682525635,
      "learning_rate": 0.00023008264462809916,
      "loss": 0.7473,
      "step": 292
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.300243616104126,
      "learning_rate": 0.00022983471074380162,
      "loss": 0.6902,
      "step": 293
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.23513919115066528,
      "learning_rate": 0.0002295867768595041,
      "loss": 0.5888,
      "step": 294
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.2077571451663971,
      "learning_rate": 0.00022933884297520658,
      "loss": 0.6256,
      "step": 295
    },
    {
      "epoch": 0.48,
      "grad_norm": 0.266201376914978,
      "learning_rate": 0.00022909090909090907,
      "loss": 0.6913,
      "step": 296
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.2239614725112915,
      "learning_rate": 0.00022884297520661152,
      "loss": 0.7369,
      "step": 297
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.21509824693202972,
      "learning_rate": 0.000228595041322314,
      "loss": 0.4445,
      "step": 298
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.21956239640712738,
      "learning_rate": 0.00022834710743801652,
      "loss": 0.6732,
      "step": 299
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.18832357227802277,
      "learning_rate": 0.000228099173553719,
      "loss": 0.6808,
      "step": 300
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.21115505695343018,
      "learning_rate": 0.0002278512396694215,
      "loss": 0.5323,
      "step": 301
    },
    {
      "epoch": 0.49,
      "grad_norm": 0.23715418577194214,
      "learning_rate": 0.00022760330578512394,
      "loss": 0.8333,
      "step": 302
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.29385048151016235,
      "learning_rate": 0.00022735537190082643,
      "loss": 0.6,
      "step": 303
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.26947689056396484,
      "learning_rate": 0.0002271074380165289,
      "loss": 0.8788,
      "step": 304
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.2778269946575165,
      "learning_rate": 0.0002268595041322314,
      "loss": 0.7073,
      "step": 305
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.20938479900360107,
      "learning_rate": 0.00022661157024793385,
      "loss": 0.6422,
      "step": 306
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.2777106761932373,
      "learning_rate": 0.00022636363636363633,
      "loss": 0.7495,
      "step": 307
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.20872819423675537,
      "learning_rate": 0.00022611570247933882,
      "loss": 0.6492,
      "step": 308
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.2752722501754761,
      "learning_rate": 0.0002258677685950413,
      "loss": 0.6014,
      "step": 309
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.24615786969661713,
      "learning_rate": 0.00022561983471074378,
      "loss": 0.6287,
      "step": 310
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.24146385490894318,
      "learning_rate": 0.00022537190082644624,
      "loss": 0.6151,
      "step": 311
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.24762235581874847,
      "learning_rate": 0.00022512396694214872,
      "loss": 0.6377,
      "step": 312
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.24630331993103027,
      "learning_rate": 0.00022487603305785124,
      "loss": 0.7255,
      "step": 313
    },
    {
      "epoch": 0.51,
      "grad_norm": 0.2922554612159729,
      "learning_rate": 0.00022462809917355372,
      "loss": 0.6645,
      "step": 314
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.21686063706874847,
      "learning_rate": 0.00022438016528925618,
      "loss": 0.5606,
      "step": 315
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.2216208428144455,
      "learning_rate": 0.00022413223140495866,
      "loss": 0.5126,
      "step": 316
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.25635436177253723,
      "learning_rate": 0.00022388429752066114,
      "loss": 0.7387,
      "step": 317
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.2786000669002533,
      "learning_rate": 0.00022363636363636363,
      "loss": 0.5941,
      "step": 318
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.26092806458473206,
      "learning_rate": 0.0002233884297520661,
      "loss": 0.7851,
      "step": 319
    },
    {
      "epoch": 0.52,
      "grad_norm": 0.23881889879703522,
      "learning_rate": 0.00022314049586776857,
      "loss": 0.598,
      "step": 320
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.23304526507854462,
      "learning_rate": 0.00022289256198347105,
      "loss": 0.7165,
      "step": 321
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.2340225875377655,
      "learning_rate": 0.00022264462809917353,
      "loss": 0.6608,
      "step": 322
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.31176140904426575,
      "learning_rate": 0.00022239669421487602,
      "loss": 0.6711,
      "step": 323
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.23832640051841736,
      "learning_rate": 0.00022214876033057847,
      "loss": 0.732,
      "step": 324
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.28845977783203125,
      "learning_rate": 0.00022190082644628096,
      "loss": 0.7968,
      "step": 325
    },
    {
      "epoch": 0.53,
      "grad_norm": 0.1978536993265152,
      "learning_rate": 0.00022165289256198344,
      "loss": 0.6592,
      "step": 326
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.26940053701400757,
      "learning_rate": 0.00022140495867768595,
      "loss": 0.7953,
      "step": 327
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.20393389463424683,
      "learning_rate": 0.00022115702479338844,
      "loss": 0.4871,
      "step": 328
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.27152347564697266,
      "learning_rate": 0.0002209090909090909,
      "loss": 0.5583,
      "step": 329
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.2883144021034241,
      "learning_rate": 0.00022066115702479338,
      "loss": 0.6156,
      "step": 330
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.1987351030111313,
      "learning_rate": 0.00022041322314049586,
      "loss": 0.5196,
      "step": 331
    },
    {
      "epoch": 0.54,
      "grad_norm": 0.2651583254337311,
      "learning_rate": 0.00022016528925619834,
      "loss": 0.6099,
      "step": 332
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.2574511468410492,
      "learning_rate": 0.0002199173553719008,
      "loss": 0.6925,
      "step": 333
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.27730292081832886,
      "learning_rate": 0.00021966942148760328,
      "loss": 0.6752,
      "step": 334
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.2001207172870636,
      "learning_rate": 0.00021942148760330577,
      "loss": 0.75,
      "step": 335
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.24222363531589508,
      "learning_rate": 0.00021917355371900825,
      "loss": 0.6364,
      "step": 336
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.26326724886894226,
      "learning_rate": 0.0002189256198347107,
      "loss": 0.673,
      "step": 337
    },
    {
      "epoch": 0.55,
      "grad_norm": 0.2272881418466568,
      "learning_rate": 0.0002186776859504132,
      "loss": 0.561,
      "step": 338
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.24880024790763855,
      "learning_rate": 0.00021842975206611567,
      "loss": 0.5552,
      "step": 339
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.2593706548213959,
      "learning_rate": 0.00021818181818181816,
      "loss": 0.5417,
      "step": 340
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.19063642621040344,
      "learning_rate": 0.00021793388429752067,
      "loss": 0.5694,
      "step": 341
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.2146475464105606,
      "learning_rate": 0.0002176859504132231,
      "loss": 0.4314,
      "step": 342
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.25150927901268005,
      "learning_rate": 0.0002174380165289256,
      "loss": 0.631,
      "step": 343
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.2753889858722687,
      "learning_rate": 0.0002171900826446281,
      "loss": 0.6859,
      "step": 344
    },
    {
      "epoch": 0.56,
      "grad_norm": 0.20773079991340637,
      "learning_rate": 0.00021694214876033058,
      "loss": 0.7515,
      "step": 345
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.2547062635421753,
      "learning_rate": 0.00021669421487603303,
      "loss": 0.7582,
      "step": 346
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.24687208235263824,
      "learning_rate": 0.00021644628099173552,
      "loss": 0.5865,
      "step": 347
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.24116279184818268,
      "learning_rate": 0.000216198347107438,
      "loss": 0.4841,
      "step": 348
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.2270282804965973,
      "learning_rate": 0.00021595041322314048,
      "loss": 0.5933,
      "step": 349
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.21436922252178192,
      "learning_rate": 0.00021570247933884297,
      "loss": 0.6959,
      "step": 350
    },
    {
      "epoch": 0.57,
      "grad_norm": 0.25802701711654663,
      "learning_rate": 0.00021545454545454542,
      "loss": 0.729,
      "step": 351
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.23808260262012482,
      "learning_rate": 0.0002152066115702479,
      "loss": 0.6346,
      "step": 352
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.23161651194095612,
      "learning_rate": 0.0002149586776859504,
      "loss": 0.6459,
      "step": 353
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.2442287802696228,
      "learning_rate": 0.00021471074380165287,
      "loss": 0.6803,
      "step": 354
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.19150683283805847,
      "learning_rate": 0.00021446280991735533,
      "loss": 0.4375,
      "step": 355
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.23142127692699432,
      "learning_rate": 0.00021421487603305781,
      "loss": 0.5505,
      "step": 356
    },
    {
      "epoch": 0.58,
      "grad_norm": 0.22447548806667328,
      "learning_rate": 0.00021396694214876033,
      "loss": 0.6368,
      "step": 357
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.25168758630752563,
      "learning_rate": 0.0002137190082644628,
      "loss": 0.6322,
      "step": 358
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.25538235902786255,
      "learning_rate": 0.0002134710743801653,
      "loss": 0.5317,
      "step": 359
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.2565425634384155,
      "learning_rate": 0.00021322314049586775,
      "loss": 0.6261,
      "step": 360
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.25399863719940186,
      "learning_rate": 0.00021297520661157023,
      "loss": 0.596,
      "step": 361
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.27143988013267517,
      "learning_rate": 0.00021272727272727272,
      "loss": 0.6691,
      "step": 362
    },
    {
      "epoch": 0.59,
      "grad_norm": 0.2387736439704895,
      "learning_rate": 0.0002124793388429752,
      "loss": 0.5288,
      "step": 363
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.2549780607223511,
      "learning_rate": 0.00021223140495867766,
      "loss": 0.7455,
      "step": 364
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.2740858793258667,
      "learning_rate": 0.00021198347107438014,
      "loss": 0.4921,
      "step": 365
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.25273847579956055,
      "learning_rate": 0.00021173553719008262,
      "loss": 0.7965,
      "step": 366
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.25858959555625916,
      "learning_rate": 0.0002114876033057851,
      "loss": 0.7303,
      "step": 367
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.2599296271800995,
      "learning_rate": 0.0002112396694214876,
      "loss": 0.6342,
      "step": 368
    },
    {
      "epoch": 0.6,
      "grad_norm": 0.21084599196910858,
      "learning_rate": 0.00021099173553719005,
      "loss": 0.633,
      "step": 369
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.24272632598876953,
      "learning_rate": 0.00021074380165289253,
      "loss": 0.6213,
      "step": 370
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.26323699951171875,
      "learning_rate": 0.00021049586776859501,
      "loss": 0.563,
      "step": 371
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.20646587014198303,
      "learning_rate": 0.00021024793388429753,
      "loss": 0.6248,
      "step": 372
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.21778297424316406,
      "learning_rate": 0.00020999999999999998,
      "loss": 0.7186,
      "step": 373
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.21315112709999084,
      "learning_rate": 0.00020975206611570247,
      "loss": 0.5961,
      "step": 374
    },
    {
      "epoch": 0.61,
      "grad_norm": 0.20787106454372406,
      "learning_rate": 0.00020950413223140495,
      "loss": 0.5917,
      "step": 375
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.23541009426116943,
      "learning_rate": 0.00020925619834710743,
      "loss": 0.7803,
      "step": 376
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.22649626433849335,
      "learning_rate": 0.00020900826446280992,
      "loss": 0.5895,
      "step": 377
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.23644742369651794,
      "learning_rate": 0.00020876033057851237,
      "loss": 0.6656,
      "step": 378
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.22934262454509735,
      "learning_rate": 0.00020851239669421486,
      "loss": 0.5933,
      "step": 379
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.289989709854126,
      "learning_rate": 0.00020826446280991734,
      "loss": 0.6852,
      "step": 380
    },
    {
      "epoch": 0.62,
      "grad_norm": 0.24489325284957886,
      "learning_rate": 0.00020801652892561982,
      "loss": 0.5546,
      "step": 381
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.27165278792381287,
      "learning_rate": 0.00020776859504132228,
      "loss": 0.6845,
      "step": 382
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.19467370212078094,
      "learning_rate": 0.00020752066115702476,
      "loss": 0.5587,
      "step": 383
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.27320200204849243,
      "learning_rate": 0.00020727272727272725,
      "loss": 0.7144,
      "step": 384
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.28100526332855225,
      "learning_rate": 0.00020702479338842973,
      "loss": 0.6914,
      "step": 385
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.3059975504875183,
      "learning_rate": 0.0002067768595041322,
      "loss": 0.6075,
      "step": 386
    },
    {
      "epoch": 0.63,
      "grad_norm": 0.24904222786426544,
      "learning_rate": 0.00020652892561983467,
      "loss": 0.5543,
      "step": 387
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.24768255650997162,
      "learning_rate": 0.00020628099173553718,
      "loss": 0.607,
      "step": 388
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.25083738565444946,
      "learning_rate": 0.00020603305785123967,
      "loss": 0.7961,
      "step": 389
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.26338303089141846,
      "learning_rate": 0.00020578512396694215,
      "loss": 0.6467,
      "step": 390
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.25761598348617554,
      "learning_rate": 0.0002055371900826446,
      "loss": 0.5891,
      "step": 391
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.2616937756538391,
      "learning_rate": 0.0002052892561983471,
      "loss": 0.5706,
      "step": 392
    },
    {
      "epoch": 0.64,
      "grad_norm": 0.18980839848518372,
      "learning_rate": 0.00020504132231404957,
      "loss": 0.4479,
      "step": 393
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.250431627035141,
      "learning_rate": 0.00020479338842975206,
      "loss": 0.6006,
      "step": 394
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.2146655172109604,
      "learning_rate": 0.0002045454545454545,
      "loss": 0.7113,
      "step": 395
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.2195209115743637,
      "learning_rate": 0.000204297520661157,
      "loss": 0.5354,
      "step": 396
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.24879257380962372,
      "learning_rate": 0.00020404958677685948,
      "loss": 0.5478,
      "step": 397
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.27159082889556885,
      "learning_rate": 0.00020380165289256196,
      "loss": 0.7681,
      "step": 398
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.20614947378635406,
      "learning_rate": 0.00020355371900826445,
      "loss": 0.6357,
      "step": 399
    },
    {
      "epoch": 0.65,
      "grad_norm": 0.25690051913261414,
      "learning_rate": 0.0002033057851239669,
      "loss": 0.5731,
      "step": 400
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.24473583698272705,
      "learning_rate": 0.0002030578512396694,
      "loss": 0.6784,
      "step": 401
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.32395297288894653,
      "learning_rate": 0.0002028099173553719,
      "loss": 0.7118,
      "step": 402
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.2975274324417114,
      "learning_rate": 0.00020256198347107438,
      "loss": 0.6504,
      "step": 403
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.2652553915977478,
      "learning_rate": 0.00020231404958677684,
      "loss": 0.6986,
      "step": 404
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.29475778341293335,
      "learning_rate": 0.00020206611570247932,
      "loss": 0.6525,
      "step": 405
    },
    {
      "epoch": 0.66,
      "grad_norm": 0.24549973011016846,
      "learning_rate": 0.0002018181818181818,
      "loss": 0.5408,
      "step": 406
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.2181435376405716,
      "learning_rate": 0.0002015702479338843,
      "loss": 0.6146,
      "step": 407
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.2682584226131439,
      "learning_rate": 0.00020132231404958677,
      "loss": 0.6368,
