{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.938897168405365,
  "eval_steps": 168,
  "global_step": 2684,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 1e-05,
      "loss": 1.3745,
      "step": 1
    },
    {
      "epoch": 0.0,
      "eval_loss": 1.6296857595443726,
      "eval_runtime": 2.6662,
      "eval_samples_per_second": 409.572,
      "eval_steps_per_second": 25.88,
      "step": 1
    },
    {
      "epoch": 0.0,
      "learning_rate": 2e-05,
      "loss": 1.42,
      "step": 2
    },
    {
      "epoch": 0.0,
      "learning_rate": 3e-05,
      "loss": 1.3057,
      "step": 3
    },
    {
      "epoch": 0.01,
      "learning_rate": 4e-05,
      "loss": 1.2307,
      "step": 4
    },
    {
      "epoch": 0.01,
      "learning_rate": 5e-05,
      "loss": 1.289,
      "step": 5
    },
    {
      "epoch": 0.01,
      "learning_rate": 6e-05,
      "loss": 1.4111,
      "step": 6
    },
    {
      "epoch": 0.01,
      "learning_rate": 7e-05,
      "loss": 1.3089,
      "step": 7
    },
    {
      "epoch": 0.01,
      "learning_rate": 8e-05,
      "loss": 1.3204,
      "step": 8
    },
    {
      "epoch": 0.01,
      "learning_rate": 9e-05,
      "loss": 1.3575,
      "step": 9
    },
    {
      "epoch": 0.01,
      "learning_rate": 0.0001,
      "loss": 1.3279,
      "step": 10
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.00011000000000000002,
      "loss": 1.3149,
      "step": 11
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.00012,
      "loss": 1.2578,
      "step": 12
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.00013000000000000002,
      "loss": 1.2849,
      "step": 13
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.00014,
      "loss": 1.2971,
      "step": 14
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.00015000000000000001,
      "loss": 1.1473,
      "step": 15
    },
    {
      "epoch": 0.02,
      "learning_rate": 0.00016,
      "loss": 1.1943,
      "step": 16
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.00017,
      "loss": 1.1877,
      "step": 17
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.00018,
      "loss": 1.1984,
      "step": 18
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.00019,
      "loss": 1.2647,
      "step": 19
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.0002,
      "loss": 1.217,
      "step": 20
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.00019999993046535236,
      "loss": 1.0274,
      "step": 21
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.00019999972186150606,
      "loss": 1.2122,
      "step": 22
    },
    {
      "epoch": 0.03,
      "learning_rate": 0.00019999937418875124,
      "loss": 1.1868,
      "step": 23
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.00019999888744757143,
      "loss": 1.2345,
      "step": 24
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.00019999826163864348,
      "loss": 1.2127,
      "step": 25
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.00019999749676283775,
      "loss": 1.2114,
      "step": 26
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.00019999659282121792,
      "loss": 1.2224,
      "step": 27
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0001999955498150411,
      "loss": 1.1517,
      "step": 28
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0001999943677457578,
      "loss": 1.1631,
      "step": 29
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0001999930466150119,
      "loss": 1.0465,
      "step": 30
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0001999915864246407,
      "loss": 1.1847,
      "step": 31
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0001999899871766749,
      "loss": 1.1238,
      "step": 32
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0001999882488733385,
      "loss": 1.1491,
      "step": 33
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.000199986371517049,
      "loss": 1.276,
      "step": 34
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.0001999843551104172,
      "loss": 1.0911,
      "step": 35
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.00019998219965624734,
      "loss": 1.1276,
      "step": 36
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.00019997990515753693,
      "loss": 1.0981,
      "step": 37
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.00019997747161747695,
      "loss": 1.0901,
      "step": 38
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.0001999748990394517,
      "loss": 1.096,
      "step": 39
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.00019997218742703887,
      "loss": 1.122,
      "step": 40
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.00019996933678400946,
      "loss": 1.1132,
      "step": 41
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.00019996634711432786,
      "loss": 1.1498,
      "step": 42
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.00019996321842215173,
      "loss": 1.0708,
      "step": 43
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0001999599507118322,
      "loss": 1.1154,
      "step": 44
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.00019995654398791355,
      "loss": 1.2118,
      "step": 45
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.00019995299825513357,
      "loss": 1.0919,
      "step": 46
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.00019994931351842327,
      "loss": 1.1364,
      "step": 47
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.00019994548978290695,
      "loss": 1.1442,
      "step": 48
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.0001999415270539023,
      "loss": 1.1248,
      "step": 49
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.00019993742533692022,
      "loss": 1.1366,
      "step": 50
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.00019993318463766495,
      "loss": 1.1437,
      "step": 51
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.000199928804962034,
      "loss": 1.1191,
      "step": 52
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0001999242863161182,
      "loss": 1.0786,
      "step": 53
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.00019991962870620153,
      "loss": 1.1951,
      "step": 54
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.00019991483213876134,
      "loss": 1.1321,
      "step": 55
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.00019990989662046818,
      "loss": 1.0876,
      "step": 56
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.0001999048221581858,
      "loss": 1.1794,
      "step": 57
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.00019989960875897126,
      "loss": 1.1796,
      "step": 58
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.00019989425643007476,
      "loss": 1.1165,
      "step": 59
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0001998887651789398,
      "loss": 1.1978,
      "step": 60
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.00019988313501320297,
      "loss": 1.1693,
      "step": 61
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.00019987736594069414,
      "loss": 1.1553,
      "step": 62
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0001998714579694363,
      "loss": 1.1959,
      "step": 63
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.00019986541110764565,
      "loss": 1.1945,
      "step": 64
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.00019985922536373146,
      "loss": 1.121,
      "step": 65
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.00019985290074629627,
      "loss": 1.122,
      "step": 66
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.00019984643726413565,
      "loss": 1.1435,
      "step": 67
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.00019983983492623833,
      "loss": 1.0413,
      "step": 68
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0001998330937417861,
      "loss": 1.078,
      "step": 69
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0001998262137201539,
      "loss": 1.0811,
      "step": 70
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.00019981919487090972,
      "loss": 1.1639,
      "step": 71
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.00019981203720381463,
      "loss": 1.164,
      "step": 72
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.00019980474072882277,
      "loss": 1.1006,
      "step": 73
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.00019979730545608126,
      "loss": 1.1926,
      "step": 74
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.0001997897313959303,
      "loss": 1.1129,
      "step": 75
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.00019978201855890308,
      "loss": 1.1367,
      "step": 76
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.00019977416695572578,
      "loss": 1.1495,
      "step": 77
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0001997661765973176,
      "loss": 1.1567,
      "step": 78
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.00019975804749479062,
      "loss": 1.2102,
      "step": 79
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.00019974977965945,
      "loss": 1.1175,
      "step": 80
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.0001997413731027937,
      "loss": 1.1243,
      "step": 81
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.00019973282783651263,
      "loss": 1.1406,
      "step": 82
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.00019972414387249072,
      "loss": 1.09,
      "step": 83
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.00019971532122280464,
      "loss": 1.0115,
      "step": 84
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.00019970635989972402,
      "loss": 1.0328,
      "step": 85
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.00019969725991571128,
      "loss": 1.1226,
      "step": 86
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.00019968802128342172,
      "loss": 1.0747,
      "step": 87
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.00019967864401570343,
      "loss": 1.119,
      "step": 88
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.00019966912812559732,
      "loss": 1.1125,
      "step": 89
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.00019965947362633708,
      "loss": 1.0734,
      "step": 90
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0001996496805313491,
      "loss": 1.1798,
      "step": 91
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.00019963974885425266,
      "loss": 1.1461,
      "step": 92
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.0001996296786088596,
      "loss": 1.0397,
      "step": 93
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.00019961946980917456,
      "loss": 1.17,
      "step": 94
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.00019960912246939485,
      "loss": 1.0679,
      "step": 95
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.00019959863660391045,
      "loss": 1.0839,
      "step": 96
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.00019958801222730394,
      "loss": 1.0937,
      "step": 97
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00019957724935435063,
      "loss": 1.1668,
      "step": 98
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00019956634800001832,
      "loss": 1.0858,
      "step": 99
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00019955530817946748,
      "loss": 1.0935,
      "step": 100
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00019954412990805107,
      "loss": 1.1046,
      "step": 101
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00019953281320131468,
      "loss": 1.1319,
      "step": 102
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00019952135807499633,
      "loss": 1.1108,
      "step": 103
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.0001995097645450266,
      "loss": 1.0485,
      "step": 104
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.00019949803262752855,
      "loss": 1.0862,
      "step": 105
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.00019948616233881768,
      "loss": 1.268,
      "step": 106
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.00019947415369540189,
      "loss": 1.0926,
      "step": 107
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.0001994620067139815,
      "loss": 1.1427,
      "step": 108
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.00019944972141144928,
      "loss": 1.0754,
      "step": 109
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.00019943729780489027,
      "loss": 1.0044,
      "step": 110
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0001994247359115819,
      "loss": 1.1304,
      "step": 111
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.00019941203574899393,
      "loss": 1.1683,
      "step": 112
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.00019939919733478838,
      "loss": 1.1559,
      "step": 113
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.00019938622068681953,
      "loss": 1.1879,
      "step": 114
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.00019937310582313392,
      "loss": 1.0613,
      "step": 115
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0001993598527619703,
      "loss": 1.1196,
      "step": 116
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0001993464615217596,
      "loss": 1.0762,
      "step": 117
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00019933293212112495,
      "loss": 1.1059,
      "step": 118
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00019931926457888156,
      "loss": 1.0831,
      "step": 119
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00019930545891403678,
      "loss": 1.0552,
      "step": 120
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00019929151514579008,
      "loss": 1.15,
      "step": 121
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00019927743329353295,
      "loss": 1.1038,
      "step": 122
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0001992632133768489,
      "loss": 1.067,
      "step": 123
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.0001992488554155135,
      "loss": 1.1311,
      "step": 124
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00019923435942949426,
      "loss": 1.1402,
      "step": 125
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00019921972543895066,
      "loss": 1.0453,
      "step": 126
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00019920495346423402,
      "loss": 1.1567,
      "step": 127