      "step": 408
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.2641114592552185,
      "learning_rate": 0.00020107438016528923,
      "loss": 0.51,
      "step": 409
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.27871838212013245,
      "learning_rate": 0.0002008264462809917,
      "loss": 0.7269,
      "step": 410
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.23890569806098938,
      "learning_rate": 0.0002005785123966942,
      "loss": 0.6444,
      "step": 411
    },
    {
      "epoch": 0.67,
      "grad_norm": 0.2451583445072174,
      "learning_rate": 0.00020033057851239668,
      "loss": 0.5806,
      "step": 412
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.2743864953517914,
      "learning_rate": 0.00020008264462809914,
      "loss": 0.6305,
      "step": 413
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.2626914978027344,
      "learning_rate": 0.00019983471074380162,
      "loss": 0.5765,
      "step": 414
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.2874875068664551,
      "learning_rate": 0.0001995867768595041,
      "loss": 0.5928,
      "step": 415
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.30499163269996643,
      "learning_rate": 0.00019933884297520661,
      "loss": 0.6271,
      "step": 416
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.30474454164505005,
      "learning_rate": 0.0001990909090909091,
      "loss": 0.6755,
      "step": 417
    },
    {
      "epoch": 0.68,
      "grad_norm": 0.1819755882024765,
      "learning_rate": 0.00019884297520661155,
      "loss": 0.394,
      "step": 418
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.25470343232154846,
      "learning_rate": 0.00019859504132231404,
      "loss": 0.7121,
      "step": 419
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.26749151945114136,
      "learning_rate": 0.00019834710743801652,
      "loss": 0.6487,
      "step": 420
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.20643912255764008,
      "learning_rate": 0.000198099173553719,
      "loss": 0.4585,
      "step": 421
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.2576930522918701,
      "learning_rate": 0.00019785123966942146,
      "loss": 0.5235,
      "step": 422
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.2899012863636017,
      "learning_rate": 0.00019760330578512395,
      "loss": 0.6292,
      "step": 423
    },
    {
      "epoch": 0.69,
      "grad_norm": 0.2541065216064453,
      "learning_rate": 0.00019735537190082643,
      "loss": 0.648,
      "step": 424
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.24382047355175018,
      "learning_rate": 0.0001971074380165289,
      "loss": 0.5939,
      "step": 425
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.22931940853595734,
      "learning_rate": 0.00019685950413223137,
      "loss": 0.6812,
      "step": 426
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.2592567205429077,
      "learning_rate": 0.00019661157024793385,
      "loss": 0.69,
      "step": 427
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.2516980767250061,
      "learning_rate": 0.00019636363636363634,
      "loss": 0.5707,
      "step": 428
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.23515059053897858,
      "learning_rate": 0.00019611570247933882,
      "loss": 0.6739,
      "step": 429
    },
    {
      "epoch": 0.7,
      "grad_norm": 0.24742184579372406,
      "learning_rate": 0.00019586776859504133,
      "loss": 0.6761,
      "step": 430
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.26232922077178955,
      "learning_rate": 0.00019561983471074376,
      "loss": 0.7071,
      "step": 431
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.2853042781352997,
      "learning_rate": 0.00019537190082644627,
      "loss": 0.7667,
      "step": 432
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.251169353723526,
      "learning_rate": 0.00019512396694214875,
      "loss": 0.6518,
      "step": 433
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.2321665734052658,
      "learning_rate": 0.00019487603305785124,
      "loss": 0.4377,
      "step": 434
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.25216928124427795,
      "learning_rate": 0.0001946280991735537,
      "loss": 0.7173,
      "step": 435
    },
    {
      "epoch": 0.71,
      "grad_norm": 0.19498330354690552,
      "learning_rate": 0.00019438016528925618,
      "loss": 0.5584,
      "step": 436
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.32786309719085693,
      "learning_rate": 0.00019413223140495866,
      "loss": 0.6583,
      "step": 437
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.25834760069847107,
      "learning_rate": 0.00019388429752066115,
      "loss": 0.4957,
      "step": 438
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.3462083041667938,
      "learning_rate": 0.00019363636363636363,
      "loss": 0.5205,
      "step": 439
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.27106693387031555,
      "learning_rate": 0.00019338842975206609,
      "loss": 0.6803,
      "step": 440
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.28165388107299805,
      "learning_rate": 0.00019314049586776857,
      "loss": 0.7049,
      "step": 441
    },
    {
      "epoch": 0.72,
      "grad_norm": 0.20732273161411285,
      "learning_rate": 0.00019289256198347105,
      "loss": 0.6407,
      "step": 442
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.2609116733074188,
      "learning_rate": 0.00019264462809917354,
      "loss": 0.5377,
      "step": 443
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.2561998963356018,
      "learning_rate": 0.000192396694214876,
      "loss": 0.6212,
      "step": 444
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.27699044346809387,
      "learning_rate": 0.00019214876033057848,
      "loss": 0.5482,
      "step": 445
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.2426328808069229,
      "learning_rate": 0.000191900826446281,
      "loss": 0.6444,
      "step": 446
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.26187026500701904,
      "learning_rate": 0.00019165289256198347,
      "loss": 0.5443,
      "step": 447
    },
    {
      "epoch": 0.73,
      "grad_norm": 0.2719630002975464,
      "learning_rate": 0.00019140495867768595,
      "loss": 0.6886,
      "step": 448
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.18477971851825714,
      "learning_rate": 0.0001911570247933884,
      "loss": 0.5292,
      "step": 449
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.2144313007593155,
      "learning_rate": 0.0001909090909090909,
      "loss": 0.4613,
      "step": 450
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.2580784857273102,
      "learning_rate": 0.00019066115702479338,
      "loss": 0.5606,
      "step": 451
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.3073588013648987,
      "learning_rate": 0.00019041322314049586,
      "loss": 0.6123,
      "step": 452
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.21787844598293304,
      "learning_rate": 0.00019016528925619832,
      "loss": 0.5939,
      "step": 453
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.255750447511673,
      "learning_rate": 0.0001899173553719008,
      "loss": 0.5739,
      "step": 454
    },
    {
      "epoch": 0.74,
      "grad_norm": 0.24147820472717285,
      "learning_rate": 0.00018966942148760329,
      "loss": 0.6026,
      "step": 455
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.26172590255737305,
      "learning_rate": 0.00018942148760330577,
      "loss": 0.5166,
      "step": 456
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.2710455358028412,
      "learning_rate": 0.00018917355371900825,
      "loss": 0.6429,
      "step": 457
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.1971074640750885,
      "learning_rate": 0.0001889256198347107,
      "loss": 0.4799,
      "step": 458
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.23394368588924408,
      "learning_rate": 0.0001886776859504132,
      "loss": 0.5491,
      "step": 459
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.22820048034191132,
      "learning_rate": 0.0001884297520661157,
      "loss": 0.5343,
      "step": 460
    },
    {
      "epoch": 0.75,
      "grad_norm": 0.23169974982738495,
      "learning_rate": 0.0001881818181818182,
      "loss": 0.5852,
      "step": 461
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.24015003442764282,
      "learning_rate": 0.00018793388429752064,
      "loss": 0.6209,
      "step": 462
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.2230776697397232,
      "learning_rate": 0.00018768595041322313,
      "loss": 0.6296,
      "step": 463
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.2518354654312134,
      "learning_rate": 0.0001874380165289256,
      "loss": 0.6167,
      "step": 464
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.338256299495697,
      "learning_rate": 0.0001871900826446281,
      "loss": 0.6512,
      "step": 465
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.23796728253364563,
      "learning_rate": 0.00018694214876033055,
      "loss": 0.8155,
      "step": 466
    },
    {
      "epoch": 0.76,
      "grad_norm": 0.31516361236572266,
      "learning_rate": 0.00018669421487603303,
      "loss": 0.8023,
      "step": 467
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.2371574491262436,
      "learning_rate": 0.00018644628099173552,
      "loss": 0.5613,
      "step": 468
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.2822033762931824,
      "learning_rate": 0.000186198347107438,
      "loss": 0.5549,
      "step": 469
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.25953295826911926,
      "learning_rate": 0.00018595041322314049,
      "loss": 0.6199,
      "step": 470
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.2478639930486679,
      "learning_rate": 0.00018570247933884294,
      "loss": 0.5806,
      "step": 471
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.2439350187778473,
      "learning_rate": 0.00018545454545454543,
      "loss": 0.6222,
      "step": 472
    },
    {
      "epoch": 0.77,
      "grad_norm": 0.24993474781513214,
      "learning_rate": 0.0001852066115702479,
      "loss": 0.6048,
      "step": 473
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.24781496822834015,
      "learning_rate": 0.00018495867768595042,
      "loss": 0.5941,
      "step": 474
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.1847202032804489,
      "learning_rate": 0.00018471074380165285,
      "loss": 0.609,
      "step": 475
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.21596528589725494,
      "learning_rate": 0.00018446280991735536,
      "loss": 0.4457,
      "step": 476
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.240879625082016,
      "learning_rate": 0.00018421487603305784,
      "loss": 0.6118,
      "step": 477
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.2898111641407013,
      "learning_rate": 0.00018396694214876033,
      "loss": 0.7725,
      "step": 478
    },
    {
      "epoch": 0.78,
      "grad_norm": 0.27428382635116577,
      "learning_rate": 0.0001837190082644628,
      "loss": 0.5366,
      "step": 479
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.23467296361923218,
      "learning_rate": 0.00018347107438016527,
      "loss": 0.6018,
      "step": 480
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.2190561592578888,
      "learning_rate": 0.00018322314049586775,
      "loss": 0.5249,
      "step": 481
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.2240625023841858,
      "learning_rate": 0.00018297520661157024,
      "loss": 0.6891,
      "step": 482
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.24726848304271698,
      "learning_rate": 0.00018272727272727272,
      "loss": 0.5545,
      "step": 483
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.3318251371383667,
      "learning_rate": 0.00018247933884297518,
      "loss": 0.4809,
      "step": 484
    },
    {
      "epoch": 0.79,
      "grad_norm": 0.2396695613861084,
      "learning_rate": 0.00018223140495867766,
      "loss": 0.4942,
      "step": 485
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.25009942054748535,
      "learning_rate": 0.00018198347107438014,
      "loss": 0.7381,
      "step": 486
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.22655311226844788,
      "learning_rate": 0.00018173553719008263,
      "loss": 0.4729,
      "step": 487
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.23187695443630219,
      "learning_rate": 0.0001814876033057851,
      "loss": 0.5719,
      "step": 488
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.2703653573989868,
      "learning_rate": 0.00018123966942148757,
      "loss": 0.6031,
      "step": 489
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.2207796424627304,
      "learning_rate": 0.00018099173553719008,
      "loss": 0.5361,
      "step": 490
    },
    {
      "epoch": 0.8,
      "grad_norm": 0.24914169311523438,
      "learning_rate": 0.00018074380165289256,
      "loss": 0.6547,
      "step": 491
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.2714746594429016,
      "learning_rate": 0.00018049586776859504,
      "loss": 0.5702,
      "step": 492
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.3201580047607422,
      "learning_rate": 0.0001802479338842975,
      "loss": 0.6119,
      "step": 493
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.2548397183418274,
      "learning_rate": 0.00017999999999999998,
      "loss": 0.5251,
      "step": 494
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.28669115900993347,
      "learning_rate": 0.00017975206611570247,
      "loss": 0.5773,
      "step": 495
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.26253971457481384,
      "learning_rate": 0.00017950413223140495,
      "loss": 0.6504,
      "step": 496
    },
    {
      "epoch": 0.81,
      "grad_norm": 0.22113384306430817,
      "learning_rate": 0.00017925619834710744,
      "loss": 0.4741,
      "step": 497
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.261636346578598,
      "learning_rate": 0.0001790082644628099,
      "loss": 0.6241,
      "step": 498
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.1780402809381485,
      "learning_rate": 0.00017876033057851238,
      "loss": 0.5207,
      "step": 499
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.26149195432662964,
      "learning_rate": 0.00017851239669421486,
      "loss": 0.5872,
      "step": 500
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.26113009452819824,
      "learning_rate": 0.00017826446280991734,
      "loss": 0.6163,
      "step": 501
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.21397502720355988,
      "learning_rate": 0.0001780165289256198,
      "loss": 0.479,
      "step": 502
    },
    {
      "epoch": 0.82,
      "grad_norm": 0.21250088512897491,
      "learning_rate": 0.00017776859504132228,
      "loss": 0.6978,
      "step": 503
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.2556426525115967,
      "learning_rate": 0.00017752066115702477,
      "loss": 0.6128,
      "step": 504
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.24139715731143951,
      "learning_rate": 0.00017727272727272728,
      "loss": 0.5066,
      "step": 505
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.23671215772628784,
      "learning_rate": 0.00017702479338842976,
      "loss": 0.5183,
      "step": 506
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.23494285345077515,
      "learning_rate": 0.00017677685950413222,
      "loss": 0.5181,
      "step": 507
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.2547609806060791,
      "learning_rate": 0.0001765289256198347,
      "loss": 0.5406,
      "step": 508
    },
    {
      "epoch": 0.83,
      "grad_norm": 0.3042651414871216,
      "learning_rate": 0.00017628099173553718,
      "loss": 0.5551,
      "step": 509
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.22910748422145844,
      "learning_rate": 0.00017603305785123967,
      "loss": 0.6373,
      "step": 510
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.19777967035770416,
      "learning_rate": 0.00017578512396694212,
      "loss": 0.5471,
      "step": 511
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.31034502387046814,
      "learning_rate": 0.0001755371900826446,
      "loss": 0.7017,
      "step": 512
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.3504410684108734,
      "learning_rate": 0.0001752892561983471,
      "loss": 0.7208,
      "step": 513
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.24271292984485626,
      "learning_rate": 0.00017504132231404958,
      "loss": 0.5563,
      "step": 514
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.27147865295410156,
      "learning_rate": 0.00017479338842975203,
      "loss": 0.5869,
      "step": 515
    },
    {
      "epoch": 0.84,
      "grad_norm": 0.2976628839969635,
      "learning_rate": 0.00017454545454545452,
      "loss": 0.5471,
      "step": 516
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.28489646315574646,
      "learning_rate": 0.000174297520661157,
      "loss": 0.6053,
      "step": 517
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.30020108819007874,
      "learning_rate": 0.00017404958677685948,
      "loss": 0.6178,
      "step": 518
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.23986253142356873,
      "learning_rate": 0.000173801652892562,
      "loss": 0.5896,
      "step": 519
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.2667832374572754,
      "learning_rate": 0.00017355371900826442,
      "loss": 0.5375,
      "step": 520
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.22176356613636017,
      "learning_rate": 0.00017330578512396693,
      "loss": 0.5723,
      "step": 521
    },
    {
      "epoch": 0.85,
      "grad_norm": 0.263257771730423,
      "learning_rate": 0.00017305785123966942,
      "loss": 0.7317,
      "step": 522
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.24838753044605255,
      "learning_rate": 0.0001728099173553719,
      "loss": 0.5849,
      "step": 523
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.24839664995670319,
      "learning_rate": 0.00017256198347107436,
      "loss": 0.6678,
      "step": 524
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.2849573493003845,