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00019919004352588767,
      "loss": 1.137,
      "step": 128
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.0001991749956446468,
      "loss": 0.9986,
      "step": 129
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00019915980984143832,
      "loss": 1.083,
      "step": 130
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.00019914448613738106,
      "loss": 1.0619,
      "step": 131
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.00019912902455378556,
      "loss": 1.1294,
      "step": 132
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.00019911342511215414,
      "loss": 1.0965,
      "step": 133
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.00019909768783418086,
      "loss": 1.0216,
      "step": 134
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.00019908181274175138,
      "loss": 1.0081,
      "step": 135
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0001990657998569432,
      "loss": 1.0246,
      "step": 136
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.0001990496492020252,
      "loss": 1.1249,
      "step": 137
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.00019903336079945804,
      "loss": 1.0518,
      "step": 138
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.00019901693467189386,
      "loss": 1.189,
      "step": 139
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.00019900037084217637,
      "loss": 1.1475,
      "step": 140
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0001989836693333408,
      "loss": 1.2259,
      "step": 141
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0001989668301686138,
      "loss": 1.0399,
      "step": 142
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0001989498533714135,
      "loss": 1.128,
      "step": 143
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.00019893273896534936,
      "loss": 1.014,
      "step": 144
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0001989154869742223,
      "loss": 1.1552,
      "step": 145
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.00019889809742202455,
      "loss": 1.1159,
      "step": 146
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0001988805703329396,
      "loss": 1.0218,
      "step": 147
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.00019886290573134228,
      "loss": 1.1723,
      "step": 148
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0001988451036417986,
      "loss": 1.2132,
      "step": 149
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.00019882716408906585,
      "loss": 1.112,
      "step": 150
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0001988090870980924,
      "loss": 1.0856,
      "step": 151
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.0001987908726940178,
      "loss": 1.0951,
      "step": 152
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.00019877252090217271,
      "loss": 1.0218,
      "step": 153
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.00019875403174807882,
      "loss": 1.0552,
      "step": 154
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.00019873540525744887,
      "loss": 1.1481,
      "step": 155
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.00019871664145618657,
      "loss": 1.169,
      "step": 156
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.00019869774037038665,
      "loss": 1.0802,
      "step": 157
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0001986787020263347,
      "loss": 1.0871,
      "step": 158
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0001986595264505072,
      "loss": 1.1022,
      "step": 159
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.00019864021366957147,
      "loss": 1.0257,
      "step": 160
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.0001986207637103857,
      "loss": 1.0986,
      "step": 161
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.00019860117659999878,
      "loss": 1.0837,
      "step": 162
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.00019858145236565037,
      "loss": 1.1895,
      "step": 163
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.00019856159103477086,
      "loss": 1.052,
      "step": 164
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.00019854159263498123,
      "loss": 1.1184,
      "step": 165
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0001985214571940931,
      "loss": 1.0895,
      "step": 166
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.00019850118474010872,
      "loss": 1.0764,
      "step": 167
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.00019848077530122083,
      "loss": 1.1387,
      "step": 168
    },
    {
      "epoch": 0.25,
      "eval_loss": 1.084919810295105,
      "eval_runtime": 2.6029,
      "eval_samples_per_second": 419.538,
      "eval_steps_per_second": 26.509,
      "step": 168
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.00019846022890581267,
      "loss": 1.0826,
      "step": 169
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.000198439545582458,
      "loss": 1.1366,
      "step": 170
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.000198418725359921,
      "loss": 1.1349,
      "step": 171
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.00019839776826715614,
      "loss": 1.0636,
      "step": 172
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.00019837667433330838,
      "loss": 1.1216,
      "step": 173
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0001983554435877128,
      "loss": 1.1051,
      "step": 174
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.00019833407605989494,
      "loss": 1.1558,
      "step": 175
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.00019831257177957044,
      "loss": 1.0364,
      "step": 176
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.00019829093077664513,
      "loss": 1.0665,
      "step": 177
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.00019826915308121504,
      "loss": 1.1994,
      "step": 178
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0001982472387235662,
      "loss": 1.1434,
      "step": 179
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0001982251877341748,
      "loss": 1.081,
      "step": 180
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.000198203000143707,
      "loss": 1.0653,
      "step": 181
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0001981806759830189,
      "loss": 1.0269,
      "step": 182
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0001981582152831566,
      "loss": 1.1167,
      "step": 183
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.00019813561807535598,
      "loss": 1.0608,
      "step": 184
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0001981128843910428,
      "loss": 1.0989,
      "step": 185
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0001980900142618327,
      "loss": 1.1405,
      "step": 186
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.00019806700771953097,
      "loss": 1.0359,
      "step": 187
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0001980438647961327,
      "loss": 1.1073,
      "step": 188
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0001980205855238225,
      "loss": 1.0338,
      "step": 189
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.00019799716993497475,
      "loss": 1.1285,
      "step": 190
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.00019797361806215332,
      "loss": 1.1277,
      "step": 191
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00019794992993811165,
      "loss": 1.119,
      "step": 192
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00019792610559579265,
      "loss": 1.1224,
      "step": 193
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00019790214506832868,
      "loss": 1.1438,
      "step": 194
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.0001978780483890414,
      "loss": 1.1462,
      "step": 195
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00019785381559144196,
      "loss": 1.042,
      "step": 196
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00019782944670923076,
      "loss": 1.1022,
      "step": 197
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00019780494177629735,
      "loss": 1.0564,
      "step": 198
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00019778030082672068,
      "loss": 1.0471,
      "step": 199
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00019775552389476864,
      "loss": 1.0636,
      "step": 200
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.0001977306110148984,
      "loss": 1.0917,
      "step": 201
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00019770556222175608,
      "loss": 1.1965,
      "step": 202
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00019768037755017685,
      "loss": 1.073,
      "step": 203
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00019765505703518496,
      "loss": 1.0636,
      "step": 204
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00019762960071199333,
      "loss": 1.087,
      "step": 205
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.000197604008616004,
      "loss": 1.0569,
      "step": 206
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00019757828078280766,
      "loss": 1.08,
      "step": 207
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00019755241724818387,
      "loss": 1.1536,
      "step": 208
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00019752641804810084,
      "loss": 1.1514,
      "step": 209
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00019750028321871546,
      "loss": 1.0691,
      "step": 210
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.00019747401279637325,
      "loss": 1.1289,
      "step": 211
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.00019744760681760832,
      "loss": 1.0834,
      "step": 212
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.00019742106531914328,
      "loss": 1.0762,
      "step": 213
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.0001973943883378892,
      "loss": 1.0913,
      "step": 214
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.00019736757591094558,
      "loss": 1.132,
      "step": 215
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.00019734062807560027,
      "loss": 1.0894,
      "step": 216
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.00019731354486932944,
      "loss": 1.0327,
      "step": 217
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.00019728632632979746,
      "loss": 1.112,
      "step": 218
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00019725897249485704,
      "loss": 1.0718,
      "step": 219
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00019723148340254892,
      "loss": 1.077,
      "step": 220
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00019720385909110198,
      "loss": 1.0335,
      "step": 221
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00019717609959893318,
      "loss": 1.0483,
      "step": 222
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00019714820496464746,
      "loss": 1.0901,
      "step": 223
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00019712017522703764,
      "loss": 0.9921,
      "step": 224
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00019709201042508455,
      "loss": 1.0829,
      "step": 225
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.0001970637105979567,
      "loss": 1.0705,
      "step": 226
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.0001970352757850105,
      "loss": 1.0481,
      "step": 227
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00019700670602579008,
      "loss": 0.9846,
      "step": 228
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.0001969780013600272,
      "loss": 1.1492,
      "step": 229
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00019694916182764113,
      "loss": 1.1745,
      "step": 230
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00019692018746873892,
      "loss": 1.0451,
      "step": 231
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00019689107832361496,
      "loss": 1.1217,
      "step": 232
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00019686183443275116,
      "loss": 1.0788,
      "step": 233
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00019683245583681675,
      "loss": 1.0703,
      "step": 234
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00019680294257666837,
      "loss": 1.1521,
      "step": 235
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.0001967732946933499,
      "loss": 1.0659,
      "step": 236
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00019674351222809242,
      "loss": 1.0321,
      "step": 237
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.0001967135952223142,
      "loss": 1.0555,
      "step": 238
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00019668354371762066,
      "loss": 1.0648,
      "step": 239
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00019665335775580415,
      "loss": 1.0723,
      "step": 240
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.0001966230373788441,
      "loss": 1.0264,
      "step": 241
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00019659258262890683,
      "loss": 1.0331,
      "step": 242
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00019656199354834558,
      "loss": 1.1514,
      "step": 243
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00019653127017970034,
      "loss": 1.069,
      "step": 244
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00019650041256569792,
      "loss": 0.9623,
      "step": 245
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00019646942074925172,
      "loss": 1.0021,
      "step": 246
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00019643829477346188,
      "loss": 1.1131,
      "step": 247
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.0001964070346816151,
      "loss": 1.1426,
      "step": 248
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.0001963756405171845,
      "loss": 1.0761,
      "step": 249
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00019634411232382978,
      "loss": 1.1112,
      "step": 250
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00019631245014539698,
      "loss": 1.081,
      "step": 251
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00019628065402591845,
      "loss": 1.1446,
      "step": 252
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00019624872400961284,