      "learning_rate": 0.00017231404958677684,
      "loss": 0.7144,
      "step": 525
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.26900768280029297,
      "learning_rate": 0.00017206611570247932,
      "loss": 0.5156,
      "step": 526
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.2212425172328949,
      "learning_rate": 0.0001718181818181818,
      "loss": 0.4551,
      "step": 527
    },
    {
      "epoch": 0.86,
      "grad_norm": 0.2066129595041275,
      "learning_rate": 0.0001715702479338843,
      "loss": 0.4193,
      "step": 528
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.2838365137577057,
      "learning_rate": 0.00017132231404958675,
      "loss": 0.6078,
      "step": 529
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.239679753780365,
      "learning_rate": 0.00017107438016528923,
      "loss": 0.616,
      "step": 530
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.23269398510456085,
      "learning_rate": 0.00017082644628099172,
      "loss": 0.542,
      "step": 531
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.23838558793067932,
      "learning_rate": 0.0001705785123966942,
      "loss": 0.5147,
      "step": 532
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.2819415330886841,
      "learning_rate": 0.00017033057851239666,
      "loss": 0.6437,
      "step": 533
    },
    {
      "epoch": 0.87,
      "grad_norm": 0.243398055434227,
      "learning_rate": 0.00017008264462809914,
      "loss": 0.6611,
      "step": 534
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.22569122910499573,
      "learning_rate": 0.00016983471074380165,
      "loss": 0.3979,
      "step": 535
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.33265820145606995,
      "learning_rate": 0.00016958677685950413,
      "loss": 0.6005,
      "step": 536
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.26828673481941223,
      "learning_rate": 0.00016933884297520662,
      "loss": 0.608,
      "step": 537
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.24439513683319092,
      "learning_rate": 0.00016909090909090907,
      "loss": 0.5572,
      "step": 538
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.22491876780986786,
      "learning_rate": 0.00016884297520661156,
      "loss": 0.7226,
      "step": 539
    },
    {
      "epoch": 0.88,
      "grad_norm": 0.24468480050563812,
      "learning_rate": 0.00016859504132231404,
      "loss": 0.4582,
      "step": 540
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.23392945528030396,
      "learning_rate": 0.00016834710743801652,
      "loss": 0.6477,
      "step": 541
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.27548858523368835,
      "learning_rate": 0.00016809917355371898,
      "loss": 0.5846,
      "step": 542
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.2861180603504181,
      "learning_rate": 0.00016785123966942146,
      "loss": 0.6412,
      "step": 543
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.24700766801834106,
      "learning_rate": 0.00016760330578512395,
      "loss": 0.6947,
      "step": 544
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.2600953280925751,
      "learning_rate": 0.00016735537190082643,
      "loss": 0.6165,
      "step": 545
    },
    {
      "epoch": 0.89,
      "grad_norm": 0.26876646280288696,
      "learning_rate": 0.00016710743801652892,
      "loss": 0.6855,
      "step": 546
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.26161080598831177,
      "learning_rate": 0.00016685950413223137,
      "loss": 0.5066,
      "step": 547
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.25190046429634094,
      "learning_rate": 0.00016661157024793386,
      "loss": 0.5902,
      "step": 548
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.25269225239753723,
      "learning_rate": 0.00016636363636363637,
      "loss": 0.7017,
      "step": 549
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.28042706847190857,
      "learning_rate": 0.00016611570247933885,
      "loss": 0.6264,
      "step": 550
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.2767360508441925,
      "learning_rate": 0.0001658677685950413,
      "loss": 0.7562,
      "step": 551
    },
    {
      "epoch": 0.9,
      "grad_norm": 0.2771216034889221,
      "learning_rate": 0.0001656198347107438,
      "loss": 0.5333,
      "step": 552
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.189210906624794,
      "learning_rate": 0.00016537190082644627,
      "loss": 0.5378,
      "step": 553
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.22517065703868866,
      "learning_rate": 0.00016512396694214876,
      "loss": 0.5292,
      "step": 554
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.2390165776014328,
      "learning_rate": 0.00016487603305785121,
      "loss": 0.4407,
      "step": 555
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.21548262238502502,
      "learning_rate": 0.0001646280991735537,
      "loss": 0.4504,
      "step": 556
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.20831167697906494,
      "learning_rate": 0.00016438016528925618,
      "loss": 0.6848,
      "step": 557
    },
    {
      "epoch": 0.91,
      "grad_norm": 0.271257609128952,
      "learning_rate": 0.00016413223140495866,
      "loss": 0.535,
      "step": 558
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.32008254528045654,
      "learning_rate": 0.00016388429752066115,
      "loss": 0.5107,
      "step": 559
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.34058302640914917,
      "learning_rate": 0.0001636363636363636,
      "loss": 0.5708,
      "step": 560
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.28070059418678284,
      "learning_rate": 0.0001633884297520661,
      "loss": 0.5086,
      "step": 561
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.25487688183784485,
      "learning_rate": 0.00016314049586776857,
      "loss": 0.5184,
      "step": 562
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.3240332007408142,
      "learning_rate": 0.00016289256198347108,
      "loss": 0.6774,
      "step": 563
    },
    {
      "epoch": 0.92,
      "grad_norm": 0.30744409561157227,
      "learning_rate": 0.0001626446280991735,
      "loss": 0.5314,
      "step": 564
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.25220754742622375,
      "learning_rate": 0.00016239669421487602,
      "loss": 0.6308,
      "step": 565
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.29116958379745483,
      "learning_rate": 0.0001621487603305785,
      "loss": 0.5685,
      "step": 566
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.23250073194503784,
      "learning_rate": 0.000161900826446281,
      "loss": 0.4318,
      "step": 567
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.2808091640472412,
      "learning_rate": 0.00016165289256198347,
      "loss": 0.6313,
      "step": 568
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.2711193561553955,
      "learning_rate": 0.00016140495867768593,
      "loss": 0.4651,
      "step": 569
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.29540935158729553,
      "learning_rate": 0.00016115702479338841,
      "loss": 0.6663,
      "step": 570
    },
    {
      "epoch": 0.93,
      "grad_norm": 0.23418714106082916,
      "learning_rate": 0.0001609090909090909,
      "loss": 0.448,
      "step": 571
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.21675793826580048,
      "learning_rate": 0.00016066115702479338,
      "loss": 0.5034,
      "step": 572
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.22451865673065186,
      "learning_rate": 0.00016041322314049584,
      "loss": 0.4476,
      "step": 573
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.26300856471061707,
      "learning_rate": 0.00016016528925619832,
      "loss": 0.6646,
      "step": 574
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.3377116918563843,
      "learning_rate": 0.0001599173553719008,
      "loss": 0.6029,
      "step": 575
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.23391880095005035,
      "learning_rate": 0.0001596694214876033,
      "loss": 0.6277,
      "step": 576
    },
    {
      "epoch": 0.94,
      "grad_norm": 0.19620922207832336,
      "learning_rate": 0.0001594214876033058,
      "loss": 0.4638,
      "step": 577
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.22981096804141998,
      "learning_rate": 0.00015917355371900823,
      "loss": 0.5826,
      "step": 578
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.34321555495262146,
      "learning_rate": 0.00015892561983471074,
      "loss": 0.5618,
      "step": 579
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.28461968898773193,
      "learning_rate": 0.00015867768595041322,
      "loss": 0.5129,
      "step": 580
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.24368269741535187,
      "learning_rate": 0.0001584297520661157,
      "loss": 0.5866,
      "step": 581
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.282255083322525,
      "learning_rate": 0.00015818181818181816,
      "loss": 0.6274,
      "step": 582
    },
    {
      "epoch": 0.95,
      "grad_norm": 0.26298072934150696,
      "learning_rate": 0.00015793388429752065,
      "loss": 0.5187,
      "step": 583
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.2671455144882202,
      "learning_rate": 0.00015768595041322313,
      "loss": 0.6878,
      "step": 584
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.2681390643119812,
      "learning_rate": 0.00015743801652892561,
      "loss": 0.5469,
      "step": 585
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.38484248518943787,
      "learning_rate": 0.0001571900826446281,
      "loss": 0.6364,
      "step": 586
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.23353587090969086,
      "learning_rate": 0.00015694214876033055,
      "loss": 0.4844,
      "step": 587
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.29452502727508545,
      "learning_rate": 0.00015669421487603304,
      "loss": 0.5059,
      "step": 588
    },
    {
      "epoch": 0.96,
      "grad_norm": 0.2460879236459732,
      "learning_rate": 0.00015644628099173552,
      "loss": 0.6495,
      "step": 589
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.30693721771240234,
      "learning_rate": 0.000156198347107438,
      "loss": 0.5165,
      "step": 590
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.2171495109796524,
      "learning_rate": 0.00015595041322314046,
      "loss": 0.6172,
      "step": 591
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.24301984906196594,
      "learning_rate": 0.00015570247933884294,
      "loss": 0.6786,
      "step": 592
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.2288222461938858,
      "learning_rate": 0.00015545454545454546,
      "loss": 0.5669,
      "step": 593
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.2407921552658081,
      "learning_rate": 0.00015520661157024794,
      "loss": 0.5968,
      "step": 594
    },
    {
      "epoch": 0.97,
      "grad_norm": 0.2591527998447418,
      "learning_rate": 0.0001549586776859504,
      "loss": 0.544,
      "step": 595
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.25770679116249084,
      "learning_rate": 0.00015471074380165288,
      "loss": 0.7177,
      "step": 596
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.2528848648071289,
      "learning_rate": 0.00015446280991735536,
      "loss": 0.4703,
      "step": 597
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.24993537366390228,
      "learning_rate": 0.00015421487603305785,
      "loss": 0.6003,
      "step": 598
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.25807908177375793,
      "learning_rate": 0.00015396694214876033,
      "loss": 0.465,
      "step": 599
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.3142452836036682,
      "learning_rate": 0.0001537190082644628,
      "loss": 0.6122,
      "step": 600
    },
    {
      "epoch": 0.98,
      "grad_norm": 0.27111849188804626,
      "learning_rate": 0.00015347107438016527,
      "loss": 0.5962,
      "step": 601
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.28503674268722534,
      "learning_rate": 0.00015322314049586775,
      "loss": 0.6667,
      "step": 602
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.27074381709098816,
      "learning_rate": 0.00015297520661157024,
      "loss": 0.6115,
      "step": 603
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.25918465852737427,
      "learning_rate": 0.0001527272727272727,
      "loss": 0.4483,
      "step": 604
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.24476633965969086,
      "learning_rate": 0.00015247933884297518,
      "loss": 0.6501,
      "step": 605
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.21205200254917145,
      "learning_rate": 0.00015223140495867766,
      "loss": 0.3914,
      "step": 606
    },
    {
      "epoch": 0.99,
      "grad_norm": 0.25496751070022583,
      "learning_rate": 0.00015198347107438017,
      "loss": 0.5335,
      "step": 607
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.27991780638694763,
      "learning_rate": 0.00015173553719008266,
      "loss": 0.6083,
      "step": 608
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.23995639383792877,
      "learning_rate": 0.0001514876033057851,
      "loss": 0.55,
      "step": 609
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.2349666953086853,
      "learning_rate": 0.0001512396694214876,
      "loss": 0.7054,
      "step": 610
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.27498871088027954,
      "learning_rate": 0.00015099173553719008,
      "loss": 0.55,
      "step": 611
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.21346105635166168,
      "learning_rate": 0.00015074380165289256,
      "loss": 0.3467,
      "step": 612
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.2638354003429413,
      "learning_rate": 0.00015049586776859502,
      "loss": 0.5624,
      "step": 613
    },
    {
      "epoch": 1.01,
      "grad_norm": 0.2751975953578949,
      "learning_rate": 0.0001502479338842975,
      "loss": 0.3814,
      "step": 614
    },
    {
      "epoch": 1.01,
      "grad_norm": 0.225106880068779,
      "learning_rate": 0.00015,
      "loss": 0.479,
      "step": 615
    },
    {
      "epoch": 1.01,
      "grad_norm": 0.22013232111930847,
      "learning_rate": 0.00014975206611570247,
      "loss": 0.5672,
      "step": 616
    },
    {
      "epoch": 1.01,
      "grad_norm": 0.21252033114433289,
      "learning_rate": 0.00014950413223140495,
      "loss": 0.546,
      "step": 617
    },
    {
      "epoch": 1.01,
      "grad_norm": 0.2847185432910919,
      "learning_rate": 0.00014925619834710744,
      "loss": 0.4434,
      "step": 618
    },
    {
      "epoch": 1.01,
      "grad_norm": 0.25599631667137146,
      "learning_rate": 0.0001490082644628099,
      "loss": 0.4713,
      "step": 619
    },
    {
      "epoch": 1.02,
      "grad_norm": 0.2719402611255646,
      "learning_rate": 0.00014876033057851238,
      "loss": 0.4475,
      "step": 620
    },
    {
      "epoch": 1.02,
      "grad_norm": 0.26454958319664,
      "learning_rate": 0.00014851239669421486,
      "loss": 0.4515,
      "step": 621
    },
    {
      "epoch": 1.02,
      "grad_norm": 0.39801672101020813,
      "learning_rate": 0.00014826446280991735,
      "loss": 0.4647,
      "step": 622
    },
    {
      "epoch": 1.02,
      "grad_norm": 0.3378361463546753,
      "learning_rate": 0.0001480165289256198,
      "loss": 0.4414,
      "step": 623
    },
    {
      "epoch": 1.02,
      "grad_norm": 0.3039036989212036,
      "learning_rate": 0.0001477685950413223,
      "loss": 0.5634,
      "step": 624
    },
    {
      "epoch": 1.02,
      "grad_norm": 0.3506157398223877,
      "learning_rate": 0.0001475206611570248,
      "loss": 0.5001,
      "step": 625
    },
    {
      "epoch": 1.02,
      "grad_norm": 0.2508845925331116,
      "learning_rate": 0.00014727272727272725,
      "loss": 0.3379,
      "step": 626
    },
    {
      "epoch": 1.03,
      "grad_norm": 0.26913216710090637,
      "learning_rate": 0.00014702479338842974,
      "loss": 0.4575,
      "step": 627
    },
    {
      "epoch": 1.03,
      "grad_norm": 0.329659640789032,
      "learning_rate": 0.00014677685950413222,
      "loss": 0.437,
      "step": 628
    },
    {
      "epoch": 1.03,
      "grad_norm": 0.2972075343132019,
      "learning_rate": 0.0001465289256198347,
      "loss": 0.5048,
      "step": 629
    },
    {
      "epoch": 1.03,
      "grad_norm": 0.3184354603290558,
      "learning_rate": 0.00014628099173553716,
      "loss": 0.4374,
      "step": 630
    },
    {
      "epoch": 1.03,
      "grad_norm": 0.3377355635166168,
      "learning_rate": 0.00014603305785123967,
      "loss": 0.4946,
      "step": 631
    },
    {
      "epoch": 1.03,
      "grad_norm": 0.29106801748275757,
      "learning_rate": 0.00014578512396694213,
      "loss": 0.5414,
      "step": 632
    },
    {
      "epoch": 1.04,
      "grad_norm": 0.22808948159217834,
      "learning_rate": 0.0001455371900826446,
      "loss": 0.3739,
      "step": 633
    },
    {
      "epoch": 1.04,
      "grad_norm": 0.27818021178245544,
      "learning_rate": 0.0001452892561983471,
      "loss": 0.4172,
      "step": 634
    },
    {
      "epoch": 1.04,
      "grad_norm": 0.25634923577308655,
      "learning_rate": 0.00014504132231404958,
      "loss": 0.4293,
      "step": 635
    },
    {
      "epoch": 1.04,
      "grad_norm": 0.30696937441825867,
      "learning_rate": 0.00014479338842975206,
      "loss": 0.4454,
      "step": 636
    },
    {
      "epoch": 1.04,
      "grad_norm": 0.26105087995529175,
      "learning_rate": 0.00014454545454545452,
      "loss": 0.2978,
      "step": 637
    },
    {
      "epoch": 1.04,
      "grad_norm": 0.3100634515285492,
      "learning_rate": 0.00014429752066115703,
      "loss": 0.4499,
      "step": 638
    },
    {
      "epoch": 1.05,
      "grad_norm": 0.27640992403030396,
      "learning_rate": 0.00014404958677685949,
      "loss": 0.3837,
      "step": 639
    },
    {
      "epoch": 1.05,
      "grad_norm": 0.24559038877487183,
      "learning_rate": 0.00014380165289256197,
      "loss": 0.3347,
      "step": 640
    },
    {
      "epoch": 1.05,
      "grad_norm": 0.2920415699481964,
      "learning_rate": 0.00014355371900826445,
      "loss": 0.4333,
      "step": 641
    },
    {
      "epoch": 1.05,
      "grad_norm": 0.3147384226322174,
      "learning_rate": 0.00014330578512396694,
      "loss": 0.4385,
      "step": 642
    },
    {
      "epoch": 1.05,
      "grad_norm": 0.35469138622283936,
      "learning_rate": 0.0001430578512396694,
      "loss": 0.5442,
      "step": 643
    },
    {
      "epoch": 1.05,
      "grad_norm": 0.2619563043117523,
      "learning_rate": 0.00014280991735537188,
      "loss": 0.3837,
      "step": 644
    },
    {
      "epoch": 1.06,
      "grad_norm": 0.32273221015930176,
      "learning_rate": 0.0001425619834710744,
      "loss": 0.4946,
      "step": 645
    },
    {
      "epoch": 1.06,
      "grad_norm": 0.2692110538482666,
      "learning_rate": 0.00014231404958677684,
      "loss": 0.4683,
      "step": 646
    },
    {
      "epoch": 1.06,
      "grad_norm": 0.35255464911460876,
      "learning_rate": 0.00014206611570247933,
      "loss": 0.5456,
      "step": 647
    },
    {
      "epoch": 1.06,
      "grad_norm": 0.29768630862236023,
      "learning_rate": 0.0001418181818181818,
      "loss": 0.3394,
      "step": 648
    },
    {
      "epoch": 1.06,
      "grad_norm": 0.30738797783851624,
      "learning_rate": 0.0001415702479338843,
      "loss": 0.3583,
      "step": 649
    },
    {
      "epoch": 1.06,
      "grad_norm": 0.33226314187049866,
      "learning_rate": 0.00014132231404958675,
      "loss": 0.4477,
      "step": 650
    },
    {
      "epoch": 1.07,
      "grad_norm": 0.2842199504375458,
      "learning_rate": 0.00014107438016528923,
      "loss": 0.4454,
      "step": 651
    },
    {
      "epoch": 1.07,
      "grad_norm": 0.28207266330718994,
      "learning_rate": 0.00014082644628099172,
      "loss": 0.3665,
      "step": 652
    },
    {
      "epoch": 1.07,
      "grad_norm": 0.2228500097990036,
      "learning_rate": 0.0001405785123966942,
      "loss": 0.3446,
      "step": 653
    },
    {
      "epoch": 1.07,
      "grad_norm": 0.2969403564929962,
      "learning_rate": 0.00014033057851239669,
      "loss": 0.377,
      "step": 654
    },
    {
      "epoch": 1.07,
      "grad_norm": 0.28087565302848816,
      "learning_rate": 0.00014008264462809917,
      "loss": 0.3683,
      "step": 655
    },
    {
      "epoch": 1.07,
      "grad_norm": 0.27268192172050476,
      "learning_rate": 0.00013983471074380165,
      "loss": 0.427,
      "step": 656
    },
    {
      "epoch": 1.08,
      "grad_norm": 0.339070588350296,
      "learning_rate": 0.0001395867768595041,
      "loss": 0.4887,
      "step": 657
    },
    {
      "epoch": 1.08,
      "grad_norm": 0.3170423209667206,
      "learning_rate": 0.0001393388429752066,
      "loss": 0.5097,
      "step": 658
    },
    {
      "epoch": 1.08,
      "grad_norm": 0.3114936947822571,
      "learning_rate": 0.00013909090909090908,
      "loss": 0.4587,
      "step": 659
    },
    {
      "epoch": 1.08,
      "grad_norm": 0.28112486004829407,
      "learning_rate": 0.00013884297520661156,
      "loss": 0.4781,
      "step": 660
    },
    {
      "epoch": 1.08,
      "grad_norm": 0.28116974234580994,
      "learning_rate": 0.00013859504132231404,
      "loss": 0.3546,
      "step": 661
    },
    {
      "epoch": 1.08,
      "grad_norm": 0.25061559677124023,
      "learning_rate": 0.00013834710743801653,
      "loss": 0.4512,
      "step": 662
    },
    {
      "epoch": 1.09,
      "grad_norm": 0.29854199290275574,
      "learning_rate": 0.00013809917355371898,
      "loss": 0.6068,
      "step": 663
    },
    {
      "epoch": 1.09,
      "grad_norm": 0.2901363670825958,
      "learning_rate": 0.00013785123966942147,
      "loss": 0.3667,
      "step": 664
    },
    {
      "epoch": 1.09,
      "grad_norm": 0.29766595363616943,
      "learning_rate": 0.00013760330578512395,
      "loss": 0.5194,
      "step": 665
    },
    {
      "epoch": 1.09,
      "grad_norm": 0.2765616476535797,
      "learning_rate": 0.00013735537190082643,
      "loss": 0.5079,
      "step": 666
    },
    {
      "epoch": 1.09,
      "grad_norm": 0.27531540393829346,
      "learning_rate": 0.00013710743801652892,
      "loss": 0.4423,
      "step": 667
    },
    {
      "epoch": 1.09,
      "grad_norm": 0.3063349425792694,
      "learning_rate": 0.0001368595041322314,
      "loss": 0.4666,
      "step": 668
    },
    {
      "epoch": 1.1,
      "grad_norm": 0.24519848823547363,
      "learning_rate": 0.00013661157024793389,
      "loss": 0.2995,
      "step": 669
    },
    {
      "epoch": 1.1,
      "grad_norm": 0.4366275370121002,
      "learning_rate": 0.00013636363636363634,
      "loss": 0.4961,
      "step": 670
    },
    {
      "epoch": 1.1,
      "grad_norm": 0.28639987111091614,
      "learning_rate": 0.00013611570247933883,
      "loss": 0.5015,
      "step": 671
    },
    {
      "epoch": 1.1,
      "grad_norm": 0.2763878107070923,
      "learning_rate": 0.0001358677685950413,
      "loss": 0.4883,
      "step": 672
    },
    {
      "epoch": 1.1,
      "grad_norm": 0.5589582324028015,
      "learning_rate": 0.0001356198347107438,
      "loss": 0.5072,
      "step": 673
    },
    {
      "epoch": 1.1,
      "grad_norm": 0.238887220621109,
      "learning_rate": 0.00013537190082644625,
      "loss": 0.411,
      "step": 674
    },
    {
      "epoch": 1.11,
      "grad_norm": 0.2899521589279175,
      "learning_rate": 0.00013512396694214876,
      "loss": 0.3478,
      "step": 675
    },
    {
      "epoch": 1.11,
      "grad_norm": 0.30960512161254883,
      "learning_rate": 0.00013487603305785124,
      "loss": 0.5058,
      "step": 676
    },
    {
      "epoch": 1.11,
      "grad_norm": 0.33305928111076355,
      "learning_rate": 0.0001346280991735537,
      "loss": 0.4528,
      "step": 677
    },
    {
      "epoch": 1.11,
      "grad_norm": 0.33324292302131653,
      "learning_rate": 0.00013438016528925618,
      "loss": 0.3523,
      "step": 678
    },
    {
      "epoch": 1.11,
      "grad_norm": 0.25855520367622375,
      "learning_rate": 0.00013413223140495867,
      "loss": 0.4257,
      "step": 679
    },
    {
      "epoch": 1.11,
      "grad_norm": 0.36000239849090576,
      "learning_rate": 0.00013388429752066115,
      "loss": 0.4963,
      "step": 680
    },
    {
      "epoch": 1.12,
      "grad_norm": 0.30540961027145386,
      "learning_rate": 0.0001336363636363636,
      "loss": 0.4706,
      "step": 681
    },
    {
      "epoch": 1.12,
      "grad_norm": 0.2791118025779724,
      "learning_rate": 0.00013338842975206612,
      "loss": 0.4543,
      "step": 682
    },
    {
      "epoch": 1.12,
      "grad_norm": 0.37401753664016724,
      "learning_rate": 0.00013314049586776857,
      "loss": 0.5614,
      "step": 683
    },
    {
      "epoch": 1.12,
      "grad_norm": 0.2772528827190399,
      "learning_rate": 0.00013289256198347106,
      "loss": 0.3881,
      "step": 684
    },
    {
      "epoch": 1.12,
      "grad_norm": 0.29219475388526917,
      "learning_rate": 0.00013264462809917354,
      "loss": 0.5418,
      "step": 685
    },
    {
      "epoch": 1.12,
      "grad_norm": 0.3255159258842468,
      "learning_rate": 0.00013239669421487603,
      "loss": 0.4669,
      "step": 686
    },
    {
      "epoch": 1.12,
      "grad_norm": 0.2640572488307953,
      "learning_rate": 0.0001321487603305785,
      "loss": 0.4156,
      "step": 687
    },
    {
      "epoch": 1.13,
      "grad_norm": 0.2618845999240875,
      "learning_rate": 0.00013190082644628097,
      "loss": 0.3537,
      "step": 688
    },
    {
      "epoch": 1.13,
      "grad_norm": 0.27396076917648315,
      "learning_rate": 0.00013165289256198348,
      "loss": 0.4391,
      "step": 689
    },
    {
      "epoch": 1.13,
      "grad_norm": 0.5098498463630676,
      "learning_rate": 0.00013140495867768593,
      "loss": 0.3863,
      "step": 690
    },
    {
      "epoch": 1.13,
      "grad_norm": 0.31764644384384155,
      "learning_rate": 0.00013115702479338842,
      "loss": 0.3874,
      "step": 691
    },
    {
      "epoch": 1.13,
      "grad_norm": 0.28738152980804443,
      "learning_rate": 0.0001309090909090909,
      "loss": 0.3209,
      "step": 692
    },
    {
      "epoch": 1.13,
      "grad_norm": 0.32756757736206055,
      "learning_rate": 0.00013066115702479338,
      "loss": 0.4614,
      "step": 693
    },
    {
      "epoch": 1.14,
      "grad_norm": 0.27650028467178345,
      "learning_rate": 0.00013041322314049584,
      "loss": 0.4717,
      "step": 694
    },
    {
      "epoch": 1.14,
      "grad_norm": 0.33100056648254395,
      "learning_rate": 0.00013016528925619832,
      "loss": 0.4317,
      "step": 695
    },
    {
      "epoch": 1.14,
      "grad_norm": 0.3200342357158661,
      "learning_rate": 0.00012991735537190083,
      "loss": 0.4494,
      "step": 696
    },
    {
      "epoch": 1.14,
      "grad_norm": 0.29615214467048645,
      "learning_rate": 0.0001296694214876033,
      "loss": 0.3786,
      "step": 697
    },
    {
      "epoch": 1.14,
      "grad_norm": 0.278094619512558,
      "learning_rate": 0.00012942148760330577,
      "loss": 0.4484,
      "step": 698
    },
    {
      "epoch": 1.14,
      "grad_norm": 0.32800769805908203,
      "learning_rate": 0.00012917355371900826,
      "loss": 0.4635,
      "step": 699
    },
    {
      "epoch": 1.15,
      "grad_norm": 0.3319619596004486,
      "learning_rate": 0.00012892561983471074,
      "loss": 0.5001,
      "step": 700
    },
    {
      "epoch": 1.15,
      "grad_norm": 0.2818608283996582,
      "learning_rate": 0.0001286776859504132,
      "loss": 0.3536,
      "step": 701
    },
    {
      "epoch": 1.15,
      "grad_norm": 0.28644126653671265,
      "learning_rate": 0.00012842975206611568,
      "loss": 0.4168,
      "step": 702
    },
    {
      "epoch": 1.15,
      "grad_norm": 0.2802482545375824,
      "learning_rate": 0.00012818181818181817,
      "loss": 0.3918,
      "step": 703
    },
    {
      "epoch": 1.15,
      "grad_norm": 0.21232947707176208,
      "learning_rate": 0.00012793388429752065,
      "loss": 0.3218,
      "step": 704
    },
    {
      "epoch": 1.15,
      "grad_norm": 0.36512815952301025,
      "learning_rate": 0.00012768595041322313,
      "loss": 0.4566,
      "step": 705
    },
    {
      "epoch": 1.16,
      "grad_norm": 0.26876160502433777,
      "learning_rate": 0.00012743801652892562,
      "loss": 0.4394,
      "step": 706
    },
    {
      "epoch": 1.16,
      "grad_norm": 0.3757662773132324,
      "learning_rate": 0.0001271900826446281,
      "loss": 0.574,
      "step": 707
    },
    {
      "epoch": 1.16,
      "grad_norm": 0.3161550760269165,
      "learning_rate": 0.00012694214876033056,
      "loss": 0.4524,
      "step": 708
    },
    {
      "epoch": 1.16,
      "grad_norm": 0.31256961822509766,
      "learning_rate": 0.00012669421487603304,
      "loss": 0.4332,
      "step": 709
    },
    {
      "epoch": 1.16,
      "grad_norm": 0.3122079074382782,
      "learning_rate": 0.00012644628099173552,
      "loss": 0.5669,
      "step": 710
    },
    {
      "epoch": 1.16,
      "grad_norm": 0.33779048919677734,
      "learning_rate": 0.000126198347107438,
      "loss": 0.515,
      "step": 711
    },
    {
      "epoch": 1.17,
      "grad_norm": 0.38516169786453247,
      "learning_rate": 0.0001259504132231405,
      "loss": 0.5502,
      "step": 712
    },
    {
      "epoch": 1.17,
      "grad_norm": 0.2803480625152588,
      "learning_rate": 0.00012570247933884297,
      "loss": 0.404,
      "step": 713
    },
    {
      "epoch": 1.17,
      "grad_norm": 0.31674399971961975,
      "learning_rate": 0.00012545454545454543,
      "loss": 0.4403,
      "step": 714
    },
    {
      "epoch": 1.17,
      "grad_norm": 0.3029496669769287,
      "learning_rate": 0.00012520661157024791,
      "loss": 0.372,
      "step": 715
    },
    {
      "epoch": 1.17,
      "grad_norm": 0.22542959451675415,
      "learning_rate": 0.0001249586776859504,
      "loss": 0.355,
      "step": 716
    },
    {
      "epoch": 1.17,
      "grad_norm": 0.32029619812965393,
      "learning_rate": 0.00012471074380165288,
      "loss": 0.4845,
      "step": 717
    },
    {
      "epoch": 1.18,
      "grad_norm": 0.34882861375808716,
      "learning_rate": 0.00012446280991735537,
      "loss": 0.4184,
      "step": 718
    },
    {
      "epoch": 1.18,
      "grad_norm": 0.3319970667362213,
      "learning_rate": 0.00012421487603305785,
      "loss": 0.5733,
      "step": 719
    },
    {
      "epoch": 1.18,
      "grad_norm": 0.2770652770996094,
      "learning_rate": 0.00012396694214876033,
      "loss": 0.4296,
      "step": 720
    },
    {
      "epoch": 1.18,
      "grad_norm": 0.3109978437423706,
      "learning_rate": 0.0001237190082644628,
      "loss": 0.3757,
      "step": 721
    },
    {
      "epoch": 1.18,
      "grad_norm": 0.23606395721435547,
      "learning_rate": 0.00012347107438016527,
      "loss": 0.2713,
      "step": 722
    },
    {
      "epoch": 1.18,
      "grad_norm": 0.304574579000473,
      "learning_rate": 0.00012322314049586776,
      "loss": 0.4451,
      "step": 723
    },
    {
      "epoch": 1.19,
      "grad_norm": 0.31314462423324585,
      "learning_rate": 0.00012297520661157024,
      "loss": 0.493,
      "step": 724
    },
    {
      "epoch": 1.19,
      "grad_norm": 0.32014840841293335,
      "learning_rate": 0.00012272727272727272,
      "loss": 0.3784,
      "step": 725
    },
    {
      "epoch": 1.19,
      "grad_norm": 0.29856279492378235,
      "learning_rate": 0.0001224793388429752,
      "loss": 0.581,
      "step": 726
    },
    {
      "epoch": 1.19,
      "grad_norm": 0.30951863527297974,
      "learning_rate": 0.0001222314049586777,
      "loss": 0.4851,
      "step": 727
    },
    {
      "epoch": 1.19,
      "grad_norm": 0.264663428068161,
      "learning_rate": 0.00012198347107438015,
      "loss": 0.431,
      "step": 728
    },
    {
      "epoch": 1.19,
      "grad_norm": 0.3092226982116699,
      "learning_rate": 0.00012173553719008264,
      "loss": 0.4553,
      "step": 729
    },
    {
      "epoch": 1.2,
      "grad_norm": 0.33568286895751953,
      "learning_rate": 0.00012148760330578511,
      "loss": 0.4894,
      "step": 730
    },
    {
      "epoch": 1.2,
      "grad_norm": 0.2966444492340088,
      "learning_rate": 0.0001212396694214876,
      "loss": 0.3855,
      "step": 731
    },
    {
      "epoch": 1.2,
      "grad_norm": 0.2829122841358185,
      "learning_rate": 0.00012099173553719007,
      "loss": 0.5328,
      "step": 732
    },
    {
      "epoch": 1.2,
      "grad_norm": 0.31785663962364197,
      "learning_rate": 0.00012074380165289255,
      "loss": 0.4142,
      "step": 733
    },
    {
      "epoch": 1.2,
      "grad_norm": 0.2983114719390869,
      "learning_rate": 0.00012049586776859502,
      "loss": 0.4168,
      "step": 734
    },
    {
      "epoch": 1.2,
      "grad_norm": 0.2514868378639221,
      "learning_rate": 0.0001202479338842975,
      "loss": 0.4728,
      "step": 735
    },
    {
      "epoch": 1.21,
      "grad_norm": 0.2959445118904114,
      "learning_rate": 0.00011999999999999999,
      "loss": 0.458,
      "step": 736
    },
    {
      "epoch": 1.21,
      "grad_norm": 0.31830325722694397,
      "learning_rate": 0.00011975206611570247,
      "loss": 0.5035,
      "step": 737
    },
    {
      "epoch": 1.21,
      "grad_norm": 0.31181418895721436,
      "learning_rate": 0.00011950413223140496,
      "loss": 0.3776,
      "step": 738
    },
    {
      "epoch": 1.21,
      "grad_norm": 0.3027549684047699,
      "learning_rate": 0.00011925619834710743,
      "loss": 0.4483,
      "step": 739
    },
    {
      "epoch": 1.21,
      "grad_norm": 0.28026890754699707,
      "learning_rate": 0.00011900826446280991,
      "loss": 0.4236,
      "step": 740
    },
    {
      "epoch": 1.21,
      "grad_norm": 0.29137665033340454,
      "learning_rate": 0.00011876033057851238,
      "loss": 0.3615,
      "step": 741
    },
    {
      "epoch": 1.21,
      "grad_norm": 0.282008558511734,
      "learning_rate": 0.00011851239669421486,
      "loss": 0.4335,
      "step": 742
    },
    {
      "epoch": 1.22,
      "grad_norm": 0.297736793756485,
      "learning_rate": 0.00011826446280991733,
      "loss": 0.4945,
      "step": 743
    },
    {
      "epoch": 1.22,
      "grad_norm": 0.3276868164539337,
      "learning_rate": 0.00011801652892561982,
      "loss": 0.5379,
      "step": 744
    },
    {
      "epoch": 1.22,
      "grad_norm": 0.3510095179080963,
      "learning_rate": 0.00011776859504132231,
      "loss": 0.3589,
      "step": 745
    },
    {
      "epoch": 1.22,
      "grad_norm": 0.29952242970466614,
      "learning_rate": 0.00011752066115702478,
      "loss": 0.3805,
      "step": 746
    },
    {
      "epoch": 1.22,
      "grad_norm": 0.220473513007164,
      "learning_rate": 0.00011727272727272727,
      "loss": 0.3978,
      "step": 747
    },
    {
      "epoch": 1.22,
      "grad_norm": 0.30668944120407104,
      "learning_rate": 0.00011702479338842974,
      "loss": 0.3577,
      "step": 748
    },
    {
      "epoch": 1.23,
      "grad_norm": 0.3152049779891968,
      "learning_rate": 0.00011677685950413222,
      "loss": 0.5186,
      "step": 749
    },
    {
      "epoch": 1.23,
      "grad_norm": 0.17376375198364258,
      "learning_rate": 0.00011652892561983469,
      "loss": 0.32,
      "step": 750
    },
    {
      "epoch": 1.23,
      "grad_norm": 0.32847121357917786,
      "learning_rate": 0.00011628099173553718,
      "loss": 0.5403,
      "step": 751
    },
    {
      "epoch": 1.23,
      "grad_norm": 0.28821662068367004,
      "learning_rate": 0.00011603305785123965,
      "loss": 0.3516,
      "step": 752
    },
    {
      "epoch": 1.23,
      "grad_norm": 0.23324501514434814,
      "learning_rate": 0.00011578512396694214,
      "loss": 0.3398,
      "step": 753
    },
    {
      "epoch": 1.23,
      "grad_norm": 0.2897385060787201,
      "learning_rate": 0.00011553719008264463,
      "loss": 0.3775,
      "step": 754
    },
    {
      "epoch": 1.24,
      "grad_norm": 0.33701419830322266,
      "learning_rate": 0.0001152892561983471,
      "loss": 0.5225,
      "step": 755
    },
    {
      "epoch": 1.24,
      "grad_norm": 0.3228382468223572,
      "learning_rate": 0.00011504132231404958,
      "loss": 0.4384,
      "step": 756
    },
    {
      "epoch": 1.24,
      "grad_norm": 0.24733024835586548,
      "learning_rate": 0.00011479338842975205,
      "loss": 0.2883,
      "step": 757