      "loss": 1.045,
      "step": 253
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00019621666014088494,
      "loss": 1.0337,
      "step": 254
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00019618446246432583,
      "loss": 1.1764,
      "step": 255
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00019615213102471257,
      "loss": 1.0323,
      "step": 256
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00019611966586700823,
      "loss": 1.0073,
      "step": 257
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00019608706703636188,
      "loss": 1.1615,
      "step": 258
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00019605433457810855,
      "loss": 1.1209,
      "step": 259
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00019602146853776894,
      "loss": 1.0721,
      "step": 260
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.0001959884689610497,
      "loss": 1.0967,
      "step": 261
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00019595533589384308,
      "loss": 1.0284,
      "step": 262
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00019592206938222703,
      "loss": 1.0148,
      "step": 263
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00019588866947246498,
      "loss": 1.1434,
      "step": 264
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00019585513621100603,
      "loss": 1.1125,
      "step": 265
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.0001958214696444846,
      "loss": 1.0812,
      "step": 266
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00019578766981972058,
      "loss": 1.0611,
      "step": 267
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00019575373678371909,
      "loss": 1.1029,
      "step": 268
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00019571967058367064,
      "loss": 1.0692,
      "step": 269
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00019568547126695083,
      "loss": 1.0581,
      "step": 270
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00019565113888112036,
      "loss": 0.9841,
      "step": 271
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00019561667347392508,
      "loss": 1.0173,
      "step": 272
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00019558207509329584,
      "loss": 1.0805,
      "step": 273
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00019554734378734824,
      "loss": 1.088,
      "step": 274
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00019551247960438296,
      "loss": 1.0481,
      "step": 275
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00019547748259288536,
      "loss": 1.1747,
      "step": 276
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.0001954423528015255,
      "loss": 1.0407,
      "step": 277
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00019540709027915818,
      "loss": 1.1412,
      "step": 278
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0001953716950748227,
      "loss": 1.075,
      "step": 279
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00019533616723774294,
      "loss": 0.9863,
      "step": 280
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0001953005068173272,
      "loss": 1.1426,
      "step": 281
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0001952647138631682,
      "loss": 1.0621,
      "step": 282
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00019522878842504295,
      "loss": 1.1007,
      "step": 283
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00019519273055291266,
      "loss": 1.0632,
      "step": 284
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00019515654029692278,
      "loss": 1.126,
      "step": 285
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.00019512021770740288,
      "loss": 1.0946,
      "step": 286
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0001950837628348665,
      "loss": 1.0639,
      "step": 287
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.00019504717573001117,
      "loss": 1.1432,
      "step": 288
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.00019501045644371832,
      "loss": 1.0619,
      "step": 289
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.0001949736050270532,
      "loss": 1.0597,
      "step": 290
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.00019493662153126481,
      "loss": 1.0743,
      "step": 291
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0001948995060077859,
      "loss": 1.1114,
      "step": 292
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.00019486225850823266,
      "loss": 1.1435,
      "step": 293
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.000194824879084405,
      "loss": 1.1396,
      "step": 294
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.00019478736778828624,
      "loss": 1.1597,
      "step": 295
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.00019474972467204297,
      "loss": 1.0976,
      "step": 296
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.00019471194978802533,
      "loss": 1.0829,
      "step": 297
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.0001946740431887665,
      "loss": 1.0437,
      "step": 298
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00019463600492698296,
      "loss": 1.0835,
      "step": 299
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00019459783505557424,
      "loss": 1.0558,
      "step": 300
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0001945595336276229,
      "loss": 1.0656,
      "step": 301
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00019452110069639452,
      "loss": 1.1487,
      "step": 302
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00019448253631533744,
      "loss": 1.1383,
      "step": 303
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00019444384053808288,
      "loss": 1.1582,
      "step": 304
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00019440501341844483,
      "loss": 0.9999,
      "step": 305
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.00019436605501041987,
      "loss": 1.1317,
      "step": 306
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.00019432696536818717,
      "loss": 1.0944,
      "step": 307
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.00019428774454610843,
      "loss": 1.1624,
      "step": 308
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.00019424839259872778,
      "loss": 1.1644,
      "step": 309
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.00019420890958077167,
      "loss": 1.0486,
      "step": 310
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.00019416929554714888,
      "loss": 1.0705,
      "step": 311
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.00019412955055295034,
      "loss": 1.023,
      "step": 312
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00019408967465344917,
      "loss": 1.1144,
      "step": 313
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00019404966790410047,
      "loss": 1.0378,
      "step": 314
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00019400953036054138,
      "loss": 1.036,
      "step": 315
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00019396926207859084,
      "loss": 1.0735,
      "step": 316
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00019392886311424973,
      "loss": 1.0259,
      "step": 317
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0001938883335237006,
      "loss": 1.1603,
      "step": 318
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.0001938476733633076,
      "loss": 1.1282,
      "step": 319
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.0001938068826896166,
      "loss": 1.063,
      "step": 320
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00019376596155935486,
      "loss": 1.1176,
      "step": 321
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00019372491002943112,
      "loss": 1.1307,
      "step": 322
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00019368372815693549,
      "loss": 1.0412,
      "step": 323
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00019364241599913924,
      "loss": 1.1353,
      "step": 324
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00019360097361349494,
      "loss": 1.1293,
      "step": 325
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.0001935594010576362,
      "loss": 1.0885,
      "step": 326
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00019351769838937775,
      "loss": 1.0944,
      "step": 327
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00019347586566671512,
      "loss": 1.1435,
      "step": 328
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.0001934339029478248,
      "loss": 1.1217,
      "step": 329
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00019339181029106404,
      "loss": 1.1801,
      "step": 330
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00019334958775497083,
      "loss": 1.1846,
      "step": 331
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00019330723539826375,
      "loss": 1.0897,
      "step": 332
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00019326475327984192,
      "loss": 1.0643,
      "step": 333
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00019322214145878487,
      "loss": 1.0246,
      "step": 334
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.0001931793999943526,
      "loss": 1.1108,
      "step": 335
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00019313652894598543,
      "loss": 1.0619,
      "step": 336
    },
    {
      "epoch": 0.5,
      "eval_loss": 1.048388123512268,
      "eval_runtime": 2.6045,
      "eval_samples_per_second": 419.273,
      "eval_steps_per_second": 26.493,
      "step": 336
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00019309352837330372,
      "loss": 1.0014,
      "step": 337
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.0001930503983361081,
      "loss": 1.0786,
      "step": 338
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00019300713889437926,
      "loss": 1.014,
      "step": 339
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00019296375010827773,
      "loss": 1.1233,
      "step": 340
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.000192920232038144,
      "loss": 1.1052,
      "step": 341
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0001928765847444984,
      "loss": 1.0138,
      "step": 342
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00019283280828804081,
      "loss": 1.1536,
      "step": 343
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00019278890272965096,
      "loss": 0.992,
      "step": 344
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0001927448681303879,
      "loss": 1.1165,
      "step": 345
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.0001927007045514903,
      "loss": 1.0565,
      "step": 346
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00019265641205437611,
      "loss": 1.0664,
      "step": 347
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.0001926119907006426,
      "loss": 1.0625,
      "step": 348
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00019256744055206622,
      "loss": 1.0393,
      "step": 349
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.0001925227616706026,
      "loss": 1.125,
      "step": 350
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00019247795411838627,
      "loss": 1.0375,
      "step": 351
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00019243301795773086,
      "loss": 1.0648,
      "step": 352
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0001923879532511287,
      "loss": 1.0903,
      "step": 353
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.000192342760061251,
      "loss": 1.1219,
      "step": 354
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00019229743845094755,
      "loss": 1.054,
      "step": 355
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0001922519884832469,
      "loss": 1.1206,
      "step": 356
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00019220641022135588,
      "loss": 1.1125,
      "step": 357
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00019216070372865996,
      "loss": 1.064,
      "step": 358
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0001921148690687228,
      "loss": 1.0843,
      "step": 359
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.00019206890630528634,
      "loss": 1.1378,
      "step": 360
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.00019202281550227064,
      "loss": 1.0399,
      "step": 361
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0001919765967237739,
      "loss": 1.1762,
      "step": 362
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0001919302500340722,
      "loss": 1.0538,
      "step": 363
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.00019188377549761963,
      "loss": 1.0343,
      "step": 364
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.0001918371731790479,
      "loss": 1.1027,
      "step": 365
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00019179044314316664,
      "loss": 1.036,
      "step": 366
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00019174358545496288,
      "loss": 1.041,
      "step": 367
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00019169660017960137,
      "loss": 1.0762,
      "step": 368
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00019164948738242409,
      "loss": 1.0807,
      "step": 369
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00019160224712895055,
      "loss": 1.037,
      "step": 370
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.00019155487948487748,
      "loss": 1.0625,
      "step": 371
    },
    {
      "epoch": 0.55,
      "learning_rate": 0.0001915073845160786,
      "loss": 1.062,
      "step": 372
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.00019145976228860496,
      "loss": 1.1882,
      "step": 373
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.00019141201286868435,
      "loss": 1.1338,
      "step": 374
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.00019136413632272163,
      "loss": 1.0174,
      "step": 375
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.00019131613271729833,
      "loss": 1.0585,
      "step": 376
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.00019126800211917276,
      "loss": 1.0495,
      "step": 377
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0001912197445952798,
      "loss": 1.123,