    },
    {
      "epoch": 1.24,
      "grad_norm": 0.2824367880821228,
      "learning_rate": 0.00011454545454545453,
      "loss": 0.3141,
      "step": 758
    },
    {
      "epoch": 1.24,
      "grad_norm": 0.27844521403312683,
      "learning_rate": 0.000114297520661157,
      "loss": 0.3327,
      "step": 759
    },
    {
      "epoch": 1.24,
      "grad_norm": 0.26114732027053833,
      "learning_rate": 0.0001140495867768595,
      "loss": 0.4071,
      "step": 760
    },
    {
      "epoch": 1.25,
      "grad_norm": 0.34284186363220215,
      "learning_rate": 0.00011380165289256197,
      "loss": 0.4619,
      "step": 761
    },
    {
      "epoch": 1.25,
      "grad_norm": 0.2463303506374359,
      "learning_rate": 0.00011355371900826446,
      "loss": 0.3038,
      "step": 762
    },
    {
      "epoch": 1.25,
      "grad_norm": 0.26452890038490295,
      "learning_rate": 0.00011330578512396693,
      "loss": 0.3603,
      "step": 763
    },
    {
      "epoch": 1.25,
      "grad_norm": 0.27888497710227966,
      "learning_rate": 0.00011305785123966941,
      "loss": 0.5109,
      "step": 764
    },
    {
      "epoch": 1.25,
      "grad_norm": 0.3039766252040863,
      "learning_rate": 0.00011280991735537189,
      "loss": 0.5377,
      "step": 765
    },
    {
      "epoch": 1.25,
      "grad_norm": 0.28995901346206665,
      "learning_rate": 0.00011256198347107436,
      "loss": 0.4797,
      "step": 766
    },
    {
      "epoch": 1.26,
      "grad_norm": 0.3420790135860443,
      "learning_rate": 0.00011231404958677686,
      "loss": 0.5209,
      "step": 767
    },
    {
      "epoch": 1.26,
      "grad_norm": 0.33119046688079834,
      "learning_rate": 0.00011206611570247933,
      "loss": 0.3709,
      "step": 768
    },
    {
      "epoch": 1.26,
      "grad_norm": 0.3408135175704956,
      "learning_rate": 0.00011181818181818181,
      "loss": 0.4389,
      "step": 769
    },
    {
      "epoch": 1.26,
      "grad_norm": 0.29120129346847534,
      "learning_rate": 0.00011157024793388428,
      "loss": 0.4327,
      "step": 770
    },
    {
      "epoch": 1.26,
      "grad_norm": 0.32718029618263245,
      "learning_rate": 0.00011132231404958677,
      "loss": 0.4859,
      "step": 771
    },
    {
      "epoch": 1.26,
      "grad_norm": 0.34422147274017334,
      "learning_rate": 0.00011107438016528924,
      "loss": 0.5184,
      "step": 772
    },
    {
      "epoch": 1.27,
      "grad_norm": 0.330323189496994,
      "learning_rate": 0.00011082644628099172,
      "loss": 0.4322,
      "step": 773
    },
    {
      "epoch": 1.27,
      "grad_norm": 0.3218427002429962,
      "learning_rate": 0.00011057851239669422,
      "loss": 0.4129,
      "step": 774
    },
    {
      "epoch": 1.27,
      "grad_norm": 0.2976725995540619,
      "learning_rate": 0.00011033057851239669,
      "loss": 0.5039,
      "step": 775
    },
    {
      "epoch": 1.27,
      "grad_norm": 0.32841789722442627,
      "learning_rate": 0.00011008264462809917,
      "loss": 0.4718,
      "step": 776
    },
    {
      "epoch": 1.27,
      "grad_norm": 0.32977914810180664,
      "learning_rate": 0.00010983471074380164,
      "loss": 0.4248,
      "step": 777
    },
    {
      "epoch": 1.27,
      "grad_norm": 0.2632751166820526,
      "learning_rate": 0.00010958677685950413,
      "loss": 0.3458,
      "step": 778
    },
    {
      "epoch": 1.28,
      "grad_norm": 0.33028510212898254,
      "learning_rate": 0.0001093388429752066,
      "loss": 0.4884,
      "step": 779
    },
    {
      "epoch": 1.28,
      "grad_norm": 0.30288752913475037,
      "learning_rate": 0.00010909090909090908,
      "loss": 0.3776,
      "step": 780
    },
    {
      "epoch": 1.28,
      "grad_norm": 0.32292476296424866,
      "learning_rate": 0.00010884297520661155,
      "loss": 0.392,
      "step": 781
    },
    {
      "epoch": 1.28,
      "grad_norm": 0.31956765055656433,
      "learning_rate": 0.00010859504132231405,
      "loss": 0.3308,
      "step": 782
    },
    {
      "epoch": 1.28,
      "grad_norm": 0.280553936958313,
      "learning_rate": 0.00010834710743801652,
      "loss": 0.5806,
      "step": 783
    },
    {
      "epoch": 1.28,
      "grad_norm": 0.35859328508377075,
      "learning_rate": 0.000108099173553719,
      "loss": 0.5059,
      "step": 784
    },
    {
      "epoch": 1.29,
      "grad_norm": 0.2944432497024536,
      "learning_rate": 0.00010785123966942148,
      "loss": 0.5132,
      "step": 785
    },
    {
      "epoch": 1.29,
      "grad_norm": 0.27504968643188477,
      "learning_rate": 0.00010760330578512395,
      "loss": 0.3741,
      "step": 786
    },
    {
      "epoch": 1.29,
      "grad_norm": 0.29401764273643494,
      "learning_rate": 0.00010735537190082644,
      "loss": 0.4992,
      "step": 787
    },
    {
      "epoch": 1.29,
      "grad_norm": 0.30569151043891907,
      "learning_rate": 0.00010710743801652891,
      "loss": 0.5029,
      "step": 788
    },
    {
      "epoch": 1.29,
      "grad_norm": 0.28654801845550537,
      "learning_rate": 0.0001068595041322314,
      "loss": 0.4618,
      "step": 789
    },
    {
      "epoch": 1.29,
      "grad_norm": 0.26424363255500793,
      "learning_rate": 0.00010661157024793387,
      "loss": 0.3929,
      "step": 790
    },
    {
      "epoch": 1.3,
      "grad_norm": 0.28117212653160095,
      "learning_rate": 0.00010636363636363636,
      "loss": 0.5116,
      "step": 791
    },
    {
      "epoch": 1.3,
      "grad_norm": 0.28402891755104065,
      "learning_rate": 0.00010611570247933883,
      "loss": 0.3758,
      "step": 792
    },
    {
      "epoch": 1.3,
      "grad_norm": 0.32903602719306946,
      "learning_rate": 0.00010586776859504131,
      "loss": 0.3594,
      "step": 793
    },
    {
      "epoch": 1.3,
      "grad_norm": 0.4285104274749756,
      "learning_rate": 0.0001056198347107438,
      "loss": 0.3007,
      "step": 794
    },
    {
      "epoch": 1.3,
      "grad_norm": 0.27649369835853577,
      "learning_rate": 0.00010537190082644627,
      "loss": 0.342,
      "step": 795
    },
    {
      "epoch": 1.3,
      "grad_norm": 0.3094039261341095,
      "learning_rate": 0.00010512396694214876,
      "loss": 0.4452,
      "step": 796
    },
    {
      "epoch": 1.3,
      "grad_norm": 0.32547199726104736,
      "learning_rate": 0.00010487603305785123,
      "loss": 0.4274,
      "step": 797
    },
    {
      "epoch": 1.31,
      "grad_norm": 0.30244141817092896,
      "learning_rate": 0.00010462809917355372,
      "loss": 0.393,
      "step": 798
    },
    {
      "epoch": 1.31,
      "grad_norm": 0.3018583655357361,
      "learning_rate": 0.00010438016528925619,
      "loss": 0.4012,
      "step": 799
    },
    {
      "epoch": 1.31,
      "grad_norm": 0.36397960782051086,
      "learning_rate": 0.00010413223140495867,
      "loss": 0.5231,
      "step": 800
    },
    {
      "epoch": 1.31,
      "grad_norm": 0.3178517520427704,
      "learning_rate": 0.00010388429752066114,
      "loss": 0.4036,
      "step": 801
    },
    {
      "epoch": 1.31,
      "grad_norm": 0.34640219807624817,
      "learning_rate": 0.00010363636363636362,
      "loss": 0.4717,
      "step": 802
    },
    {
      "epoch": 1.31,
      "grad_norm": 0.302775114774704,
      "learning_rate": 0.0001033884297520661,
      "loss": 0.4207,
      "step": 803
    },
    {
      "epoch": 1.32,
      "grad_norm": 0.30845245718955994,
      "learning_rate": 0.00010314049586776859,
      "loss": 0.3976,
      "step": 804
    },
    {
      "epoch": 1.32,
      "grad_norm": 0.2689266502857208,
      "learning_rate": 0.00010289256198347107,
      "loss": 0.3777,
      "step": 805
    },
    {
      "epoch": 1.32,
      "grad_norm": 0.33539149165153503,
      "learning_rate": 0.00010264462809917354,
      "loss": 0.3896,
      "step": 806
    },
    {
      "epoch": 1.32,
      "grad_norm": 0.2548604905605316,
      "learning_rate": 0.00010239669421487603,
      "loss": 0.4026,
      "step": 807
    },
    {
      "epoch": 1.32,
      "grad_norm": 0.5050720572471619,
      "learning_rate": 0.0001021487603305785,
      "loss": 0.4008,
      "step": 808
    },
    {
      "epoch": 1.32,
      "grad_norm": 0.2518717646598816,
      "learning_rate": 0.00010190082644628098,
      "loss": 0.348,
      "step": 809
    },
    {
      "epoch": 1.33,
      "grad_norm": 0.39397895336151123,
      "learning_rate": 0.00010165289256198345,
      "loss": 0.5369,
      "step": 810
    },
    {
      "epoch": 1.33,
      "grad_norm": 0.3471471965312958,
      "learning_rate": 0.00010140495867768595,
      "loss": 0.5272,
      "step": 811
    },
    {
      "epoch": 1.33,
      "grad_norm": 0.4147883355617523,
      "learning_rate": 0.00010115702479338842,
      "loss": 0.427,
      "step": 812
    },
    {
      "epoch": 1.33,
      "grad_norm": 0.2932160794734955,
      "learning_rate": 0.0001009090909090909,
      "loss": 0.3274,
      "step": 813
    },
    {
      "epoch": 1.33,
      "grad_norm": 0.28647059202194214,
      "learning_rate": 0.00010066115702479339,
      "loss": 0.3346,
      "step": 814
    },
    {
      "epoch": 1.33,
      "grad_norm": 0.28154057264328003,
      "learning_rate": 0.00010041322314049586,
      "loss": 0.3785,
      "step": 815
    },
    {
      "epoch": 1.34,
      "grad_norm": 0.25706711411476135,
      "learning_rate": 0.00010016528925619834,
      "loss": 0.3261,
      "step": 816
    },
    {
      "epoch": 1.34,
      "grad_norm": 0.3318668603897095,
      "learning_rate": 9.991735537190081e-05,
      "loss": 0.4362,
      "step": 817
    },
    {
      "epoch": 1.34,
      "grad_norm": 0.33185282349586487,
      "learning_rate": 9.966942148760331e-05,
      "loss": 0.5219,
      "step": 818
    },
    {
      "epoch": 1.34,
      "grad_norm": 0.2683846056461334,
      "learning_rate": 9.942148760330578e-05,
      "loss": 0.3657,
      "step": 819
    },
    {
      "epoch": 1.34,
      "grad_norm": 0.2643420100212097,
      "learning_rate": 9.917355371900826e-05,
      "loss": 0.4697,
      "step": 820
    },
    {
      "epoch": 1.34,
      "grad_norm": 0.32440856099128723,
      "learning_rate": 9.892561983471073e-05,
      "loss": 0.5572,
      "step": 821
    },
    {
      "epoch": 1.35,
      "grad_norm": 0.22183597087860107,
      "learning_rate": 9.867768595041321e-05,
      "loss": 0.3379,
      "step": 822
    },
    {
      "epoch": 1.35,
      "grad_norm": 0.26266101002693176,
      "learning_rate": 9.842975206611568e-05,
      "loss": 0.439,
      "step": 823
    },
    {
      "epoch": 1.35,
      "grad_norm": 0.2978360950946808,
      "learning_rate": 9.818181818181817e-05,
      "loss": 0.4654,
      "step": 824
    },
    {
      "epoch": 1.35,
      "grad_norm": 0.2713984251022339,
      "learning_rate": 9.793388429752067e-05,
      "loss": 0.2983,
      "step": 825
    },
    {
      "epoch": 1.35,
      "grad_norm": 0.2561984956264496,
      "learning_rate": 9.768595041322314e-05,
      "loss": 0.3381,
      "step": 826
    },
    {
      "epoch": 1.35,
      "grad_norm": 0.2766323983669281,
      "learning_rate": 9.743801652892562e-05,
      "loss": 0.4167,
      "step": 827
    },
    {
      "epoch": 1.36,
      "grad_norm": 0.33810022473335266,
      "learning_rate": 9.719008264462809e-05,
      "loss": 0.3793,
      "step": 828
    },
    {
      "epoch": 1.36,
      "grad_norm": 0.3332251310348511,
      "learning_rate": 9.694214876033057e-05,
      "loss": 0.5517,
      "step": 829
    },
    {
      "epoch": 1.36,
      "grad_norm": 0.2713959515094757,
      "learning_rate": 9.669421487603304e-05,
      "loss": 0.3583,
      "step": 830
    },
    {
      "epoch": 1.36,
      "grad_norm": 0.2778157889842987,
      "learning_rate": 9.644628099173553e-05,
      "loss": 0.3089,
      "step": 831
    },
    {
      "epoch": 1.36,
      "grad_norm": 0.33538392186164856,
      "learning_rate": 9.6198347107438e-05,
      "loss": 0.3776,
      "step": 832
    },
    {
      "epoch": 1.36,
      "grad_norm": 0.32728123664855957,
      "learning_rate": 9.59504132231405e-05,
      "loss": 0.434,
      "step": 833
    },
    {
      "epoch": 1.37,
      "grad_norm": 0.30630162358283997,
      "learning_rate": 9.570247933884298e-05,
      "loss": 0.3913,
      "step": 834
    },
    {
      "epoch": 1.37,
      "grad_norm": 0.2960034906864166,
      "learning_rate": 9.545454545454545e-05,
      "loss": 0.4368,
      "step": 835
    },
    {
      "epoch": 1.37,
      "grad_norm": 0.35711923241615295,
      "learning_rate": 9.520661157024793e-05,
      "loss": 0.399,
      "step": 836
    },
    {
      "epoch": 1.37,
      "grad_norm": 0.30195897817611694,
      "learning_rate": 9.49586776859504e-05,
      "loss": 0.4421,
      "step": 837
    },
    {
      "epoch": 1.37,
      "grad_norm": 0.3220643401145935,
      "learning_rate": 9.471074380165288e-05,
      "loss": 0.3441,
      "step": 838
    },
    {
      "epoch": 1.37,
      "grad_norm": 0.3709239661693573,
      "learning_rate": 9.446280991735535e-05,
      "loss": 0.4095,
      "step": 839
    },
    {
      "epoch": 1.38,
      "grad_norm": 0.40360063314437866,
      "learning_rate": 9.421487603305785e-05,
      "loss": 0.5692,
      "step": 840
    },
    {
      "epoch": 1.38,
      "grad_norm": 0.32428041100502014,
      "learning_rate": 9.396694214876032e-05,
      "loss": 0.4306,
      "step": 841
    },
    {
      "epoch": 1.38,
      "grad_norm": 0.2750518321990967,
      "learning_rate": 9.37190082644628e-05,
      "loss": 0.3905,
      "step": 842
    },
    {
      "epoch": 1.38,
      "grad_norm": 0.331478476524353,
      "learning_rate": 9.347107438016528e-05,
      "loss": 0.6008,
      "step": 843
    },
    {
      "epoch": 1.38,
      "grad_norm": 0.3165242671966553,
      "learning_rate": 9.322314049586776e-05,
      "loss": 0.4624,
      "step": 844
    },
    {
      "epoch": 1.38,
      "grad_norm": 0.26457470655441284,
      "learning_rate": 9.297520661157024e-05,
      "loss": 0.4462,
      "step": 845
    },
    {
      "epoch": 1.39,
      "grad_norm": 0.3557126522064209,
      "learning_rate": 9.272727272727271e-05,
      "loss": 0.5737,
      "step": 846
    },
    {
      "epoch": 1.39,
      "grad_norm": 0.3306926488876343,
      "learning_rate": 9.247933884297521e-05,
      "loss": 0.4597,
      "step": 847
    },
    {
      "epoch": 1.39,
      "grad_norm": 0.24906127154827118,
      "learning_rate": 9.223140495867768e-05,
      "loss": 0.378,
      "step": 848
    },
    {
      "epoch": 1.39,
      "grad_norm": 0.29440054297447205,
      "learning_rate": 9.198347107438016e-05,
      "loss": 0.4562,
      "step": 849
    },
    {
      "epoch": 1.39,
      "grad_norm": 0.34878161549568176,
      "learning_rate": 9.173553719008263e-05,
      "loss": 0.4546,
      "step": 850
    },
    {
      "epoch": 1.39,
      "grad_norm": 0.3725307583808899,
      "learning_rate": 9.148760330578512e-05,
      "loss": 0.4119,
      "step": 851
    },
    {
      "epoch": 1.4,
      "grad_norm": 0.30648747086524963,
      "learning_rate": 9.123966942148759e-05,
      "loss": 0.4428,
      "step": 852
    },
    {
      "epoch": 1.4,
      "grad_norm": 0.2755535840988159,
      "learning_rate": 9.099173553719007e-05,
      "loss": 0.3592,
      "step": 853
    },
    {
      "epoch": 1.4,
      "grad_norm": 0.2802577614784241,
      "learning_rate": 9.074380165289255e-05,
      "loss": 0.472,
      "step": 854
    },
    {
      "epoch": 1.4,
      "grad_norm": 0.28871360421180725,
      "learning_rate": 9.049586776859504e-05,
      "loss": 0.4532,
      "step": 855
    },
    {
      "epoch": 1.4,
      "grad_norm": 0.37071362137794495,
      "learning_rate": 9.024793388429752e-05,
      "loss": 0.3426,
      "step": 856
    },
    {
      "epoch": 1.4,
      "grad_norm": 0.30081430077552795,
      "learning_rate": 8.999999999999999e-05,
      "loss": 0.4069,
      "step": 857
    },
    {
      "epoch": 1.4,
      "grad_norm": 0.3186596930027008,
      "learning_rate": 8.975206611570248e-05,
      "loss": 0.4997,
      "step": 858
    },
    {
      "epoch": 1.41,
      "grad_norm": 0.286479115486145,
      "learning_rate": 8.950413223140495e-05,
      "loss": 0.3902,
      "step": 859
    },
    {
      "epoch": 1.41,
      "grad_norm": 0.3457258939743042,
      "learning_rate": 8.925619834710743e-05,
      "loss": 0.4339,
      "step": 860
    },
    {
      "epoch": 1.41,
      "grad_norm": 0.30513113737106323,
      "learning_rate": 8.90082644628099e-05,
      "loss": 0.3414,
      "step": 861
    },
    {
      "epoch": 1.41,
      "grad_norm": 0.30697953701019287,
      "learning_rate": 8.876033057851238e-05,
      "loss": 0.4657,
      "step": 862
    },
    {
      "epoch": 1.41,
      "grad_norm": 0.3395203649997711,
      "learning_rate": 8.851239669421488e-05,
      "loss": 0.3945,
      "step": 863
    },
    {
      "epoch": 1.41,
      "grad_norm": 0.43322789669036865,
      "learning_rate": 8.826446280991735e-05,
      "loss": 0.5337,
      "step": 864
    },
    {
      "epoch": 1.42,
      "grad_norm": 0.3421814739704132,
      "learning_rate": 8.801652892561983e-05,
      "loss": 0.4481,
      "step": 865
    },
    {
      "epoch": 1.42,
      "grad_norm": 0.24497461318969727,
      "learning_rate": 8.77685950413223e-05,
      "loss": 0.4199,
      "step": 866
    },
    {
      "epoch": 1.42,
      "grad_norm": 0.3835270404815674,
      "learning_rate": 8.752066115702479e-05,
      "loss": 0.5534,
      "step": 867
    },
    {
      "epoch": 1.42,
      "grad_norm": 0.3144569396972656,
      "learning_rate": 8.727272727272726e-05,
      "loss": 0.4563,
      "step": 868
    },
    {
      "epoch": 1.42,
      "grad_norm": 0.2757865786552429,
      "learning_rate": 8.702479338842974e-05,
      "loss": 0.4241,
      "step": 869
    },
    {
      "epoch": 1.42,
      "grad_norm": 0.28413090109825134,
      "learning_rate": 8.677685950413221e-05,
      "loss": 0.3484,
      "step": 870
    },
    {
      "epoch": 1.43,
      "grad_norm": 0.27918362617492676,
      "learning_rate": 8.652892561983471e-05,
      "loss": 0.4133,
      "step": 871
    },
    {
      "epoch": 1.43,
      "grad_norm": 0.3901917040348053,
      "learning_rate": 8.628099173553718e-05,
      "loss": 0.4755,
      "step": 872
    },
    {
      "epoch": 1.43,
      "grad_norm": 0.34810692071914673,
      "learning_rate": 8.603305785123966e-05,
      "loss": 0.4516,
      "step": 873
    },
    {
      "epoch": 1.43,
      "grad_norm": 0.3317393958568573,
      "learning_rate": 8.578512396694215e-05,
      "loss": 0.4995,
      "step": 874
    },
    {
      "epoch": 1.43,
      "grad_norm": 0.26235052943229675,
      "learning_rate": 8.553719008264462e-05,
      "loss": 0.3348,
      "step": 875
    },
    {
      "epoch": 1.43,
      "grad_norm": 0.2735447585582733,
      "learning_rate": 8.52892561983471e-05,
      "loss": 0.2932,
      "step": 876
    },
    {
      "epoch": 1.44,
      "grad_norm": 0.30968329310417175,
      "learning_rate": 8.504132231404957e-05,
      "loss": 0.3783,
      "step": 877
    },
    {
      "epoch": 1.44,
      "grad_norm": 0.30193984508514404,
      "learning_rate": 8.479338842975207e-05,
      "loss": 0.4357,
      "step": 878
    },
    {
      "epoch": 1.44,
      "grad_norm": 0.3407258987426758,
      "learning_rate": 8.454545454545454e-05,
      "loss": 0.4821,
      "step": 879
    },
    {
      "epoch": 1.44,
      "grad_norm": 0.28090009093284607,
      "learning_rate": 8.429752066115702e-05,
      "loss": 0.4158,
      "step": 880
    },
    {
      "epoch": 1.44,
      "grad_norm": 0.2898884415626526,
      "learning_rate": 8.404958677685949e-05,
      "loss": 0.3091,
      "step": 881
    },
    {
      "epoch": 1.44,
      "grad_norm": 0.31658637523651123,
      "learning_rate": 8.380165289256197e-05,
      "loss": 0.3773,
      "step": 882
    },
    {
      "epoch": 1.45,
      "grad_norm": 0.2722189724445343,
      "learning_rate": 8.355371900826446e-05,
      "loss": 0.4483,
      "step": 883
    },
    {
      "epoch": 1.45,
      "grad_norm": 0.23621954023838043,
      "learning_rate": 8.330578512396693e-05,
      "loss": 0.3112,
      "step": 884
    },
    {
      "epoch": 1.45,
      "grad_norm": 0.3659461438655853,
      "learning_rate": 8.305785123966942e-05,
      "loss": 0.4507,
      "step": 885
    },
    {
      "epoch": 1.45,
      "grad_norm": 0.3253099322319031,
      "learning_rate": 8.28099173553719e-05,
      "loss": 0.4854,
      "step": 886
    },
    {
      "epoch": 1.45,
      "grad_norm": 0.3201637864112854,
      "learning_rate": 8.256198347107438e-05,
      "loss": 0.5687,
      "step": 887
    },
    {
      "epoch": 1.45,
      "grad_norm": 0.4112270772457123,
      "learning_rate": 8.231404958677685e-05,
      "loss": 0.3742,
      "step": 888
    },
    {
      "epoch": 1.46,
      "grad_norm": 0.3146194517612457,
      "learning_rate": 8.206611570247933e-05,
      "loss": 0.4869,
      "step": 889
    },
    {
      "epoch": 1.46,
      "grad_norm": 0.34321263432502747,
      "learning_rate": 8.18181818181818e-05,
      "loss": 0.5154,
      "step": 890
    },
    {
      "epoch": 1.46,
      "grad_norm": 0.2986968159675598,
      "learning_rate": 8.157024793388429e-05,
      "loss": 0.647,
      "step": 891
    },
    {
      "epoch": 1.46,
      "grad_norm": 0.3427133858203888,
      "learning_rate": 8.132231404958676e-05,
      "loss": 0.3912,
      "step": 892
    },
    {
      "epoch": 1.46,
      "grad_norm": 0.3434309661388397,
      "learning_rate": 8.107438016528925e-05,
      "loss": 0.51,
      "step": 893
    },
    {
      "epoch": 1.46,
      "grad_norm": 0.32024991512298584,
      "learning_rate": 8.082644628099174e-05,
      "loss": 0.387,
      "step": 894
    },
    {
      "epoch": 1.47,
      "grad_norm": 0.2961815595626831,
      "learning_rate": 8.057851239669421e-05,
      "loss": 0.3909,
      "step": 895
    },
    {
      "epoch": 1.47,
      "grad_norm": 0.3219030201435089,
      "learning_rate": 8.033057851239669e-05,
      "loss": 0.3911,
      "step": 896
    },
    {
      "epoch": 1.47,
      "grad_norm": 0.2776000201702118,
      "learning_rate": 8.008264462809916e-05,
      "loss": 0.3625,
      "step": 897
    },
    {
      "epoch": 1.47,
      "grad_norm": 0.31484290957450867,
      "learning_rate": 7.983471074380164e-05,
      "loss": 0.6162,
      "step": 898
    },
    {
      "epoch": 1.47,
      "grad_norm": 0.2789134085178375,
      "learning_rate": 7.958677685950411e-05,
      "loss": 0.3199,
      "step": 899
    },
    {
      "epoch": 1.47,
      "grad_norm": 0.27821627259254456,
      "learning_rate": 7.933884297520661e-05,
      "loss": 0.4295,
      "step": 900
    },
    {
      "epoch": 1.48,
      "grad_norm": 0.3022254705429077,
      "learning_rate": 7.909090909090908e-05,
      "loss": 0.309,
      "step": 901
    },
    {
      "epoch": 1.48,
      "grad_norm": 0.24830293655395508,
      "learning_rate": 7.884297520661157e-05,
      "loss": 0.3833,
      "step": 902
    },
    {
      "epoch": 1.48,
      "grad_norm": 0.31184327602386475,
      "learning_rate": 7.859504132231405e-05,
      "loss": 0.3715,
      "step": 903
    },
    {
      "epoch": 1.48,
      "grad_norm": 0.2993053197860718,
      "learning_rate": 7.834710743801652e-05,
      "loss": 0.3825,
      "step": 904
    },
    {
      "epoch": 1.48,
      "grad_norm": 0.3385005295276642,
      "learning_rate": 7.8099173553719e-05,
      "loss": 0.4868,
      "step": 905
    },
    {
      "epoch": 1.48,
      "grad_norm": 0.26812323927879333,
      "learning_rate": 7.785123966942147e-05,
      "loss": 0.2925,
      "step": 906
    },
    {
      "epoch": 1.49,
      "grad_norm": 0.3275848925113678,
      "learning_rate": 7.760330578512397e-05,
      "loss": 0.3657,
      "step": 907
    },
    {
      "epoch": 1.49,
      "grad_norm": 0.2972089350223541,
      "learning_rate": 7.735537190082644e-05,
      "loss": 0.4396,
      "step": 908
    },
    {
      "epoch": 1.49,
      "grad_norm": 0.27619728446006775,
      "learning_rate": 7.710743801652892e-05,
      "loss": 0.3946,
      "step": 909
    },
    {
      "epoch": 1.49,
      "grad_norm": 0.30436667799949646,
      "learning_rate": 7.68595041322314e-05,
      "loss": 0.4177,
      "step": 910
    },
    {
      "epoch": 1.49,
      "grad_norm": 0.2652393877506256,
      "learning_rate": 7.661157024793388e-05,
      "loss": 0.3165,
      "step": 911
    },
    {
      "epoch": 1.49,
      "grad_norm": 0.28303712606430054,
      "learning_rate": 7.636363636363635e-05,
      "loss": 0.4829,
      "step": 912
    },
    {
      "epoch": 1.49,
      "grad_norm": 0.33964964747428894,
      "learning_rate": 7.611570247933883e-05,
      "loss": 0.5043,
      "step": 913
    },
    {
      "epoch": 1.5,
      "grad_norm": 0.2591302692890167,
      "learning_rate": 7.586776859504133e-05,
      "loss": 0.3814,
      "step": 914
    },
    {
      "epoch": 1.5,
      "grad_norm": 0.3488747179508209,
      "learning_rate": 7.56198347107438e-05,
      "loss": 0.5233,
      "step": 915
    },
    {
      "epoch": 1.5,
      "grad_norm": 0.29015597701072693,
      "learning_rate": 7.537190082644628e-05,
      "loss": 0.4672,
      "step": 916
    },
    {
      "epoch": 1.5,
      "grad_norm": 0.31618839502334595,
      "learning_rate": 7.512396694214875e-05,
      "loss": 0.4538,
      "step": 917
    },
    {
      "epoch": 1.5,
      "grad_norm": 0.35049545764923096,
      "learning_rate": 7.487603305785124e-05,
      "loss": 0.4089,
      "step": 918
    },
    {
      "epoch": 1.5,
      "grad_norm": 0.34093132615089417,
      "learning_rate": 7.462809917355372e-05,
      "loss": 0.4306,
      "step": 919
    },
    {
      "epoch": 1.51,
      "grad_norm": 0.30601584911346436,
      "learning_rate": 7.438016528925619e-05,
      "loss": 0.4396,
      "step": 920
    },
    {
      "epoch": 1.51,
      "grad_norm": 0.45013612508773804,
      "learning_rate": 7.413223140495867e-05,
      "loss": 0.4477,
      "step": 921
    },
    {
      "epoch": 1.51,
      "grad_norm": 0.30486834049224854,
      "learning_rate": 7.388429752066116e-05,
      "loss": 0.3777,
      "step": 922
    },
    {
      "epoch": 1.51,
      "grad_norm": 0.3926061689853668,
      "learning_rate": 7.363636363636363e-05,
      "loss": 0.3532,
      "step": 923
    },
    {
      "epoch": 1.51,
      "grad_norm": 0.3843371272087097,
      "learning_rate": 7.338842975206611e-05,
      "loss": 0.5182,
      "step": 924
    },
    {
      "epoch": 1.51,
      "grad_norm": 0.30922451615333557,
      "learning_rate": 7.314049586776858e-05,
      "loss": 0.4361,
      "step": 925
    },
    {
      "epoch": 1.52,
      "grad_norm": 0.3367323875427246,
      "learning_rate": 7.289256198347106e-05,
      "loss": 0.3809,
      "step": 926
    },
    {
      "epoch": 1.52,
      "grad_norm": 0.39369019865989685,
      "learning_rate": 7.264462809917355e-05,
      "loss": 0.3623,
      "step": 927
    },
    {
      "epoch": 1.52,
      "grad_norm": 0.3159162104129791,
      "learning_rate": 7.239669421487603e-05,
      "loss": 0.5059,
      "step": 928
    },
    {
      "epoch": 1.52,
      "grad_norm": 0.34716740250587463,
      "learning_rate": 7.214876033057851e-05,
      "loss": 0.4201,
      "step": 929
    },
    {
      "epoch": 1.52,
      "grad_norm": 0.20480923354625702,
      "learning_rate": 7.190082644628098e-05,
      "loss": 0.2699,
      "step": 930
    },
    {
      "epoch": 1.52,
      "grad_norm": 0.3518913686275482,
      "learning_rate": 7.165289256198347e-05,
      "loss": 0.5337,
      "step": 931
    },
    {
      "epoch": 1.53,
      "grad_norm": 0.28605952858924866,
      "learning_rate": 7.140495867768594e-05,
      "loss": 0.44,
      "step": 932
    },
    {
      "epoch": 1.53,
      "grad_norm": 0.28229033946990967,
      "learning_rate": 7.115702479338842e-05,
      "loss": 0.3534,
      "step": 933
    },
    {
      "epoch": 1.53,
      "grad_norm": 0.3456754684448242,
      "learning_rate": 7.09090909090909e-05,
      "loss": 0.3952,
      "step": 934
    },
    {
      "epoch": 1.53,
      "grad_norm": 0.27707159519195557,
      "learning_rate": 7.066115702479338e-05,
      "loss": 0.3667,
      "step": 935
    },
    {
      "epoch": 1.53,
      "grad_norm": 0.2811780273914337,
      "learning_rate": 7.041322314049586e-05,
      "loss": 0.3954,
      "step": 936
    },
    {
      "epoch": 1.53,
      "grad_norm": 0.3099793493747711,
      "learning_rate": 7.016528925619834e-05,
      "loss": 0.441,
      "step": 937
    },
    {
      "epoch": 1.54,
      "grad_norm": 0.4153590500354767,
      "learning_rate": 6.991735537190083e-05,
      "loss": 0.4462,
      "step": 938
    },
    {
      "epoch": 1.54,
      "grad_norm": 0.2945801615715027,
      "learning_rate": 6.96694214876033e-05,
      "loss": 0.4535,
      "step": 939
    },
    {
      "epoch": 1.54,
      "grad_norm": 0.2930592894554138,
      "learning_rate": 6.942148760330578e-05,
      "loss": 0.5566,
      "step": 940
    },
    {
      "epoch": 1.54,
      "grad_norm": 0.3034913241863251,
      "learning_rate": 6.917355371900826e-05,
      "loss": 0.4695,
      "step": 941
    },
    {
      "epoch": 1.54,
      "grad_norm": 0.3054913878440857,
      "learning_rate": 6.892561983471073e-05,
      "loss": 0.3921,
      "step": 942
    },
    {
      "epoch": 1.54,
      "grad_norm": 0.3297981917858124,
      "learning_rate": 6.867768595041322e-05,
      "loss": 0.5057,
      "step": 943
    },
    {
      "epoch": 1.55,
      "grad_norm": 0.23640452325344086,
      "learning_rate": 6.84297520661157e-05,
      "loss": 0.329,
      "step": 944
    },
    {
      "epoch": 1.55,
      "grad_norm": 0.2970188856124878,
      "learning_rate": 6.818181818181817e-05,
      "loss": 0.4376,
      "step": 945
    },
    {
      "epoch": 1.55,
      "grad_norm": 0.3243064880371094,
      "learning_rate": 6.793388429752065e-05,
      "loss": 0.4922,
      "step": 946
    },
    {
      "epoch": 1.55,
      "grad_norm": 0.4473859667778015,
      "learning_rate": 6.768595041322312e-05,
      "loss": 0.5245,
      "step": 947
    },
    {
      "epoch": 1.55,
      "grad_norm": 0.2901310622692108,
      "learning_rate": 6.743801652892562e-05,
      "loss": 0.4996,
      "step": 948
    },
    {
      "epoch": 1.55,
      "grad_norm": 0.3633457124233246,
      "learning_rate": 6.719008264462809e-05,
      "loss": 0.4669,
      "step": 949
    },
    {
      "epoch": 1.56,
      "grad_norm": 0.33570581674575806,
      "learning_rate": 6.694214876033058e-05,
      "loss": 0.404,
      "step": 950
    },
    {
      "epoch": 1.56,
      "grad_norm": 0.26466354727745056,
      "learning_rate": 6.669421487603306e-05,
      "loss": 0.2881,
      "step": 951
    },
    {
      "epoch": 1.56,
      "grad_norm": 0.29028353095054626,
      "learning_rate": 6.644628099173553e-05,
      "loss": 0.3607,
      "step": 952
    },
    {
      "epoch": 1.56,
      "grad_norm": 0.2878669798374176,
      "learning_rate": 6.619834710743801e-05,
      "loss": 0.4415,
      "step": 953
    },
    {
      "epoch": 1.56,
      "grad_norm": 0.33260804414749146,
      "learning_rate": 6.595041322314048e-05,
      "loss": 0.4424,
      "step": 954
    },
    {
      "epoch": 1.56,
      "grad_norm": 0.3135119378566742,
      "learning_rate": 6.570247933884297e-05,
      "loss": 0.4276,
      "step": 955
    },
    {
      "epoch": 1.57,
      "grad_norm": 0.2714795470237732,
      "learning_rate": 6.545454545454545e-05,
      "loss": 0.2789,
      "step": 956
    },
    {
      "epoch": 1.57,
      "grad_norm": 0.3564438819885254,
      "learning_rate": 6.520661157024792e-05,
      "loss": 0.4683,
      "step": 957
    },
    {
      "epoch": 1.57,
      "grad_norm": 0.3303399682044983,
      "learning_rate": 6.495867768595042e-05,
      "loss": 0.4657,
      "step": 958
    },
    {
      "epoch": 1.57,
      "grad_norm": 0.30086350440979004,
      "learning_rate": 6.471074380165289e-05,
      "loss": 0.3296,
      "step": 959
    },
    {
      "epoch": 1.57,
      "grad_norm": 0.34699100255966187,
      "learning_rate": 6.446280991735537e-05,
      "loss": 0.3543,
      "step": 960
    },
    {
      "epoch": 1.57,
      "grad_norm": 0.326579213142395,
      "learning_rate": 6.421487603305784e-05,
      "loss": 0.4001,
      "step": 961
    },
    {
      "epoch": 1.58,
      "grad_norm": 0.3462665379047394,
      "learning_rate": 6.396694214876032e-05,
      "loss": 0.3999,
      "step": 962
    },
    {
      "epoch": 1.58,
      "grad_norm": 0.3408821225166321,
      "learning_rate": 6.371900826446281e-05,
      "loss": 0.3614,
      "step": 963
    },
    {
      "epoch": 1.58,
      "grad_norm": 0.3061428666114807,
      "learning_rate": 6.347107438016528e-05,
      "loss": 0.4127,
      "step": 964
    },
    {
      "epoch": 1.58,
      "grad_norm": 0.30745938420295715,
      "learning_rate": 6.322314049586776e-05,
      "loss": 0.3965,
      "step": 965
    },
    {
      "epoch": 1.58,
      "grad_norm": 0.33782872557640076,
      "learning_rate": 6.297520661157025e-05,
      "loss": 0.5026,
      "step": 966
    },
    {
      "epoch": 1.58,
      "grad_norm": 0.3501698076725006,
      "learning_rate": 6.272727272727272e-05,
      "loss": 0.4731,
      "step": 967
    },
    {
      "epoch": 1.58,
      "grad_norm": 0.3578520119190216,
      "learning_rate": 6.24793388429752e-05,
      "loss": 0.4302,
      "step": 968
    },
    {
      "epoch": 1.59,
      "grad_norm": 0.30132660269737244,
      "learning_rate": 6.223140495867768e-05,
      "loss": 0.3784,
      "step": 969
    },
    {
      "epoch": 1.59,
      "grad_norm": 0.29198774695396423,
      "learning_rate": 6.198347107438017e-05,
      "loss": 0.396,
      "step": 970
    },
    {
      "epoch": 1.59,
      "grad_norm": 0.3028549551963806,
      "learning_rate": 6.173553719008264e-05,
      "loss": 0.3531,
      "step": 971
    },
    {
      "epoch": 1.59,
      "grad_norm": 0.3193860352039337,
      "learning_rate": 6.148760330578512e-05,
      "loss": 0.5261,
      "step": 972
    },
    {
      "epoch": 1.59,
      "grad_norm": 0.330228716135025,
      "learning_rate": 6.12396694214876e-05,
      "loss": 0.3853,
      "step": 973
    },
    {
      "epoch": 1.59,
      "grad_norm": 0.2856347858905792,
      "learning_rate": 6.0991735537190074e-05,
      "loss": 0.4543,
      "step": 974
    },
    {
      "epoch": 1.6,
      "grad_norm": 0.3663886487483978,
      "learning_rate": 6.074380165289256e-05,
      "loss": 0.3821,
      "step": 975
    },
    {
      "epoch": 1.6,
      "grad_norm": 0.3297857642173767,
      "learning_rate": 6.0495867768595034e-05,
      "loss": 0.4504,
      "step": 976
    },
    {
      "epoch": 1.6,
      "grad_norm": 0.29853883385658264,
      "learning_rate": 6.024793388429751e-05,
      "loss": 0.3528,
      "step": 977
    },
    {
      "epoch": 1.6,
      "grad_norm": 0.3246425986289978,
      "learning_rate": 5.9999999999999995e-05,
      "loss": 0.3986,
      "step": 978
    },
    {
      "epoch": 1.6,
      "grad_norm": 0.3537238836288452,
      "learning_rate": 5.975206611570248e-05,
      "loss": 0.3776,
      "step": 979
    },
    {
      "epoch": 1.6,
      "grad_norm": 0.2915757894515991,
      "learning_rate": 5.9504132231404955e-05,
      "loss": 0.2895,
      "step": 980
    },
    {
      "epoch": 1.61,
      "grad_norm": 0.30707284808158875,
      "learning_rate": 5.925619834710743e-05,
      "loss": 0.3238,
      "step": 981
    },
    {
      "epoch": 1.61,
      "grad_norm": 0.301845520734787,
      "learning_rate": 5.900826446280991e-05,
      "loss": 0.4031,
      "step": 982
    },
    {
      "epoch": 1.61,
      "grad_norm": 0.24002347886562347,
      "learning_rate": 5.876033057851239e-05,
      "loss": 0.3477,
      "step": 983
    },
    {
      "epoch": 1.61,
      "grad_norm": 0.3008634150028229,
      "learning_rate": 5.851239669421487e-05,
      "loss": 0.4595,
      "step": 984
    },
    {
      "epoch": 1.61,
      "grad_norm": 0.32416027784347534,
      "learning_rate": 5.8264462809917346e-05,
      "loss": 0.403,
      "step": 985
    },
    {
      "epoch": 1.61,
      "grad_norm": 0.3158760368824005,
      "learning_rate": 5.801652892561982e-05,
      "loss": 0.305,
      "step": 986
    },
    {
      "epoch": 1.62,
      "grad_norm": 0.33743736147880554,
      "learning_rate": 5.7768595041322313e-05,
      "loss": 0.4867,
      "step": 987
    },
    {
      "epoch": 1.62,
      "grad_norm": 0.3402981460094452,
      "learning_rate": 5.752066115702479e-05,
      "loss": 0.3982,
      "step": 988
    },
    {
      "epoch": 1.62,
      "grad_norm": 0.3389660716056824,
      "learning_rate": 5.727272727272727e-05,
      "loss": 0.4311,
      "step": 989
    },
    {
      "epoch": 1.62,
      "grad_norm": 0.47749587893486023,
      "learning_rate": 5.702479338842975e-05,
      "loss": 0.3775,