      "step": 378
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.00019117136021273075,
      "loss": 1.0517,
      "step": 379
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0001911228490388136,
      "loss": 1.0545,
      "step": 380
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.00019107421114099237,
      "loss": 1.0302,
      "step": 381
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.00019102544658690748,
      "loss": 1.0908,
      "step": 382
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.00019097655544437545,
      "loss": 1.1425,
      "step": 383
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.00019092753778138886,
      "loss": 1.0686,
      "step": 384
    },
    {
      "epoch": 0.57,
      "learning_rate": 0.0001908783936661162,
      "loss": 1.06,
      "step": 385
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0001908291231669019,
      "loss": 1.1296,
      "step": 386
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00019077972635226604,
      "loss": 1.1029,
      "step": 387
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00019073020329090444,
      "loss": 1.0469,
      "step": 388
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0001906805540516885,
      "loss": 1.0427,
      "step": 389
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.000190630778703665,
      "loss": 1.0075,
      "step": 390
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.00019058087731605624,
      "loss": 1.1146,
      "step": 391
    },
    {
      "epoch": 0.58,
      "learning_rate": 0.0001905308499582597,
      "loss": 1.1161,
      "step": 392
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00019048069669984802,
      "loss": 1.1419,
      "step": 393
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00019043041761056907,
      "loss": 1.1586,
      "step": 394
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00019038001276034557,
      "loss": 1.0765,
      "step": 395
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00019032948221927524,
      "loss": 1.1225,
      "step": 396
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.0001902788260576305,
      "loss": 1.0247,
      "step": 397
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.00019022804434585852,
      "loss": 1.135,
      "step": 398
    },
    {
      "epoch": 0.59,
      "learning_rate": 0.0001901771371545811,
      "loss": 1.1122,
      "step": 399
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00019012610455459446,
      "loss": 1.075,
      "step": 400
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00019007494661686935,
      "loss": 1.1121,
      "step": 401
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.0001900236634125507,
      "loss": 1.0531,
      "step": 402
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00018997225501295772,
      "loss": 1.0561,
      "step": 403
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00018992072148958368,
      "loss": 1.0803,
      "step": 404
    },
    {
      "epoch": 0.6,
      "learning_rate": 0.00018986906291409595,
      "loss": 1.0579,
      "step": 405
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00018981727935833567,
      "loss": 1.0614,
      "step": 406
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.0001897653708943179,
      "loss": 0.9982,
      "step": 407
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00018971333759423142,
      "loss": 1.1498,
      "step": 408
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00018966117953043852,
      "loss": 1.1165,
      "step": 409
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00018960889677547505,
      "loss": 1.1155,
      "step": 410
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00018955648940205028,
      "loss": 1.0017,
      "step": 411
    },
    {
      "epoch": 0.61,
      "learning_rate": 0.00018950395748304678,
      "loss": 1.0556,
      "step": 412
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.00018945130109152033,
      "loss": 1.0248,
      "step": 413
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.00018939852030069981,
      "loss": 1.0155,
      "step": 414
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.00018934561518398706,
      "loss": 1.0248,
      "step": 415
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.00018929258581495685,
      "loss": 0.9835,
      "step": 416
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.0001892394322673568,
      "loss": 1.1602,
      "step": 417
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.0001891861546151071,
      "loss": 1.021,
      "step": 418
    },
    {
      "epoch": 0.62,
      "learning_rate": 0.00018913275293230069,
      "loss": 1.0526,
      "step": 419
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.00018907922729320285,
      "loss": 1.0585,
      "step": 420
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.00018902557777225135,
      "loss": 1.0327,
      "step": 421
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.00018897180444405614,
      "loss": 1.0448,
      "step": 422
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.0001889179073833995,
      "loss": 1.0776,
      "step": 423
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.0001888638866652356,
      "loss": 1.0748,
      "step": 424
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.0001888097423646907,
      "loss": 1.0482,
      "step": 425
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.00018875547455706295,
      "loss": 1.0394,
      "step": 426
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.00018870108331782217,
      "loss": 1.0646,
      "step": 427
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.00018864656872260985,
      "loss": 1.0338,
      "step": 428
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.00018859193084723913,
      "loss": 0.9848,
      "step": 429
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.0001885371697676944,
      "loss": 1.0587,
      "step": 430
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.0001884822855601316,
      "loss": 1.0711,
      "step": 431
    },
    {
      "epoch": 0.64,
      "learning_rate": 0.00018842727830087778,
      "loss": 1.0964,
      "step": 432
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.00018837214806643115,
      "loss": 1.0254,
      "step": 433
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.00018831689493346095,
      "loss": 1.0748,
      "step": 434
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.00018826151897880728,
      "loss": 1.0797,
      "step": 435
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.00018820602027948114,
      "loss": 1.1068,
      "step": 436
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.00018815039891266418,
      "loss": 1.081,
      "step": 437
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.0001880946549557086,
      "loss": 1.0685,
      "step": 438
    },
    {
      "epoch": 0.65,
      "learning_rate": 0.00018803878848613716,
      "loss": 1.0916,
      "step": 439
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.00018798279958164295,
      "loss": 1.115,
      "step": 440
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.00018792668832008936,
      "loss": 1.0048,
      "step": 441
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.0001878704547795099,
      "loss": 1.0386,
      "step": 442
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.00018781409903810821,
      "loss": 1.0283,
      "step": 443
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.00018775762117425777,
      "loss": 1.085,
      "step": 444
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.00018770102126650198,
      "loss": 1.0582,
      "step": 445
    },
    {
      "epoch": 0.66,
      "learning_rate": 0.00018764429939355392,
      "loss": 1.0705,
      "step": 446
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.0001875874556342963,
      "loss": 1.1426,
      "step": 447
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.00018753049006778132,
      "loss": 1.0337,
      "step": 448
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.0001874734027732306,
      "loss": 1.0993,
      "step": 449
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.00018741619383003507,
      "loss": 1.0661,
      "step": 450
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.00018735886331775476,
      "loss": 1.0564,
      "step": 451
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.00018730141131611882,
      "loss": 1.0989,
      "step": 452
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.0001872438379050254,
      "loss": 1.0984,
      "step": 453
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.00018718614316454133,
      "loss": 1.1173,
      "step": 454
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.00018712832717490235,
      "loss": 1.1005,
      "step": 455
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.00018707039001651277,
      "loss": 1.0008,
      "step": 456
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.00018701233176994533,
      "loss": 1.0701,
      "step": 457
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.00018695415251594123,
      "loss": 1.0831,
      "step": 458
    },
    {
      "epoch": 0.68,
      "learning_rate": 0.00018689585233541003,
      "loss": 1.1165,
      "step": 459
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.00018683743130942928,
      "loss": 1.0884,
      "step": 460
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.00018677888951924474,
      "loss": 0.9882,
      "step": 461
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.00018672022704627002,
      "loss": 1.086,
      "step": 462
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.00018666144397208668,
      "loss": 1.0545,
      "step": 463
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.00018660254037844388,
      "loss": 1.0274,
      "step": 464
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.0001865435163472584,
      "loss": 1.0795,
      "step": 465
    },
    {
      "epoch": 0.69,
      "learning_rate": 0.00018648437196061462,
      "loss": 1.022,
      "step": 466
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.0001864251073007642,
      "loss": 1.0717,
      "step": 467
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.00018636572245012606,
      "loss": 1.1501,
      "step": 468
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.0001863062174912863,
      "loss": 1.1034,
      "step": 469
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.00018624659250699805,
      "loss": 1.0784,
      "step": 470
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.00018618684758018136,
      "loss": 1.1274,
      "step": 471
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.0001861269827939231,
      "loss": 1.0643,
      "step": 472
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.00018606699823147676,
      "loss": 1.1394,
      "step": 473
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.00018600689397626246,
      "loss": 0.9665,
      "step": 474
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.00018594667011186678,
      "loss": 1.058,
      "step": 475
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.00018588632672204264,
      "loss": 1.0706,
      "step": 476
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.0001858258638907091,
      "loss": 1.0414,
      "step": 477
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.00018576528170195146,
      "loss": 1.1,
      "step": 478
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.00018570458024002093,
      "loss": 1.1114,
      "step": 479
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.00018564375958933459,
      "loss": 1.0596,
      "step": 480
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.0001855828198344753,
      "loss": 1.0897,
      "step": 481
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.00018552176106019155,
      "loss": 1.0316,
      "step": 482
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.00018546058335139733,
      "loss": 1.0516,
      "step": 483
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.0001853992867931721,
      "loss": 1.0477,
      "step": 484
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.00018533787147076048,
      "loss": 1.0432,
      "step": 485
    },
    {
      "epoch": 0.72,
      "learning_rate": 0.00018527633746957234,
      "loss": 1.0568,
      "step": 486
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.00018521468487518264,
      "loss": 1.114,
      "step": 487
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.00018515291377333112,
      "loss": 1.0664,
      "step": 488
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0001850910242499225,
      "loss": 1.0162,
      "step": 489
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0001850290163910261,
      "loss": 1.0829,
      "step": 490
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.00018496689028287572,
      "loss": 1.1078,
      "step": 491
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.0001849046460118698,
      "loss": 1.0533,
      "step": 492
    },
    {
      "epoch": 0.73,
      "learning_rate": 0.00018484228366457095,
      "loss": 1.0923,
      "step": 493
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.00018477980332770607,
      "loss": 1.0516,
      "step": 494
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.00018471720508816614,
      "loss": 0.9826,
      "step": 495
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.00018465448903300606,
      "loss": 1.1581,
      "step": 496
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.0001845916552494446,
      "loss": 1.1268,
      "step": 497
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.00018452870382486432,
      "loss": 1.0483,
      "step": 498
    },
    {
      "epoch": 0.74,
      "learning_rate": 0.00018446563484681127,
      "loss": 1.1792,
      "step": 499
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.00018440244840299506,
      "loss": 1.0918,
      "step": 500
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.0001843391445812886,
      "loss": 0.9691,
      "step": 501
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.00018427572346972805,
      "loss": 1.0581,
      "step": 502
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.0001842121851565128,
      "loss": 1.0072,
      "step": 503
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.00018414852973000503,
      "loss": 0.9686,
      "step": 504
    },
    {
      "epoch": 0.75,
      "eval_loss": 1.0276715755462646,
      "eval_runtime": 2.6054,
      "eval_samples_per_second": 419.124,
      "eval_steps_per_second": 26.483,
      "step": 504
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.00018408475727872995,
      "loss": 1.1221,
      "step": 505
    },
    {
      "epoch": 0.75,
      "learning_rate": 0.00018402086789137546,
      "loss": 1.087,
      "step": 506
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.00018395686165679202,
      "loss": 1.0599,
      "step": 507
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.00018389273866399275,
      "loss": 1.1844,
      "step": 508
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.00018382849900215294,
      "loss": 1.046,
      "step": 509
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.00018376414276061032,
      "loss": 0.9691,
      "step": 510
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.00018369967002886464,
      "loss": 1.0996,
      "step": 511
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.0001836350808965776,
      "loss": 1.083,
      "step": 512
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.00018357037545357297,
      "loss": 1.0371,
      "step": 513
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.00018350555378983608,
      "loss": 1.018,
      "step": 514
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.00018344061599551398,
      "loss": 1.095,
      "step": 515
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.00018337556216091517,
      "loss": 1.0871,
      "step": 516
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.0001833103923765096,
      "loss": 1.0774,
      "step": 517
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.00018324510673292842,
      "loss": 1.0337,
      "step": 518
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.0001831797053209639,
      "loss": 1.0059,
      "step": 519
    },
    {
      "epoch": 0.77,
      "learning_rate": 0.00018311418823156936,
      "loss": 1.0744,
      "step": 520
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.00018304855555585894,
      "loss": 0.9732,
      "step": 521
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.00018298280738510752,
      "loss": 1.1176,
      "step": 522
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.00018291694381075056,
      "loss": 1.1485,
      "step": 523
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.00018285096492438424,
      "loss": 1.1044,
      "step": 524
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.00018278487081776476,
      "loss": 0.9812,
      "step": 525
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.00018271866158280884,
      "loss": 1.0966,
      "step": 526
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.0001826523373115931,
      "loss": 1.2406,
      "step": 527
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.0001825858980963543,
      "loss": 1.0727,
      "step": 528
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.000182519344029489,
      "loss": 0.9966,
      "step": 529
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.00018245267520355346,
      "loss": 1.081,
      "step": 530
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.00018238589171126353,
      "loss": 1.1104,
      "step": 531
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.00018231899364549455,
      "loss": 1.0535,
      "step": 532
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.00018225198109928114,
      "loss": 1.0801,
      "step": 533
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.00018218485416581726,
      "loss": 1.0726,
      "step": 534
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.00018211761293845585,
      "loss": 1.0923,
      "step": 535
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.00018205025751070875,
      "loss": 1.0551,
      "step": 536
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.00018198278797624675,
      "loss": 1.0495,
      "step": 537
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.0001819152044288992,
      "loss": 1.0589,
      "step": 538
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.00018184750696265408,
      "loss": 1.0487,
      "step": 539
    },
    {
      "epoch": 0.8,
      "learning_rate": 0.0001817796956716578,
      "loss": 1.0491,
      "step": 540
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.000181711770650215,
      "loss": 1.0981,
      "step": 541
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.00018164373199278856,
      "loss": 1.1706,
      "step": 542
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.0001815755797939994,
      "loss": 1.1024,
      "step": 543
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.00018150731414862622,
      "loss": 1.0488,
      "step": 544
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.00018143893515160564,
      "loss": 1.165,
      "step": 545
    },
    {
      "epoch": 0.81,
      "learning_rate": 0.00018137044289803181,
      "loss": 1.0346,
      "step": 546
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.00018130183748315645,
      "loss": 1.1179,
      "step": 547
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.0001812331190023886,
      "loss": 1.0027,
      "step": 548
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.00018116428755129459,
      "loss": 1.1106,
      "step": 549
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.00018109534322559783,
      "loss": 1.0479,
      "step": 550
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.00018102628612117865,
      "loss": 1.0046,
      "step": 551
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.0001809571163340744,
      "loss": 0.9883,
      "step": 552
    },
    {
      "epoch": 0.82,
      "learning_rate": 0.00018088783396047893,
      "loss": 1.1018,
      "step": 553
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.00018081843909674276,
      "loss": 1.1389,
      "step": 554
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.00018074893183937283,
      "loss": 1.0751,
      "step": 555
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.00018067931228503246,
      "loss": 1.1475,
      "step": 556
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.00018060958053054096,
      "loss": 1.0829,
      "step": 557
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.00018053973667287387,
      "loss": 1.0272,
      "step": 558
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.00018046978080916252,
      "loss": 1.0668,
      "step": 559
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.00018039971303669407,
      "loss": 1.0988,
      "step": 560
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.00018032953345291123,
      "loss": 1.0339,
      "step": 561
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.0001802592421554123,
      "loss": 1.0654,
      "step": 562
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.00018018883924195085,
      "loss": 1.0157,
      "step": 563
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.00018011832481043576,
      "loss": 1.0738,
      "step": 564
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.0001800476989589309,
      "loss": 1.0742,
      "step": 565
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.0001799769617856552,
      "loss": 0.9861,
      "step": 566
    },
    {
      "epoch": 0.85,
      "learning_rate": 0.0001799061133889823,
      "loss": 1.0788,
      "step": 567
    },
    {
      "epoch": 0.85,
      "learning_rate": 0.00017983515386744061,
      "loss": 1.0539,
      "step": 568
    },
    {
      "epoch": 0.85,
      "learning_rate": 0.00017976408331971298,
      "loss": 1.0875,
      "step": 569
    },
    {
      "epoch": 0.85,
      "learning_rate": 0.0001796929018446368,
      "loss": 1.0765,
      "step": 570
    },
    {
      "epoch": 0.85,
      "learning_rate": 0.00017962160954120354,
      "loss": 1.1336,
      "step": 571
    },
    {
      "epoch": 0.85,
      "learning_rate": 0.000179550206508559,
      "loss": 0.9674,
      "step": 572
    },
    {
      "epoch": 0.85,
      "learning_rate": 0.00017947869284600282,
      "loss": 1.0607,
      "step": 573
    },
    {
      "epoch": 0.86,
      "learning_rate": 0.0001794070686529886,
      "loss": 0.9959,
      "step": 574
    },
    {
      "epoch": 0.86,
      "learning_rate": 0.00017933533402912354,
      "loss": 1.038,
      "step": 575
    },
    {
      "epoch": 0.86,
      "learning_rate": 0.0001792634890741685,
      "loss": 1.1342,
      "step": 576
    },
    {
      "epoch": 0.86,
      "learning_rate": 0.00017919153388803774,
      "loss": 1.0941,
      "step": 577
    },
    {
      "epoch": 0.86,
      "learning_rate": 0.00017911946857079888,
      "loss": 1.1286,
      "step": 578
    },
    {
      "epoch": 0.86,
      "learning_rate": 0.00017904729322267256,
      "loss": 1.0354,
      "step": 579
    },
    {
      "epoch": 0.86,
      "learning_rate": 0.0001789750079440326,
      "loss": 1.1314,
      "step": 580
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.00017890261283540562,
      "loss": 1.0365,
      "step": 581
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.00017883010799747099,
      "loss": 1.091,
      "step": 582
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.00017875749353106062,
      "loss": 0.9995,
      "step": 583
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.000178684769537159,
      "loss": 1.0435,
      "step": 584
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.00017861193611690287,
      "loss": 1.0555,
      "step": 585
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.00017853899337158112,
      "loss": 1.0637,
      "step": 586
    },
    {
      "epoch": 0.87,
      "learning_rate": 0.00017846594140263474,
      "loss": 1.064,
      "step": 587
    },
    {
      "epoch": 0.88,
      "learning_rate": 0.00017839278031165658,
      "loss": 0.9879,
      "step": 588
    },
    {
      "epoch": 0.88,
      "learning_rate": 0.00017831951020039126,
      "loss": 1.0846,
      "step": 589
    },
    {
      "epoch": 0.88,
      "learning_rate": 0.000178246131170735,
      "loss": 1.0373,
      "step": 590
    },
    {
      "epoch": 0.88,
      "learning_rate": 0.00017817264332473546,
      "loss": 1.0377,
      "step": 591
    },
    {
      "epoch": 0.88,
      "learning_rate": 0.00017809904676459177,
      "loss": 1.0932,
      "step": 592
    },
    {
      "epoch": 0.88,
      "learning_rate": 0.00017802534159265404,
      "loss": 1.085,
      "step": 593
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.0001779515279114236,
      "loss": 1.0975,
      "step": 594
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.0001778776058235526,
      "loss": 1.1283,
      "step": 595
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.00017780357543184397,
      "loss": 1.0652,
      "step": 596
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.00017772943683925122,
      "loss": 1.0336,
      "step": 597
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.00017765519014887842,
      "loss": 0.9761,
      "step": 598
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.0001775808354639799,
      "loss": 1.0688,
      "step": 599
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.00017750637288796016,
      "loss": 1.1031,
      "step": 600
    },
    {
      "epoch": 0.9,
      "learning_rate": 0.00017743180252437383,
      "loss": 1.083,
      "step": 601
    },
    {
      "epoch": 0.9,
      "learning_rate": 0.00017735712447692538,
      "loss": 1.1612,
      "step": 602
    },
    {
      "epoch": 0.9,
      "learning_rate": 0.00017728233884946903,
      "loss": 1.1618,
      "step": 603
    },
    {
      "epoch": 0.9,
      "learning_rate": 0.00017720744574600863,
      "loss": 1.144,
      "step": 604
    },
    {
      "epoch": 0.9,
      "learning_rate": 0.0001771324452706975,
      "loss": 1.1174,
      "step": 605
    },
    {
      "epoch": 0.9,
      "learning_rate": 0.00017705733752783825,
      "loss": 0.9728,
      "step": 606
    },
    {
      "epoch": 0.9,
      "learning_rate": 0.0001769821226218827,
      "loss": 1.0599,
      "step": 607
    },
    {
      "epoch": 0.91,
      "learning_rate": 0.0001769068006574317,
      "loss": 1.0639,
      "step": 608
    },
    {
      "epoch": 0.91,
      "learning_rate": 0.00017683137173923495,
      "loss": 1.1278,
      "step": 609
    },
    {
      "epoch": 0.91,
      "learning_rate": 0.00017675583597219095,
      "loss": 0.9925,
      "step": 610
    },
    {
      "epoch": 0.91,
      "learning_rate": 0.0001766801934613467,
      "loss": 1.0457,
      "step": 611
    },
    {
      "epoch": 0.91,
      "learning_rate": 0.0001766044443118978,
      "loss": 1.0348,
      "step": 612
    },
    {
      "epoch": 0.91,
      "learning_rate": 0.000176528588629188,
      "loss": 1.022,
      "step": 613
    },
    {
      "epoch": 0.92,
      "learning_rate": 0.00017645262651870926,
      "loss": 1.0027,
      "step": 614
    },
    {
      "epoch": 0.92,
      "learning_rate": 0.00017637655808610156,
      "loss": 1.0491,
      "step": 615
    },
    {
      "epoch": 0.92,
      "learning_rate": 0.00017630038343715275,
      "loss": 1.0413,
      "step": 616
    },
    {
      "epoch": 0.92,
      "learning_rate": 0.00017622410267779834,
      "loss": 1.0358,
      "step": 617
    },
    {
      "epoch": 0.92,
      "learning_rate": 0.00017614771591412148,
      "loss": 1.1125,
      "step": 618
    },
    {
      "epoch": 0.92,
      "learning_rate": 0.00017607122325235267,
      "loss": 1.1185,
      "step": 619
    },
    {
      "epoch": 0.92,
      "learning_rate": 0.00017599462479886974,
      "loss": 1.0738,
      "step": 620
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.00017591792066019765,
      "loss": 1.102,
      "step": 621
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.00017584111094300827,
      "loss": 1.065,
      "step": 622
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.0001757641957541203,
      "loss": 1.0514,
      "step": 623
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.0001756871752004992,
      "loss": 1.0396,
      "step": 624
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.00017561004938925688,
      "loss": 1.1027,
      "step": 625
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.00017553281842765169,
      "loss": 1.0223,
      "step": 626
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.00017545548242308816,
      "loss": 1.1793,