      "step": 990
    },
    {
      "epoch": 1.62,
      "grad_norm": 0.27538084983825684,
      "learning_rate": 5.677685950413223e-05,
      "loss": 0.3568,
      "step": 991
    },
    {
      "epoch": 1.62,
      "grad_norm": 0.33023789525032043,
      "learning_rate": 5.6528925619834704e-05,
      "loss": 0.4225,
      "step": 992
    },
    {
      "epoch": 1.63,
      "grad_norm": 0.28135445713996887,
      "learning_rate": 5.628099173553718e-05,
      "loss": 0.3658,
      "step": 993
    },
    {
      "epoch": 1.63,
      "grad_norm": 0.3511416018009186,
      "learning_rate": 5.6033057851239665e-05,
      "loss": 0.3928,
      "step": 994
    },
    {
      "epoch": 1.63,
      "grad_norm": 0.2987925708293915,
      "learning_rate": 5.578512396694214e-05,
      "loss": 0.4015,
      "step": 995
    },
    {
      "epoch": 1.63,
      "grad_norm": 0.3340010344982147,
      "learning_rate": 5.553719008264462e-05,
      "loss": 0.4566,
      "step": 996
    },
    {
      "epoch": 1.63,
      "grad_norm": 0.23461014032363892,
      "learning_rate": 5.528925619834711e-05,
      "loss": 0.3556,
      "step": 997
    },
    {
      "epoch": 1.63,
      "grad_norm": 0.3425525724887848,
      "learning_rate": 5.5041322314049586e-05,
      "loss": 0.3736,
      "step": 998
    },
    {
      "epoch": 1.64,
      "grad_norm": 0.33320698142051697,
      "learning_rate": 5.479338842975206e-05,
      "loss": 0.3926,
      "step": 999
    },
    {
      "epoch": 1.64,
      "grad_norm": 0.26936790347099304,
      "learning_rate": 5.454545454545454e-05,
      "loss": 0.3587,
      "step": 1000
    },
    {
      "epoch": 1.64,
      "grad_norm": 0.322934091091156,
      "learning_rate": 5.429752066115702e-05,
      "loss": 0.3119,
      "step": 1001
    },
    {
      "epoch": 1.64,
      "grad_norm": 0.3295484483242035,
      "learning_rate": 5.40495867768595e-05,
      "loss": 0.3257,
      "step": 1002
    },
    {
      "epoch": 1.64,
      "grad_norm": 0.2893584370613098,
      "learning_rate": 5.380165289256198e-05,
      "loss": 0.3451,
      "step": 1003
    },
    {
      "epoch": 1.64,
      "grad_norm": 0.3215138912200928,
      "learning_rate": 5.3553719008264454e-05,
      "loss": 0.4104,
      "step": 1004
    },
    {
      "epoch": 1.65,
      "grad_norm": 0.19545914232730865,
      "learning_rate": 5.330578512396694e-05,
      "loss": 0.2245,
      "step": 1005
    },
    {
      "epoch": 1.65,
      "grad_norm": 0.2952648103237152,
      "learning_rate": 5.3057851239669414e-05,
      "loss": 0.3393,
      "step": 1006
    },
    {
      "epoch": 1.65,
      "grad_norm": 0.34105175733566284,
      "learning_rate": 5.28099173553719e-05,
      "loss": 0.519,
      "step": 1007
    },
    {
      "epoch": 1.65,
      "grad_norm": 0.3435216546058655,
      "learning_rate": 5.256198347107438e-05,
      "loss": 0.4968,
      "step": 1008
    },
    {
      "epoch": 1.65,
      "grad_norm": 0.29052355885505676,
      "learning_rate": 5.231404958677686e-05,
      "loss": 0.4419,
      "step": 1009
    },
    {
      "epoch": 1.65,
      "grad_norm": 0.3326230049133301,
      "learning_rate": 5.2066115702479335e-05,
      "loss": 0.4461,
      "step": 1010
    },
    {
      "epoch": 1.66,
      "grad_norm": 0.35595494508743286,
      "learning_rate": 5.181818181818181e-05,
      "loss": 0.4886,
      "step": 1011
    },
    {
      "epoch": 1.66,
      "grad_norm": 0.3467525541782379,
      "learning_rate": 5.1570247933884295e-05,
      "loss": 0.4671,
      "step": 1012
    },
    {
      "epoch": 1.66,
      "grad_norm": 0.29460448026657104,
      "learning_rate": 5.132231404958677e-05,
      "loss": 0.3872,
      "step": 1013
    },
    {
      "epoch": 1.66,
      "grad_norm": 0.273575097322464,
      "learning_rate": 5.107438016528925e-05,
      "loss": 0.3603,
      "step": 1014
    },
    {
      "epoch": 1.66,
      "grad_norm": 0.3603818416595459,
      "learning_rate": 5.0826446280991726e-05,
      "loss": 0.3539,
      "step": 1015
    },
    {
      "epoch": 1.66,
      "grad_norm": 0.31469517946243286,
      "learning_rate": 5.057851239669421e-05,
      "loss": 0.3988,
      "step": 1016
    },
    {
      "epoch": 1.67,
      "grad_norm": 0.3218969702720642,
      "learning_rate": 5.033057851239669e-05,
      "loss": 0.4366,
      "step": 1017
    },
    {
      "epoch": 1.67,
      "grad_norm": 0.34077420830726624,
      "learning_rate": 5.008264462809917e-05,
      "loss": 0.4248,
      "step": 1018
    },
    {
      "epoch": 1.67,
      "grad_norm": 0.322591096162796,
      "learning_rate": 4.9834710743801654e-05,
      "loss": 0.5081,
      "step": 1019
    },
    {
      "epoch": 1.67,
      "grad_norm": 0.35607361793518066,
      "learning_rate": 4.958677685950413e-05,
      "loss": 0.3596,
      "step": 1020
    },
    {
      "epoch": 1.67,
      "grad_norm": 0.2865798771381378,
      "learning_rate": 4.933884297520661e-05,
      "loss": 0.2703,
      "step": 1021
    },
    {
      "epoch": 1.67,
      "grad_norm": 0.30387502908706665,
      "learning_rate": 4.9090909090909084e-05,
      "loss": 0.3051,
      "step": 1022
    },
    {
      "epoch": 1.67,
      "grad_norm": 0.3474448323249817,
      "learning_rate": 4.884297520661157e-05,
      "loss": 0.2851,
      "step": 1023
    },
    {
      "epoch": 1.68,
      "grad_norm": 0.3696686625480652,
      "learning_rate": 4.8595041322314045e-05,
      "loss": 0.4403,
      "step": 1024
    },
    {
      "epoch": 1.68,
      "grad_norm": 0.33602291345596313,
      "learning_rate": 4.834710743801652e-05,
      "loss": 0.4134,
      "step": 1025
    },
    {
      "epoch": 1.68,
      "grad_norm": 0.27331918478012085,
      "learning_rate": 4.8099173553719e-05,
      "loss": 0.3303,
      "step": 1026
    },
    {
      "epoch": 1.68,
      "grad_norm": 0.3705825209617615,
      "learning_rate": 4.785123966942149e-05,
      "loss": 0.3411,
      "step": 1027
    },
    {
      "epoch": 1.68,
      "grad_norm": 0.4541082978248596,
      "learning_rate": 4.7603305785123966e-05,
      "loss": 0.4263,
      "step": 1028
    },
    {
      "epoch": 1.68,
      "grad_norm": 0.29885897040367126,
      "learning_rate": 4.735537190082644e-05,
      "loss": 0.5602,
      "step": 1029
    },
    {
      "epoch": 1.69,
      "grad_norm": 0.35169675946235657,
      "learning_rate": 4.7107438016528926e-05,
      "loss": 0.4409,
      "step": 1030
    },
    {
      "epoch": 1.69,
      "grad_norm": 0.41590291261672974,
      "learning_rate": 4.68595041322314e-05,
      "loss": 0.4355,
      "step": 1031
    },
    {
      "epoch": 1.69,
      "grad_norm": 0.33613288402557373,
      "learning_rate": 4.661157024793388e-05,
      "loss": 0.4399,
      "step": 1032
    },
    {
      "epoch": 1.69,
      "grad_norm": 0.3519938886165619,
      "learning_rate": 4.6363636363636356e-05,
      "loss": 0.4464,
      "step": 1033
    },
    {
      "epoch": 1.69,
      "grad_norm": 0.2981269359588623,
      "learning_rate": 4.611570247933884e-05,
      "loss": 0.3667,
      "step": 1034
    },
    {
      "epoch": 1.69,
      "grad_norm": 0.32030418515205383,
      "learning_rate": 4.586776859504132e-05,
      "loss": 0.3759,
      "step": 1035
    },
    {
      "epoch": 1.7,
      "grad_norm": 0.39815372228622437,
      "learning_rate": 4.5619834710743794e-05,
      "loss": 0.3259,
      "step": 1036
    },
    {
      "epoch": 1.7,
      "grad_norm": 0.33106112480163574,
      "learning_rate": 4.537190082644628e-05,
      "loss": 0.4985,
      "step": 1037
    },
    {
      "epoch": 1.7,
      "grad_norm": 0.3748137950897217,
      "learning_rate": 4.512396694214876e-05,
      "loss": 0.5177,
      "step": 1038
    },
    {
      "epoch": 1.7,
      "grad_norm": 0.31328514218330383,
      "learning_rate": 4.487603305785124e-05,
      "loss": 0.3406,
      "step": 1039
    },
    {
      "epoch": 1.7,
      "grad_norm": 0.35391247272491455,
      "learning_rate": 4.4628099173553715e-05,
      "loss": 0.4216,
      "step": 1040
    },
    {
      "epoch": 1.7,
      "grad_norm": 0.37352749705314636,
      "learning_rate": 4.438016528925619e-05,
      "loss": 0.4936,
      "step": 1041
    },
    {
      "epoch": 1.71,
      "grad_norm": 0.20523978769779205,
      "learning_rate": 4.4132231404958675e-05,
      "loss": 0.2241,
      "step": 1042
    },
    {
      "epoch": 1.71,
      "grad_norm": 0.26052072644233704,
      "learning_rate": 4.388429752066115e-05,
      "loss": 0.352,
      "step": 1043
    },
    {
      "epoch": 1.71,
      "grad_norm": 0.30189159512519836,
      "learning_rate": 4.363636363636363e-05,
      "loss": 0.3956,
      "step": 1044
    },
    {
      "epoch": 1.71,
      "grad_norm": 0.28206998109817505,
      "learning_rate": 4.3388429752066106e-05,
      "loss": 0.3073,
      "step": 1045
    },
    {
      "epoch": 1.71,
      "grad_norm": 0.3497346341609955,
      "learning_rate": 4.314049586776859e-05,
      "loss": 0.4544,
      "step": 1046
    },
    {
      "epoch": 1.71,
      "grad_norm": 0.31490492820739746,
      "learning_rate": 4.289256198347107e-05,
      "loss": 0.4809,
      "step": 1047
    },
    {
      "epoch": 1.72,
      "grad_norm": 0.26548659801483154,
      "learning_rate": 4.264462809917355e-05,
      "loss": 0.3189,
      "step": 1048
    },
    {
      "epoch": 1.72,
      "grad_norm": 0.40890252590179443,
      "learning_rate": 4.239669421487603e-05,
      "loss": 0.4825,
      "step": 1049
    },
    {
      "epoch": 1.72,
      "grad_norm": 0.392419695854187,
      "learning_rate": 4.214876033057851e-05,
      "loss": 0.3518,
      "step": 1050
    },
    {
      "epoch": 1.72,
      "grad_norm": 0.3267776370048523,
      "learning_rate": 4.190082644628099e-05,
      "loss": 0.5964,
      "step": 1051
    },
    {
      "epoch": 1.72,
      "grad_norm": 0.29872927069664,
      "learning_rate": 4.1652892561983464e-05,
      "loss": 0.3496,
      "step": 1052
    },
    {
      "epoch": 1.72,
      "grad_norm": 0.3140263259410858,
      "learning_rate": 4.140495867768595e-05,
      "loss": 0.3496,
      "step": 1053
    },
    {
      "epoch": 1.73,
      "grad_norm": 0.35923945903778076,
      "learning_rate": 4.1157024793388424e-05,
      "loss": 0.4328,
      "step": 1054
    },
    {
      "epoch": 1.73,
      "grad_norm": 0.24899311363697052,
      "learning_rate": 4.09090909090909e-05,
      "loss": 0.3662,
      "step": 1055
    },
    {
      "epoch": 1.73,
      "grad_norm": 0.300325870513916,
      "learning_rate": 4.066115702479338e-05,
      "loss": 0.3714,
      "step": 1056
    },
    {
      "epoch": 1.73,
      "grad_norm": 0.26927053928375244,
      "learning_rate": 4.041322314049587e-05,
      "loss": 0.3518,
      "step": 1057
    },
    {
      "epoch": 1.73,
      "grad_norm": 0.28170421719551086,
      "learning_rate": 4.0165289256198345e-05,
      "loss": 0.4214,
      "step": 1058
    },
    {
      "epoch": 1.73,
      "grad_norm": 0.3097275197505951,
      "learning_rate": 3.991735537190082e-05,
      "loss": 0.3387,
      "step": 1059
    },
    {
      "epoch": 1.74,
      "grad_norm": 0.36259180307388306,
      "learning_rate": 3.9669421487603306e-05,
      "loss": 0.4968,
      "step": 1060
    },
    {
      "epoch": 1.74,
      "grad_norm": 0.3555668592453003,
      "learning_rate": 3.942148760330578e-05,
      "loss": 0.4415,
      "step": 1061
    },
    {
      "epoch": 1.74,
      "grad_norm": 0.2894740104675293,
      "learning_rate": 3.917355371900826e-05,
      "loss": 0.3911,
      "step": 1062
    },
    {
      "epoch": 1.74,
      "grad_norm": 0.3361656665802002,
      "learning_rate": 3.8925619834710736e-05,
      "loss": 0.4286,
      "step": 1063
    },
    {
      "epoch": 1.74,
      "grad_norm": 0.33269697427749634,
      "learning_rate": 3.867768595041322e-05,
      "loss": 0.5162,
      "step": 1064
    },
    {
      "epoch": 1.74,
      "grad_norm": 0.3324260711669922,
      "learning_rate": 3.84297520661157e-05,
      "loss": 0.4073,
      "step": 1065
    },
    {
      "epoch": 1.75,
      "grad_norm": 0.3037840723991394,
      "learning_rate": 3.8181818181818174e-05,
      "loss": 0.4084,
      "step": 1066
    },
    {
      "epoch": 1.75,
      "grad_norm": 0.29843100905418396,
      "learning_rate": 3.7933884297520664e-05,
      "loss": 0.4028,
      "step": 1067
    },
    {
      "epoch": 1.75,
      "grad_norm": 0.24433061480522156,
      "learning_rate": 3.768595041322314e-05,
      "loss": 0.3769,
      "step": 1068
    },
    {
      "epoch": 1.75,
      "grad_norm": 0.31540754437446594,
      "learning_rate": 3.743801652892562e-05,
      "loss": 0.4006,
      "step": 1069
    },
    {
      "epoch": 1.75,
      "grad_norm": 0.3915780186653137,
      "learning_rate": 3.7190082644628094e-05,
      "loss": 0.3859,
      "step": 1070
    },
    {
      "epoch": 1.75,
      "grad_norm": 0.7843402028083801,
      "learning_rate": 3.694214876033058e-05,
      "loss": 0.4284,
      "step": 1071
    },
    {
      "epoch": 1.76,
      "grad_norm": 0.3000487685203552,
      "learning_rate": 3.6694214876033055e-05,
      "loss": 0.6066,
      "step": 1072
    },
    {
      "epoch": 1.76,
      "grad_norm": 0.2342897206544876,
      "learning_rate": 3.644628099173553e-05,
      "loss": 0.3012,
      "step": 1073
    },
    {
      "epoch": 1.76,
      "grad_norm": 0.3100823760032654,
      "learning_rate": 3.6198347107438015e-05,
      "loss": 0.4236,
      "step": 1074
    },
    {
      "epoch": 1.76,
      "grad_norm": 0.3442421853542328,
      "learning_rate": 3.595041322314049e-05,
      "loss": 0.4716,
      "step": 1075
    },
    {
      "epoch": 1.76,
      "grad_norm": 0.2785506546497345,
      "learning_rate": 3.570247933884297e-05,
      "loss": 0.307,
      "step": 1076
    },
    {
      "epoch": 1.76,
      "grad_norm": 0.333635151386261,
      "learning_rate": 3.545454545454545e-05,
      "loss": 0.4521,
      "step": 1077
    },
    {
      "epoch": 1.77,
      "grad_norm": 0.3365010619163513,
      "learning_rate": 3.520661157024793e-05,
      "loss": 0.4522,
      "step": 1078
    },
    {
      "epoch": 1.77,
      "grad_norm": 0.31510964035987854,
      "learning_rate": 3.495867768595041e-05,
      "loss": 0.4101,
      "step": 1079
    },
    {
      "epoch": 1.77,
      "grad_norm": 0.2939818501472473,
      "learning_rate": 3.471074380165289e-05,
      "loss": 0.378,
      "step": 1080
    },
    {
      "epoch": 1.77,
      "grad_norm": 0.33073171973228455,
      "learning_rate": 3.446280991735537e-05,
      "loss": 0.4319,
      "step": 1081
    },
    {
      "epoch": 1.77,
      "grad_norm": 0.306769460439682,
      "learning_rate": 3.421487603305785e-05,
      "loss": 0.4584,
      "step": 1082
    },
    {
      "epoch": 1.77,
      "grad_norm": 0.3151317536830902,
      "learning_rate": 3.396694214876033e-05,
      "loss": 0.3202,
      "step": 1083
    },
    {
      "epoch": 1.77,
      "grad_norm": 0.313348650932312,
      "learning_rate": 3.371900826446281e-05,
      "loss": 0.4051,
      "step": 1084
    },
    {
      "epoch": 1.78,
      "grad_norm": 0.3377431333065033,
      "learning_rate": 3.347107438016529e-05,
      "loss": 0.3842,
      "step": 1085
    },
    {
      "epoch": 1.78,
      "grad_norm": 0.31378257274627686,
      "learning_rate": 3.3223140495867765e-05,
      "loss": 0.377,
      "step": 1086
    },
    {
      "epoch": 1.78,
      "grad_norm": 0.31627315282821655,
      "learning_rate": 3.297520661157024e-05,
      "loss": 0.4278,
      "step": 1087
    },
    {
      "epoch": 1.78,
      "grad_norm": 0.2957272529602051,
      "learning_rate": 3.2727272727272725e-05,
      "loss": 0.3384,
      "step": 1088
    },
    {
      "epoch": 1.78,
      "grad_norm": 0.3261624872684479,
      "learning_rate": 3.247933884297521e-05,
      "loss": 0.4525,
      "step": 1089
    },
    {
      "epoch": 1.78,
      "grad_norm": 0.28680557012557983,
      "learning_rate": 3.2231404958677685e-05,
      "loss": 0.3627,
      "step": 1090
    },
    {
      "epoch": 1.79,
      "grad_norm": 0.29543063044548035,
      "learning_rate": 3.198347107438016e-05,
      "loss": 0.2922,
      "step": 1091
    },
    {
      "epoch": 1.79,
      "grad_norm": 0.3554795980453491,
      "learning_rate": 3.173553719008264e-05,
      "loss": 0.4692,
      "step": 1092
    },
    {
      "epoch": 1.79,
      "grad_norm": 0.28728747367858887,
      "learning_rate": 3.148760330578512e-05,
      "loss": 0.2595,
      "step": 1093
    },
    {
      "epoch": 1.79,
      "grad_norm": 0.3099517524242401,
      "learning_rate": 3.12396694214876e-05,
      "loss": 0.3912,
      "step": 1094
    },
    {
      "epoch": 1.79,
      "grad_norm": 0.3173176050186157,
      "learning_rate": 3.099173553719008e-05,
      "loss": 0.4186,
      "step": 1095
    },
    {
      "epoch": 1.79,
      "grad_norm": 0.3445116877555847,
      "learning_rate": 3.074380165289256e-05,
      "loss": 0.4966,
      "step": 1096
    },
    {
      "epoch": 1.8,
      "grad_norm": 0.32030245661735535,
      "learning_rate": 3.0495867768595037e-05,
      "loss": 0.4671,
      "step": 1097
    },
    {
      "epoch": 1.8,
      "grad_norm": 0.3321797847747803,
      "learning_rate": 3.0247933884297517e-05,
      "loss": 0.4265,
      "step": 1098
    },
    {
      "epoch": 1.8,
      "grad_norm": 0.36085036396980286,
      "learning_rate": 2.9999999999999997e-05,
      "loss": 0.4324,
      "step": 1099
    },
    {
      "epoch": 1.8,
      "grad_norm": 0.2999497950077057,
      "learning_rate": 2.9752066115702478e-05,
      "loss": 0.3173,
      "step": 1100
    },
    {
      "epoch": 1.8,
      "grad_norm": 0.31063607335090637,
      "learning_rate": 2.9504132231404954e-05,
      "loss": 0.4941,
      "step": 1101
    },
    {
      "epoch": 1.8,
      "grad_norm": 0.2864468991756439,
      "learning_rate": 2.9256198347107435e-05,
      "loss": 0.4309,
      "step": 1102
    },
    {
      "epoch": 1.81,
      "grad_norm": 0.2904879152774811,
      "learning_rate": 2.900826446280991e-05,
      "loss": 0.4782,
      "step": 1103
    },
    {
      "epoch": 1.81,
      "grad_norm": 0.31169822812080383,
      "learning_rate": 2.8760330578512395e-05,
      "loss": 0.4881,
      "step": 1104
    },
    {
      "epoch": 1.81,
      "grad_norm": 0.3462170660495758,
      "learning_rate": 2.8512396694214875e-05,
      "loss": 0.3551,
      "step": 1105
    },
    {
      "epoch": 1.81,
      "grad_norm": 0.3066549301147461,