      "step": 627
    },
    {
      "epoch": 0.94,
      "learning_rate": 0.00017537804148311695,
      "loss": 1.0642,
      "step": 628
    },
    {
      "epoch": 0.94,
      "learning_rate": 0.00017530049571543464,
      "loss": 1.0682,
      "step": 629
    },
    {
      "epoch": 0.94,
      "learning_rate": 0.00017522284522788353,
      "loss": 1.0476,
      "step": 630
    },
    {
      "epoch": 0.94,
      "learning_rate": 0.00017514509012845164,
      "loss": 1.1064,
      "step": 631
    },
    {
      "epoch": 0.94,
      "learning_rate": 0.00017506723052527242,
      "loss": 1.0258,
      "step": 632
    },
    {
      "epoch": 0.94,
      "learning_rate": 0.00017498926652662476,
      "loss": 1.1954,
      "step": 633
    },
    {
      "epoch": 0.94,
      "learning_rate": 0.0001749111982409325,
      "loss": 1.0637,
      "step": 634
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.00017483302577676475,
      "loss": 0.9685,
      "step": 635
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.00017475474924283536,
      "loss": 1.0465,
      "step": 636
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.000174676368748003,
      "loss": 1.0161,
      "step": 637
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.00017459788440127083,
      "loss": 1.0479,
      "step": 638
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.00017451929631178648,
      "loss": 1.1166,
      "step": 639
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.0001744406045888419,
      "loss": 1.0634,
      "step": 640
    },
    {
      "epoch": 0.96,
      "learning_rate": 0.00017436180934187308,
      "loss": 1.0826,
      "step": 641
    },
    {
      "epoch": 0.96,
      "learning_rate": 0.00017428291068046,
      "loss": 1.07,
      "step": 642
    },
    {
      "epoch": 0.96,
      "learning_rate": 0.00017420390871432647,
      "loss": 1.1167,
      "step": 643
    },
    {
      "epoch": 0.96,
      "learning_rate": 0.00017412480355334005,
      "loss": 1.0347,
      "step": 644
    },
    {
      "epoch": 0.96,
      "learning_rate": 0.00017404559530751162,
      "loss": 1.0393,
      "step": 645
    },
    {
      "epoch": 0.96,
      "learning_rate": 0.00017396628408699555,
      "loss": 1.1108,
      "step": 646
    },
    {
      "epoch": 0.96,
      "learning_rate": 0.00017388687000208946,
      "loss": 1.006,
      "step": 647
    },
    {
      "epoch": 0.97,
      "learning_rate": 0.0001738073531632339,
      "loss": 1.0932,
      "step": 648
    },
    {
      "epoch": 0.97,
      "learning_rate": 0.0001737277336810124,
      "loss": 1.0123,
      "step": 649
    },
    {
      "epoch": 0.97,
      "learning_rate": 0.00017364801166615124,
      "loss": 1.1273,
      "step": 650
    },
    {
      "epoch": 0.97,
      "learning_rate": 0.0001735681872295192,
      "loss": 0.9893,
      "step": 651
    },
    {
      "epoch": 0.97,
      "learning_rate": 0.0001734882604821276,
      "loss": 1.0699,
      "step": 652
    },
    {
      "epoch": 0.97,
      "learning_rate": 0.00017340823153513002,
      "loss": 1.0901,
      "step": 653
    },
    {
      "epoch": 0.97,
      "learning_rate": 0.00017332810049982208,
      "loss": 1.0212,
      "step": 654
    },
    {
      "epoch": 0.98,
      "learning_rate": 0.00017324786748764155,
      "loss": 0.9898,
      "step": 655
    },
    {
      "epoch": 0.98,
      "learning_rate": 0.00017316753261016783,
      "loss": 1.0899,
      "step": 656
    },
    {
      "epoch": 0.98,
      "learning_rate": 0.00017308709597912213,
      "loss": 1.085,
      "step": 657
    },
    {
      "epoch": 0.98,
      "learning_rate": 0.00017300655770636708,
      "loss": 1.091,
      "step": 658
    },
    {
      "epoch": 0.98,
      "learning_rate": 0.00017292591790390665,
      "loss": 1.0502,
      "step": 659
    },
    {
      "epoch": 0.98,
      "learning_rate": 0.0001728451766838861,
      "loss": 1.2131,
      "step": 660
    },
    {
      "epoch": 0.99,
      "learning_rate": 0.00017276433415859167,
      "loss": 1.1256,
      "step": 661
    },
    {
      "epoch": 0.99,
      "learning_rate": 0.00017268339044045042,
      "loss": 1.0577,
      "step": 662
    },
    {
      "epoch": 0.99,
      "learning_rate": 0.00017260234564203032,
      "loss": 1.0012,
      "step": 663
    },
    {
      "epoch": 0.99,
      "learning_rate": 0.00017252119987603973,
      "loss": 1.0611,
      "step": 664
    },
    {
      "epoch": 0.99,
      "learning_rate": 0.00017243995325532755,
      "loss": 1.1251,
      "step": 665
    },
    {
      "epoch": 0.99,
      "learning_rate": 0.00017235860589288277,
      "loss": 1.0959,
      "step": 666
    },
    {
      "epoch": 0.99,
      "learning_rate": 0.0001722771579018347,
      "loss": 1.1413,
      "step": 667
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.00017219560939545246,
      "loss": 1.0728,
      "step": 668
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.00017211396048714498,
      "loss": 1.0461,
      "step": 669
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.0001720322112904608,
      "loss": 1.1084,
      "step": 670
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.00017195036191908797,
      "loss": 1.1316,
      "step": 671
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.00017186841248685383,
      "loss": 1.0816,
      "step": 672
    },
    {
      "epoch": 1.0,
      "eval_loss": 1.0170178413391113,
      "eval_runtime": 2.6119,
      "eval_samples_per_second": 418.079,
      "eval_steps_per_second": 26.417,
      "step": 672
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.0001717863631077249,
      "loss": 1.0711,
      "step": 673
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.00017170421389580667,
      "loss": 1.1245,
      "step": 674
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.00017162196496534342,
      "loss": 1.0519,
      "step": 675
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.0001715396164307182,
      "loss": 1.104,
      "step": 676
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.00017145716840645254,
      "loss": 1.1193,
      "step": 677
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.00017137462100720631,
      "loss": 1.1238,
      "step": 678
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.00017129197434777763,
      "loss": 1.004,
      "step": 679
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.00017120922854310257,
      "loss": 1.0426,
      "step": 680
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.00017112638370825515,
      "loss": 1.0308,
      "step": 681
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.00017104343995844715,
      "loss": 1.0892,
      "step": 682
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.00017096039740902784,
      "loss": 1.0115,
      "step": 683
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.00017087725617548385,
      "loss": 1.1011,
      "step": 684
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.00017079401637343914,
      "loss": 0.9829,
      "step": 685
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.00017071067811865476,
      "loss": 0.9738,
      "step": 686
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.0001706272415270286,
      "loss": 1.0563,
      "step": 687
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.00017054370671459532,
      "loss": 1.0153,
      "step": 688
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.0001704600737975262,
      "loss": 1.0638,
      "step": 689
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.000170376342892129,
      "loss": 1.0053,
      "step": 690
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.00017029251411484765,
      "loss": 1.0178,
      "step": 691
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.00017020858758226229,
      "loss": 1.0755,
      "step": 692
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.00017012456341108885,
      "loss": 0.9365,
      "step": 693
    },
    {
      "epoch": 1.01,
      "learning_rate": 0.00017004044171817925,
      "loss": 1.0666,
      "step": 694
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.00016995622262052092,
      "loss": 1.041,
      "step": 695
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.00016987190623523674,
      "loss": 1.0387,
      "step": 696
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.00016978749267958495,
      "loss": 0.9332,
      "step": 697
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.00016970298207095885,
      "loss": 1.0737,
      "step": 698
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.00016961837452688676,
      "loss": 0.992,
      "step": 699
    },
    {
      "epoch": 1.02,
      "learning_rate": 0.00016953367016503182,
      "loss": 0.9997,
      "step": 700
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.00016944886910319173,
      "loss": 1.1054,
      "step": 701
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.00016936397145929878,
      "loss": 0.9876,
      "step": 702
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.00016927897735141952,
      "loss": 1.0158,
      "step": 703
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.00016919388689775464,
      "loss": 0.9771,
      "step": 704
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.00016910870021663883,
      "loss": 0.942,
      "step": 705
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.00016902341742654065,
      "loss": 1.0217,
      "step": 706
    },
    {
      "epoch": 1.03,
      "learning_rate": 0.00016893803864606222,
      "loss": 1.0346,
      "step": 707
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.00016885256399393924,
      "loss": 0.9891,
      "step": 708
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.00016876699358904068,
      "loss": 0.9697,
      "step": 709
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.00016868132755036875,
      "loss": 1.0062,
      "step": 710
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.00016859556599705856,
      "loss": 0.9822,
      "step": 711
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.0001685097090483781,
      "loss": 1.0921,
      "step": 712
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.00016842375682372805,
      "loss": 1.0126,
      "step": 713
    },
    {
      "epoch": 1.04,
      "learning_rate": 0.00016833770944264153,
      "loss": 1.0043,
      "step": 714
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.00016825156702478407,
      "loss": 0.952,
      "step": 715
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.00016816532968995328,
      "loss": 1.0423,
      "step": 716
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.00016807899755807886,
      "loss": 1.0465,
      "step": 717
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.00016799257074922224,
      "loss": 0.9827,
      "step": 718
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.00016790604938357663,
      "loss": 0.9798,
      "step": 719
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.00016781943358146664,
      "loss": 1.0268,
      "step": 720
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.00016773272346334828,
      "loss": 1.0007,
      "step": 721
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.0001676459191498087,
      "loss": 0.9989,
      "step": 722
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.00016755902076156604,
      "loss": 0.9374,
      "step": 723
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.00016747202841946928,
      "loss": 1.0031,
      "step": 724
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.00016738494224449802,
      "loss": 0.9751,
      "step": 725
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.00016729776235776246,
      "loss": 1.1055,
      "step": 726
    },
    {
      "epoch": 1.06,
      "learning_rate": 0.00016721048888050302,
      "loss": 1.0527,
      "step": 727
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.0001671231219340903,
      "loss": 1.0048,
      "step": 728
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.0001670356616400249,
      "loss": 0.957,
      "step": 729
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.00016694810811993723,
      "loss": 1.0598,
      "step": 730
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.00016686046149558736,
      "loss": 1.02,
      "step": 731
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.00016677272188886483,
      "loss": 0.9973,
      "step": 732
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.00016668488942178856,
      "loss": 1.0685,
      "step": 733
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.00016659696421650645,
      "loss": 0.9783,
      "step": 734
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.00016650894639529544,
      "loss": 0.9767,
      "step": 735
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.00016642083608056141,
      "loss": 1.0192,
      "step": 736
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.00016633263339483866,
      "loss": 1.0121,
      "step": 737
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.00016624433846079012,
      "loss": 0.9817,
      "step": 738
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.00016615595140120686,
      "loss": 1.1145,
      "step": 739
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.00016606747233900815,
      "loss": 0.9862,
      "step": 740
    },
    {
      "epoch": 1.08,
      "learning_rate": 0.00016597890139724125,
      "loss": 1.0606,
      "step": 741
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0001658902386990811,
      "loss": 1.0416,
      "step": 742
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0001658014843678303,
      "loss": 0.9971,
      "step": 743
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.00016571263852691888,
      "loss": 1.0318,
      "step": 744
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0001656237012999041,
      "loss": 1.0633,
      "step": 745
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.0001655346728104704,
      "loss": 1.0418,
      "step": 746
    },
    {
      "epoch": 1.09,
      "learning_rate": 0.00016544555318242897,
      "loss": 0.9308,
      "step": 747
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.00016535634253971794,
      "loss": 1.1049,
      "step": 748
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.0001652670410064019,
      "loss": 0.9377,
      "step": 749
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.00016517764870667182,
      "loss": 0.9934,