      "learning_rate": 2.8264462809917352e-05,
      "loss": 0.4522,
      "step": 1106
    },
    {
      "epoch": 1.81,
      "grad_norm": 0.33785369992256165,
      "learning_rate": 2.8016528925619832e-05,
      "loss": 0.3763,
      "step": 1107
    },
    {
      "epoch": 1.81,
      "grad_norm": 0.2975507378578186,
      "learning_rate": 2.776859504132231e-05,
      "loss": 0.3193,
      "step": 1108
    },
    {
      "epoch": 1.82,
      "grad_norm": 0.31934845447540283,
      "learning_rate": 2.7520661157024793e-05,
      "loss": 0.2994,
      "step": 1109
    },
    {
      "epoch": 1.82,
      "grad_norm": 0.29450473189353943,
      "learning_rate": 2.727272727272727e-05,
      "loss": 0.4279,
      "step": 1110
    },
    {
      "epoch": 1.82,
      "grad_norm": 0.3054717779159546,
      "learning_rate": 2.702479338842975e-05,
      "loss": 0.4687,
      "step": 1111
    },
    {
      "epoch": 1.82,
      "grad_norm": 0.32938167452812195,
      "learning_rate": 2.6776859504132227e-05,
      "loss": 0.4815,
      "step": 1112
    },
    {
      "epoch": 1.82,
      "grad_norm": 0.2678495943546295,
      "learning_rate": 2.6528925619834707e-05,
      "loss": 0.3116,
      "step": 1113
    },
    {
      "epoch": 1.82,
      "grad_norm": 0.26357004046440125,
      "learning_rate": 2.628099173553719e-05,
      "loss": 0.3286,
      "step": 1114
    },
    {
      "epoch": 1.83,
      "grad_norm": 0.3359578251838684,
      "learning_rate": 2.6033057851239667e-05,
      "loss": 0.4137,
      "step": 1115
    },
    {
      "epoch": 1.83,
      "grad_norm": 0.3395717442035675,
      "learning_rate": 2.5785123966942148e-05,
      "loss": 0.3812,
      "step": 1116
    },
    {
      "epoch": 1.83,
      "grad_norm": 0.29891693592071533,
      "learning_rate": 2.5537190082644625e-05,
      "loss": 0.2989,
      "step": 1117
    },
    {
      "epoch": 1.83,
      "grad_norm": 0.402649462223053,
      "learning_rate": 2.5289256198347105e-05,
      "loss": 0.4333,
      "step": 1118
    },
    {
      "epoch": 1.83,
      "grad_norm": 0.3397662341594696,
      "learning_rate": 2.5041322314049585e-05,
      "loss": 0.4188,
      "step": 1119
    },
    {
      "epoch": 1.83,
      "grad_norm": 0.33743607997894287,
      "learning_rate": 2.4793388429752065e-05,
      "loss": 0.5309,
      "step": 1120
    },
    {
      "epoch": 1.84,
      "grad_norm": 0.3248274624347687,
      "learning_rate": 2.4545454545454542e-05,
      "loss": 0.3905,
      "step": 1121
    },
    {
      "epoch": 1.84,
      "grad_norm": 0.3567257821559906,
      "learning_rate": 2.4297520661157022e-05,
      "loss": 0.4107,
      "step": 1122
    },
    {
      "epoch": 1.84,
      "grad_norm": 0.4383893311023712,
      "learning_rate": 2.40495867768595e-05,
      "loss": 0.5024,
      "step": 1123
    },
    {
      "epoch": 1.84,
      "grad_norm": 0.2777807414531708,
      "learning_rate": 2.3801652892561983e-05,
      "loss": 0.3289,
      "step": 1124
    },
    {
      "epoch": 1.84,
      "grad_norm": 0.3409118950366974,
      "learning_rate": 2.3553719008264463e-05,
      "loss": 0.5199,
      "step": 1125
    },
    {
      "epoch": 1.84,
      "grad_norm": 0.3060845732688904,
      "learning_rate": 2.330578512396694e-05,
      "loss": 0.412,
      "step": 1126
    },
    {
      "epoch": 1.85,
      "grad_norm": 0.3366425335407257,
      "learning_rate": 2.305785123966942e-05,
      "loss": 0.484,
      "step": 1127
    },
    {
      "epoch": 1.85,
      "grad_norm": 0.36060798168182373,
      "learning_rate": 2.2809917355371897e-05,
      "loss": 0.5543,
      "step": 1128
    },
    {
      "epoch": 1.85,
      "grad_norm": 0.25729015469551086,
      "learning_rate": 2.256198347107438e-05,
      "loss": 0.2763,
      "step": 1129
    },
    {
      "epoch": 1.85,
      "grad_norm": 0.2890430688858032,
      "learning_rate": 2.2314049586776857e-05,
      "loss": 0.3762,
      "step": 1130
    },
    {
      "epoch": 1.85,
      "grad_norm": 0.31579041481018066,
      "learning_rate": 2.2066115702479338e-05,
      "loss": 0.396,
      "step": 1131
    },
    {
      "epoch": 1.85,
      "grad_norm": 0.3136342763900757,
      "learning_rate": 2.1818181818181814e-05,
      "loss": 0.4134,
      "step": 1132
    },
    {
      "epoch": 1.86,
      "grad_norm": 0.37239784002304077,
      "learning_rate": 2.1570247933884295e-05,
      "loss": 0.4666,
      "step": 1133
    },
    {
      "epoch": 1.86,
      "grad_norm": 0.2847795784473419,
      "learning_rate": 2.1322314049586775e-05,
      "loss": 0.3481,
      "step": 1134
    },
    {
      "epoch": 1.86,
      "grad_norm": 0.27870920300483704,
      "learning_rate": 2.1074380165289255e-05,
      "loss": 0.2669,
      "step": 1135
    },
    {
      "epoch": 1.86,
      "grad_norm": 0.2700231969356537,
      "learning_rate": 2.0826446280991732e-05,
      "loss": 0.2798,
      "step": 1136
    },
    {
      "epoch": 1.86,
      "grad_norm": 0.3257925510406494,
      "learning_rate": 2.0578512396694212e-05,
      "loss": 0.4931,
      "step": 1137
    },
    {
      "epoch": 1.86,
      "grad_norm": 0.2964242994785309,
      "learning_rate": 2.033057851239669e-05,
      "loss": 0.429,
      "step": 1138
    },
    {
      "epoch": 1.86,
      "grad_norm": 0.32561832666397095,
      "learning_rate": 2.0082644628099173e-05,
      "loss": 0.3467,
      "step": 1139
    },
    {
      "epoch": 1.87,
      "grad_norm": 0.27957382798194885,
      "learning_rate": 1.9834710743801653e-05,
      "loss": 0.2686,
      "step": 1140
    },
    {
      "epoch": 1.87,
      "grad_norm": 0.3476884663105011,
      "learning_rate": 1.958677685950413e-05,
      "loss": 0.4814,
      "step": 1141
    },
    {
      "epoch": 1.87,
      "grad_norm": 0.2950107753276825,
      "learning_rate": 1.933884297520661e-05,
      "loss": 0.3578,
      "step": 1142
    },
    {
      "epoch": 1.87,
      "grad_norm": 0.30689096450805664,
      "learning_rate": 1.9090909090909087e-05,
      "loss": 0.3725,
      "step": 1143
    },
    {
      "epoch": 1.87,
      "grad_norm": 0.430915504693985,
      "learning_rate": 1.884297520661157e-05,
      "loss": 0.4766,
      "step": 1144
    },
    {
      "epoch": 1.87,
      "grad_norm": 0.3086168169975281,
      "learning_rate": 1.8595041322314047e-05,
      "loss": 0.5506,
      "step": 1145
    },
    {
      "epoch": 1.88,
      "grad_norm": 0.3441203534603119,
      "learning_rate": 1.8347107438016527e-05,
      "loss": 0.4251,
      "step": 1146
    },
    {
      "epoch": 1.88,
      "grad_norm": 0.2828252613544464,
      "learning_rate": 1.8099173553719008e-05,
      "loss": 0.3,
      "step": 1147
    },
    {
      "epoch": 1.88,
      "grad_norm": 0.33563023805618286,
      "learning_rate": 1.7851239669421485e-05,
      "loss": 0.4082,
      "step": 1148
    },
    {
      "epoch": 1.88,
      "grad_norm": 0.33100175857543945,
      "learning_rate": 1.7603305785123965e-05,
      "loss": 0.5853,
      "step": 1149
    },
    {
      "epoch": 1.88,
      "grad_norm": 0.3554556369781494,
      "learning_rate": 1.7355371900826445e-05,
      "loss": 0.5677,
      "step": 1150
    },
    {
      "epoch": 1.88,
      "grad_norm": 0.32995131611824036,
      "learning_rate": 1.7107438016528925e-05,
      "loss": 0.3315,
      "step": 1151
    },
    {
      "epoch": 1.89,
      "grad_norm": 0.3160393238067627,
      "learning_rate": 1.6859504132231405e-05,
      "loss": 0.3632,
      "step": 1152
    },
    {
      "epoch": 1.89,
      "grad_norm": 0.3632807433605194,
      "learning_rate": 1.6611570247933882e-05,
      "loss": 0.4053,
      "step": 1153
    },
    {
      "epoch": 1.89,
      "grad_norm": 0.2931605279445648,
      "learning_rate": 1.6363636363636363e-05,
      "loss": 0.358,
      "step": 1154
    },
    {
      "epoch": 1.89,
      "grad_norm": 0.32687610387802124,
      "learning_rate": 1.6115702479338843e-05,
      "loss": 0.4584,
      "step": 1155
    },
    {
      "epoch": 1.89,
      "grad_norm": 0.3283078074455261,
      "learning_rate": 1.586776859504132e-05,
      "loss": 0.3818,
      "step": 1156
    },
    {
      "epoch": 1.89,
      "grad_norm": 0.31993189454078674,
      "learning_rate": 1.56198347107438e-05,
      "loss": 0.3714,
      "step": 1157
    },
    {
      "epoch": 1.9,
      "grad_norm": 0.2674204409122467,
      "learning_rate": 1.537190082644628e-05,
      "loss": 0.3943,
      "step": 1158
    },
    {
      "epoch": 1.9,
      "grad_norm": 0.3968242406845093,
      "learning_rate": 1.5123966942148759e-05,
      "loss": 0.4465,
      "step": 1159
    },
    {
      "epoch": 1.9,
      "grad_norm": 0.2870213985443115,
      "learning_rate": 1.4876033057851239e-05,
      "loss": 0.3616,
      "step": 1160
    },
    {
      "epoch": 1.9,
      "grad_norm": 0.29502633213996887,
      "learning_rate": 1.4628099173553717e-05,
      "loss": 0.4112,
      "step": 1161
    },
    {
      "epoch": 1.9,
      "grad_norm": 0.36414459347724915,
      "learning_rate": 1.4380165289256198e-05,
      "loss": 0.3928,
      "step": 1162
    },
    {
      "epoch": 1.9,
      "grad_norm": 0.274940550327301,
      "learning_rate": 1.4132231404958676e-05,
      "loss": 0.3971,
      "step": 1163
    },
    {
      "epoch": 1.91,
      "grad_norm": 0.3382115364074707,
      "learning_rate": 1.3884297520661155e-05,
      "loss": 0.339,
      "step": 1164
    },
    {
      "epoch": 1.91,
      "grad_norm": 0.32059189677238464,
      "learning_rate": 1.3636363636363635e-05,
      "loss": 0.4632,
      "step": 1165
    },
    {
      "epoch": 1.91,
      "grad_norm": 0.40788954496383667,
      "learning_rate": 1.3388429752066113e-05,
      "loss": 0.4729,
      "step": 1166
    },
    {
      "epoch": 1.91,
      "grad_norm": 0.4415609836578369,
      "learning_rate": 1.3140495867768595e-05,
      "loss": 0.4311,
      "step": 1167
    },
    {
      "epoch": 1.91,
      "grad_norm": 0.29439279437065125,
      "learning_rate": 1.2892561983471074e-05,
      "loss": 0.3428,
      "step": 1168
    },
    {
      "epoch": 1.91,
      "grad_norm": 0.38421952724456787,
      "learning_rate": 1.2644628099173552e-05,
      "loss": 0.5504,
      "step": 1169
    },
    {
      "epoch": 1.92,
      "grad_norm": 0.2757047116756439,
      "learning_rate": 1.2396694214876033e-05,
      "loss": 0.3488,
      "step": 1170
    },
    {
      "epoch": 1.92,
      "grad_norm": 0.27029332518577576,
      "learning_rate": 1.2148760330578511e-05,
      "loss": 0.3922,
      "step": 1171
    },
    {
      "epoch": 1.92,
      "grad_norm": 0.29828086495399475,
      "learning_rate": 1.1900826446280991e-05,
      "loss": 0.3484,
      "step": 1172
    },
    {
      "epoch": 1.92,
      "grad_norm": 0.3248095214366913,
      "learning_rate": 1.165289256198347e-05,
      "loss": 0.4166,
      "step": 1173
    },
    {
      "epoch": 1.92,
      "grad_norm": 0.3183375895023346,
      "learning_rate": 1.1404958677685948e-05,
      "loss": 0.4207,
      "step": 1174
    },
    {
      "epoch": 1.92,
      "grad_norm": 0.38209760189056396,
      "learning_rate": 1.1157024793388429e-05,
      "loss": 0.4136,
      "step": 1175
    },
    {
      "epoch": 1.93,
      "grad_norm": 0.31191781163215637,
      "learning_rate": 1.0909090909090907e-05,
      "loss": 0.3821,
      "step": 1176
    },
    {
      "epoch": 1.93,
      "grad_norm": 0.3147072494029999,
      "learning_rate": 1.0661157024793387e-05,
      "loss": 0.2973,
      "step": 1177
    },
    {
      "epoch": 1.93,
      "grad_norm": 0.346629798412323,
      "learning_rate": 1.0413223140495866e-05,
      "loss": 0.5924,
      "step": 1178
    },
    {
      "epoch": 1.93,
      "grad_norm": 0.30329591035842896,
      "learning_rate": 1.0165289256198345e-05,
      "loss": 0.4802,
      "step": 1179
    },
    {
      "epoch": 1.93,
      "grad_norm": 0.3608144521713257,
      "learning_rate": 9.917355371900826e-06,
      "loss": 0.4187,
      "step": 1180
    },
    {
      "epoch": 1.93,
      "grad_norm": 0.3330174684524536,
      "learning_rate": 9.669421487603305e-06,
      "loss": 0.4585,
      "step": 1181
    },
    {
      "epoch": 1.94,
      "grad_norm": 0.2880091071128845,
      "learning_rate": 9.421487603305785e-06,
      "loss": 0.3926,
      "step": 1182
    },
    {
      "epoch": 1.94,
      "grad_norm": 0.2711026668548584,
      "learning_rate": 9.173553719008264e-06,
      "loss": 0.3128,
      "step": 1183
    },
    {
      "epoch": 1.94,
      "grad_norm": 0.3472573161125183,
      "learning_rate": 8.925619834710742e-06,
      "loss": 0.3626,
      "step": 1184
    },
    {
      "epoch": 1.94,
      "grad_norm": 0.29903772473335266,
      "learning_rate": 8.677685950413222e-06,
      "loss": 0.3778,
      "step": 1185
    },
    {
      "epoch": 1.94,
      "grad_norm": 0.309654176235199,
      "learning_rate": 8.429752066115703e-06,
      "loss": 0.3965,
      "step": 1186
    },
    {
      "epoch": 1.94,
      "grad_norm": 0.3163444399833679,
      "learning_rate": 8.181818181818181e-06,
      "loss": 0.3207,
      "step": 1187
    },
    {
      "epoch": 1.95,
      "grad_norm": 0.3754628300666809,
      "learning_rate": 7.93388429752066e-06,
      "loss": 0.3954,
      "step": 1188
    },
    {
      "epoch": 1.95,
      "grad_norm": 0.2967177629470825,
      "learning_rate": 7.68595041322314e-06,
      "loss": 0.4092,
      "step": 1189
    },
    {
      "epoch": 1.95,
      "grad_norm": 0.37930914759635925,
      "learning_rate": 7.438016528925619e-06,
      "loss": 0.5038,
      "step": 1190
    },
    {
      "epoch": 1.95,
      "grad_norm": 0.31978312134742737,
      "learning_rate": 7.190082644628099e-06,
      "loss": 0.3039,
      "step": 1191
    },
    {
      "epoch": 1.95,
      "grad_norm": 0.34556475281715393,
      "learning_rate": 6.942148760330577e-06,
      "loss": 0.3749,
      "step": 1192
    },
    {
      "epoch": 1.95,
      "grad_norm": 0.33958449959754944,
      "learning_rate": 6.694214876033057e-06,
      "loss": 0.4974,
      "step": 1193
    },
    {
      "epoch": 1.95,
      "grad_norm": 0.34213709831237793,
      "learning_rate": 6.446280991735537e-06,
      "loss": 0.4874,
      "step": 1194
    },
    {
      "epoch": 1.96,
      "grad_norm": 0.3194979131221771,
      "learning_rate": 6.198347107438016e-06,
      "loss": 0.4415,
      "step": 1195
    },
    {
      "epoch": 1.96,
      "grad_norm": 0.3170003890991211,
      "learning_rate": 5.950413223140496e-06,
      "loss": 0.299,
      "step": 1196
    },
    {
      "epoch": 1.96,
      "grad_norm": 0.35796797275543213,
      "learning_rate": 5.702479338842974e-06,
      "loss": 0.4516,
      "step": 1197
    },
    {
      "epoch": 1.96,
      "grad_norm": 0.36410433053970337,
      "learning_rate": 5.454545454545454e-06,
      "loss": 0.3137,
      "step": 1198
    },
    {
      "epoch": 1.96,
      "grad_norm": 0.27563753724098206,
      "learning_rate": 5.206611570247933e-06,
      "loss": 0.3465,
      "step": 1199
    },
    {
      "epoch": 1.96,
      "grad_norm": 0.3430056869983673,
      "learning_rate": 4.958677685950413e-06,
      "loss": 0.5325,
      "step": 1200
    },
    {
      "epoch": 1.97,
      "grad_norm": 0.3032241463661194,
      "learning_rate": 4.710743801652893e-06,
      "loss": 0.3802,
      "step": 1201
    },
    {
      "epoch": 1.97,
      "grad_norm": 0.3008878231048584,
      "learning_rate": 4.462809917355371e-06,
      "loss": 0.3674,
      "step": 1202
    },
    {
      "epoch": 1.97,
      "grad_norm": 0.34465453028678894,
      "learning_rate": 4.214876033057851e-06,
      "loss": 0.3465,
      "step": 1203
    },
    {
      "epoch": 1.97,
      "grad_norm": 0.3217530846595764,
      "learning_rate": 3.96694214876033e-06,
      "loss": 0.395,
      "step": 1204
    },
    {
      "epoch": 1.97,
      "grad_norm": 0.3256390690803528,
      "learning_rate": 3.7190082644628097e-06,
      "loss": 0.2928,
      "step": 1205
    },
    {
      "epoch": 1.97,
      "grad_norm": 0.404376357793808,
      "learning_rate": 3.4710743801652887e-06,
      "loss": 0.5579,
      "step": 1206
    },
    {
      "epoch": 1.98,
      "grad_norm": 0.2786218822002411,
      "learning_rate": 3.2231404958677685e-06,
      "loss": 0.3842,
      "step": 1207
    },
    {
      "epoch": 1.98,
      "grad_norm": 0.339501291513443,
      "learning_rate": 2.975206611570248e-06,
      "loss": 0.4061,
      "step": 1208
    },
    {
      "epoch": 1.98,
      "grad_norm": 0.3386409878730774,
      "learning_rate": 2.727272727272727e-06,
      "loss": 0.3452,
      "step": 1209
    },
    {
      "epoch": 1.98,
      "grad_norm": 0.36449265480041504,
      "learning_rate": 2.4793388429752066e-06,
      "loss": 0.3769,
      "step": 1210
    },
    {
      "epoch": 1.98,
      "grad_norm": 0.3336932361125946,
      "learning_rate": 2.2314049586776856e-06,
      "loss": 0.4361,
      "step": 1211
    },
    {
      "epoch": 1.98,
      "grad_norm": 0.28075236082077026,
      "learning_rate": 1.983471074380165e-06,
      "loss": 0.3614,
      "step": 1212
    },
    {
      "epoch": 1.99,
      "grad_norm": 0.31337854266166687,
      "learning_rate": 1.7355371900826443e-06,
      "loss": 0.37,
      "step": 1213
    },
    {
      "epoch": 1.99,
      "grad_norm": 0.3034374415874481,
      "learning_rate": 1.487603305785124e-06,
      "loss": 0.274,
      "step": 1214
    },
    {
      "epoch": 1.99,
      "grad_norm": 0.3485061526298523,
      "learning_rate": 1.2396694214876033e-06,
      "loss": 0.425,
      "step": 1215
    },
    {
      "epoch": 1.99,
      "grad_norm": 0.24720066785812378,
      "learning_rate": 9.917355371900825e-07,
      "loss": 0.307,
      "step": 1216
    },
    {
      "epoch": 1.99,
      "grad_norm": 0.2727121412754059,
      "learning_rate": 7.43801652892562e-07,
      "loss": 0.2991,
      "step": 1217
    },
    {
      "epoch": 1.99,
      "grad_norm": 0.33211690187454224,
      "learning_rate": 4.958677685950412e-07,
      "loss": 0.5309,
      "step": 1218
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.328895628452301,
      "learning_rate": 2.479338842975206e-07,
      "loss": 0.3547,
      "step": 1219
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.2642543315887451,
      "learning_rate": 0.0,
      "loss": 0.3047,
      "step": 1220
    }
  ],
  "logging_steps": 1,
  "max_steps": 1220,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 610,
  "total_flos": 2.626577866972938e+18,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}