      "step": 750
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.000165088165764845,
      "loss": 1.0467,
      "step": 751
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.00016499859230536466,
      "loss": 1.0172,
      "step": 752
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.0001649089284528001,
      "loss": 0.9922,
      "step": 753
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.00016481917433184607,
      "loss": 1.0373,
      "step": 754
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.0001647293300673231,
      "loss": 1.0377,
      "step": 755
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.00016463939578417692,
      "loss": 0.9991,
      "step": 756
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.00016454937160747854,
      "loss": 1.0657,
      "step": 757
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.00016445925766242391,
      "loss": 0.9954,
      "step": 758
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.0001643690540743339,
      "loss": 1.018,
      "step": 759
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.00016427876096865394,
      "loss": 1.01,
      "step": 760
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.0001641883784709541,
      "loss": 0.9318,
      "step": 761
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.0001640979067069286,
      "loss": 1.0174,
      "step": 762
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.00016400734580239594,
      "loss": 1.0886,
      "step": 763
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.0001639166958832985,
      "loss": 1.0316,
      "step": 764
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.0001638259570757025,
      "loss": 1.0514,
      "step": 765
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.0001637351295057978,
      "loss": 0.9914,
      "step": 766
    },
    {
      "epoch": 1.12,
      "learning_rate": 0.00016364421329989755,
      "loss": 1.0529,
      "step": 767
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.00016355320858443842,
      "loss": 0.9689,
      "step": 768
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.00016346211548597995,
      "loss": 1.0398,
      "step": 769
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.0001633709341312046,
      "loss": 1.0127,
      "step": 770
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.00016327966464691778,
      "loss": 1.1388,
      "step": 771
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.00016318830716004722,
      "loss": 0.9659,
      "step": 772
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.00016309686179764317,
      "loss": 0.9907,
      "step": 773
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.00016300532868687806,
      "loss": 0.9168,
      "step": 774
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.0001629137079550463,
      "loss": 1.06,
      "step": 775
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.00016282199972956425,
      "loss": 0.9826,
      "step": 776
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.00016273020413796983,
      "loss": 1.0496,
      "step": 777
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.0001626383213079226,
      "loss": 1.0245,
      "step": 778
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.00016254635136720328,
      "loss": 1.036,
      "step": 779
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.0001624542944437139,
      "loss": 1.0283,
      "step": 780
    },
    {
      "epoch": 1.14,
      "learning_rate": 0.00016236215066547734,
      "loss": 1.0078,
      "step": 781
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.00016226992016063723,
      "loss": 0.9819,
      "step": 782
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.00016217760305745803,
      "loss": 1.0687,
      "step": 783
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.0001620851994843244,
      "loss": 1.0523,
      "step": 784
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.00016199270956974128,
      "loss": 1.0279,
      "step": 785
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.00016190013344233388,
      "loss": 1.0559,
      "step": 786
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.00016180747123084705,
      "loss": 1.0844,
      "step": 787
    },
    {
      "epoch": 1.15,
      "learning_rate": 0.00016171472306414554,
      "loss": 1.0724,
      "step": 788
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.00016162188907121354,
      "loss": 0.9696,
      "step": 789
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.00016152896938115464,
      "loss": 0.9551,
      "step": 790
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.0001614359641231916,
      "loss": 1.0032,
      "step": 791
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.0001613428734266662,
      "loss": 1.1404,
      "step": 792
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.00016124969742103897,
      "loss": 1.0329,
      "step": 793
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.00016115643623588915,
      "loss": 1.039,
      "step": 794
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0001610630900009144,
      "loss": 1.0231,
      "step": 795
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0001609696588459307,
      "loss": 1.0659,
      "step": 796
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.00016087614290087208,
      "loss": 1.0029,
      "step": 797
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.0001607825422957905,
      "loss": 0.985,
      "step": 798
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.00016068885716085567,
      "loss": 0.9392,
      "step": 799
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.00016059508762635482,
      "loss": 1.006,
      "step": 800
    },
    {
      "epoch": 1.17,
      "learning_rate": 0.00016050123382269264,
      "loss": 1.0748,
      "step": 801
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.0001604072958803909,
      "loss": 1.1378,
      "step": 802
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.00016031327393008845,
      "loss": 1.058,
      "step": 803
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.00016021916810254097,
      "loss": 0.9827,
      "step": 804
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.00016012497852862075,
      "loss": 0.9572,
      "step": 805
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.00016003070533931657,
      "loss": 1.0042,
      "step": 806
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.00015993634866573347,
      "loss": 0.9521,
      "step": 807
    },
    {
      "epoch": 1.18,
      "learning_rate": 0.0001598419086390927,
      "loss": 0.9395,
      "step": 808
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.00015974738539073125,
      "loss": 1.0902,
      "step": 809
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.00015965277905210195,
      "loss": 1.0408,
      "step": 810
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.00015955808975477319,
      "loss": 1.0436,
      "step": 811
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.00015946331763042867,
      "loss": 1.0845,
      "step": 812
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.00015936846281086736,
      "loss": 1.0752,
      "step": 813
    },
    {
      "epoch": 1.19,
      "learning_rate": 0.00015927352542800317,
      "loss": 1.0832,
      "step": 814
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.00015917850561386488,
      "loss": 0.9901,
      "step": 815
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.00015908340350059583,
      "loss": 1.0311,
      "step": 816
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.00015898821922045385,
      "loss": 0.9858,
      "step": 817
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.0001588929529058111,
      "loss": 0.9541,
      "step": 818
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.00015879760468915372,
      "loss": 0.9516,
      "step": 819
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.00015870217470308188,
      "loss": 1.0791,
      "step": 820
    },
    {
      "epoch": 1.2,
      "learning_rate": 0.00015860666308030932,
      "loss": 0.9099,
      "step": 821
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.00015851106995366337,
      "loss": 1.0983,
      "step": 822
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.00015841539545608478,
      "loss": 0.9951,
      "step": 823
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.00015831963972062733,
      "loss": 0.9661,
      "step": 824
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.00015822380288045792,
      "loss": 1.0111,
      "step": 825
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.0001581278850688561,
      "loss": 1.0436,
      "step": 826
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.00015803188641921417,
      "loss": 1.0916,
      "step": 827
    },
    {
      "epoch": 1.21,
      "learning_rate": 0.0001579358070650367,
      "loss": 1.0347,
      "step": 828
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.0001578396471399406,
      "loss": 0.9978,
      "step": 829
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.0001577434067776548,
      "loss": 1.0036,
      "step": 830
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.00015764708611202015,
      "loss": 1.0387,
      "step": 831
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.00015755068527698902,
      "loss": 1.0172,
      "step": 832
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.00015745420440662543,
      "loss": 0.9723,
      "step": 833
    },
    {
      "epoch": 1.22,
      "learning_rate": 0.0001573576436351046,
      "loss": 0.9662,
      "step": 834
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.000157261003096713,
      "loss": 0.9849,
      "step": 835
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.00015716428292584787,
      "loss": 1.0198,
      "step": 836
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.00015706748325701732,
      "loss": 0.9015,
      "step": 837
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.0001569706042248399,
      "loss": 1.001,
      "step": 838
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.0001568736459640447,
      "loss": 0.9681,
      "step": 839
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.00015677660860947078,
      "loss": 1.0513,
      "step": 840
    },
    {
      "epoch": 1.23,
      "eval_loss": 1.008791446685791,
      "eval_runtime": 2.6133,
      "eval_samples_per_second": 417.867,
      "eval_steps_per_second": 26.404,
      "step": 840
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.0001566794922960674,
      "loss": 0.9829,
      "step": 841
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.00015658229715889347,
      "loss": 1.0362,
      "step": 842
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.00015648502333311757,
      "loss": 0.9736,
      "step": 843
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.0001563876709540178,
      "loss": 1.0457,
      "step": 844
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.00015629024015698136,
      "loss": 0.9786,
      "step": 845
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.00015619273107750462,
      "loss": 1.0595,
      "step": 846
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.0001560951438511927,
      "loss": 1.0307,
      "step": 847
    },
    {
      "epoch": 1.24,
      "learning_rate": 0.00015599747861375955,
      "loss": 1.0386,
      "step": 848
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.00015589973550102747,
      "loss": 0.9916,
      "step": 849
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.00015580191464892716,
      "loss": 0.9652,
      "step": 850
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.00015570401619349736,
      "loss": 0.9691,
      "step": 851
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.00015560604027088477,
      "loss": 1.0006,
      "step": 852
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.00015550798701734385,
      "loss": 1.0271,
      "step": 853
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.00015540985656923645,
      "loss": 1.0591,
      "step": 854
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.00015531164906303207,
      "loss": 0.967,
      "step": 855
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.00015521336463530705,
      "loss": 1.0466,
      "step": 856
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.0001551150034227449,
      "loss": 0.9953,
      "step": 857
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.0001550165655621359,
      "loss": 0.9899,
      "step": 858
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.00015491805119037684,
      "loss": 0.9742,
      "step": 859
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.00015481946044447099,
      "loss": 0.9865,
      "step": 860
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.0001547207934615278,
      "loss": 0.9384,
      "step": 861
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.00015462205037876275,
      "loss": 1.0216,
      "step": 862
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.00015452323133349714,
      "loss": 0.9467,
      "step": 863
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.0001544243364631579,
      "loss": 1.0038,
      "step": 864
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.0001543253659052775,
      "loss": 0.978,
      "step": 865
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.00015422631979749354,
      "loss": 1.0434,
      "step": 866
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.00015412719827754873,
      "loss": 1.0091,
      "step": 867
    },
    {
      "epoch": 1.27,
      "learning_rate": 0.00015402800148329071,
      "loss": 0.9598,
      "step": 868
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.00015392872955267175,
      "loss": 0.9876,
      "step": 869
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.00015382938262374865,
      "loss": 0.9559,
      "step": 870
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.0001537299608346824,
      "loss": 0.9984,
      "step": 871
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.00015363046432373824,
      "loss": 1.0171,
      "step": 872
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.0001535308932292853,
      "loss": 1.0188,
      "step": 873
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.00015343124768979637,
      "loss": 0.9613,
      "step": 874
    },
    {
      "epoch": 1.28,
      "learning_rate": 0.00015333152784384777,
      "loss": 0.9572,
      "step": 875
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.0001532317338301192,
      "loss": 1.0093,
      "step": 876
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.00015313186578739353,
      "loss": 0.9935,
      "step": 877
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.0001530319238545565,
      "loss": 1.0371,
      "step": 878
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.00015293190817059667,
      "loss": 1.0022,
      "step": 879
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.00015283181887460517,
      "loss": 1.0033,
      "step": 880
    },
    {
      "epoch": 1.29,
      "learning_rate": 0.00015273165610577542,
      "loss": 0.9986,
      "step": 881
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.00015263142000340312,
      "loss": 1.0495,
      "step": 882
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.0001525311107068859,
      "loss": 1.017,
      "step": 883
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.00015243072835572318,
      "loss": 0.9757,
      "step": 884
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.000152330273089516,
      "loss": 1.0342,
      "step": 885
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.0001522297450479668,
      "loss": 1.0059,
      "step": 886
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.00015212914437087922,
      "loss": 0.9845,
      "step": 887
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.0001520284711981579,
      "loss": 1.0365,
      "step": 888
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.0001519277256698083,
      "loss": 0.9521,
      "step": 889
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.0001518269079259366,
      "loss": 1.0867,
      "step": 890
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.00015172601810674915,
      "loss": 1.0444,
      "step": 891
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.00015162505635255287,
      "loss": 1.077,
      "step": 892
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.00015152402280375454,
      "loss": 0.9883,
      "step": 893
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.0001514229176008607,
      "loss": 0.9819,
      "step": 894
    },
    {
      "epoch": 1.31,
      "learning_rate": 0.00015132174088447776,
      "loss": 0.9912,
      "step": 895
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.00015122049279531143,
      "loss": 0.9575,
      "step": 896
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.00015111917347416671,
      "loss": 1.0356,
      "step": 897
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.00015101778306194765,
      "loss": 0.9963,
      "step": 898
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.0001509163216996572,
      "loss": 0.9728,
      "step": 899
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.00015081478952839693,
      "loss": 1.0402,
      "step": 900
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.00015071318668936695,
      "loss": 1.0287,
      "step": 901
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.00015061151332386566,
      "loss": 1.0505,
      "step": 902
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.00015050976957328938,
      "loss": 0.9814,
      "step": 903
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.00015040795557913245,
      "loss": 1.0083,
      "step": 904
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.00015030607148298696,
      "loss": 1.0871,
      "step": 905
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.00015020411742654237,
      "loss": 1.0943,
      "step": 906
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.0001501020935515854,
      "loss": 1.0631,
      "step": 907
    },
    {
      "epoch": 1.33,
      "learning_rate": 0.00015000000000000001,
      "loss": 1.0615,
      "step": 908
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.00014989783691376696,
      "loss": 0.8933,
      "step": 909
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.0001497956044349637,
      "loss": 1.012,
      "step": 910
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.00014969330270576427,
      "loss": 0.9215,
      "step": 911
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.00014959093186843895,
      "loss": 0.9894,
      "step": 912
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.00014948849206535412,
      "loss": 1.0053,
      "step": 913
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.00014938598343897214,
      "loss": 1.0975,
      "step": 914
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.00014928340613185097,
      "loss": 1.068,
      "step": 915
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.0001491807602866442,
      "loss": 0.9838,
      "step": 916
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.00014907804604610063,
      "loss": 1.1493,
      "step": 917
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.00014897526355306428,
      "loss": 0.9491,
      "step": 918
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.000148872412950474,
      "loss": 1.0252,
      "step": 919
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.00014876949438136347,
      "loss": 0.9555,
      "step": 920
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.00014866650798886074,
      "loss": 0.9831,
      "step": 921
    },
    {
      "epoch": 1.35,
      "learning_rate": 0.0001485634539161883,
      "loss": 1.0957,
      "step": 922
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.0001484603323066627,
      "loss": 0.9606,
      "step": 923
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.00014835714330369446,
      "loss": 1.0643,
      "step": 924
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.00014825388705078777,
      "loss": 1.0219,
      "step": 925
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.00014815056369154038,
      "loss": 1.1315,
      "step": 926
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.0001480471733696434,
      "loss": 1.0406,
      "step": 927
    },
    {
      "epoch": 1.36,
      "learning_rate": 0.000147943716228881,
      "loss": 1.0284,
      "step": 928
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.00014784019241313026,
      "loss": 1.035,
      "step": 929
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.00014773660206636105,
      "loss": 1.0562,
      "step": 930
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.0001476329453326357,
      "loss": 0.9813,
      "step": 931
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.000147529222356109,
      "loss": 1.0865,
      "step": 932
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.0001474254332810277,
      "loss": 1.0074,
      "step": 933
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.00014732157825173044,
      "loss": 1.0855,
      "step": 934
    },
    {
      "epoch": 1.37,
      "learning_rate": 0.00014721765741264786,
      "loss": 0.9785,
      "step": 935
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.0001471136709083018,
      "loss": 1.011,
      "step": 936
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.00014700961888330563,
      "loss": 1.0484,
      "step": 937
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.0001469055014823637,
      "loss": 1.0435,
      "step": 938
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.00014680131885027141,
      "loss": 1.0176,
      "step": 939
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.00014669707113191483,
      "loss": 0.9542,
      "step": 940
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.00014659275847227042,
      "loss": 0.9526,
      "step": 941
    },
    {
      "epoch": 1.38,
      "learning_rate": 0.00014648838101640518,
      "loss": 0.9681,
      "step": 942
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.00014638393890947603,
      "loss": 0.9072,
      "step": 943
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.0001462794322967299,
      "loss": 0.9939,
      "step": 944
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.00014617486132350343,
      "loss": 1.018,
      "step": 945
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.0001460702261352226,
      "loss": 0.8993,
      "step": 946
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.00014596552687740302,
      "loss": 1.0134,
      "step": 947
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.00014586076369564908,
      "loss": 0.947,
      "step": 948
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.00014575593673565426,
      "loss": 1.0697,
      "step": 949
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.00014565104614320065,
      "loss": 1.006,
      "step": 950
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.00014554609206415885,
      "loss": 1.0262,
      "step": 951
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.00014544107464448775,
      "loss": 0.9809,
      "step": 952
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.0001453359940302344,
      "loss": 0.981,
      "step": 953
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.00014523085036753354,
      "loss": 0.9925,
      "step": 954
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.00014512564380260787,
      "loss": 1.0199,
      "step": 955
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.00014502037448176734,
      "loss": 0.9715,
      "step": 956
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.00014491504255140927,
      "loss": 1.0072,
      "step": 957
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.000144809648158018,
      "loss": 1.0659,
      "step": 958
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.00014470419144816483,
      "loss": 1.0538,
      "step": 959
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.0001445986725685076,
      "loss": 1.0571,
      "step": 960
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.00014449309166579072,
      "loss": 0.9701,
      "step": 961
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.00014438744888684482,
      "loss": 0.9618,
      "step": 962
    },
    {
      "epoch": 1.42,
      "learning_rate": 0.0001442817443785865,
      "loss": 0.9179,
      "step": 963
    },
    {
      "epoch": 1.42,
      "learning_rate": 0.00014417597828801832,
      "loss": 1.0613,
      "step": 964
    },
    {
      "epoch": 1.42,
      "learning_rate": 0.00014407015076222846,
      "loss": 0.9558,
      "step": 965
    },
    {
      "epoch": 1.42,
      "learning_rate": 0.00014396426194839042,
      "loss": 0.9823,
      "step": 966
    },
    {
      "epoch": 1.42,
      "learning_rate": 0.00014385831199376317,
      "loss": 0.9968,
      "step": 967
    },
    {
      "epoch": 1.42,
      "learning_rate": 0.00014375230104569044,
      "loss": 0.9829,
      "step": 968
    },
    {
      "epoch": 1.42,
      "learning_rate": 0.00014364622925160098,
      "loss": 1.0552,
      "step": 969
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.00014354009675900803,
      "loss": 0.993,
      "step": 970
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.00014343390371550935,
      "loss": 1.0927,
      "step": 971
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.00014332765026878687,
      "loss": 1.0387,
      "step": 972
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.00014322133656660647,
      "loss": 0.9255,
      "step": 973
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.00014311496275681783,
      "loss": 1.0093,
      "step": 974
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.00014300852898735435,
      "loss": 1.0078,
      "step": 975
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.00014290203540623267,
      "loss": 0.9161,
      "step": 976
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.00014279548216155266,
      "loss": 1.03,
      "step": 977
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.00014268886940149714,
      "loss": 1.0364,
      "step": 978
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.0001425821972743318,
      "loss": 0.9768,
      "step": 979
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.0001424754659284048,
      "loss": 1.1229,
      "step": 980
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.0001423686755121466,
      "loss": 1.0362,
      "step": 981
    },
    {
      "epoch": 1.44,
      "learning_rate": 0.00014226182617406996,
      "loss": 0.9522,
      "step": 982
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.00014215491806276944,
      "loss": 1.0479,
      "step": 983
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.00014204795132692144,
      "loss": 1.0671,
      "step": 984
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.00014194092611528384,
      "loss": 0.8983,
      "step": 985
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.00014183384257669581,
      "loss": 1.004,
      "step": 986
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.00014172670086007774,
      "loss": 1.0972,
      "step": 987
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.00014161950111443077,
      "loss": 1.0198,
      "step": 988
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.00014151224348883692,
      "loss": 1.0257,
      "step": 989
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.00014140492813245856,
      "loss": 0.9717,
      "step": 990
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.0001412975551945384,
      "loss": 0.9455,
      "step": 991
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.0001411901248243993,
      "loss": 1.0372,
      "step": 992
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.0001410826371714438,
      "loss": 0.9961,
      "step": 993
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.00014097509238515432,
      "loss": 1.0599,
      "step": 994
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.00014086749061509258,
      "loss": 1.0166,
      "step": 995
    },
    {
      "epoch": 1.46,
      "learning_rate": 0.00014075983201089964,
      "loss": 1.0254,
      "step": 996
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.00014065211672229555,
      "loss": 0.9979,
      "step": 997
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.00014054434489907915,
      "loss": 1.0365,
      "step": 998
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.00014043651669112808,
      "loss": 1.0075,
      "step": 999
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.00014032863224839814,
      "loss": 0.9743,
      "step": 1000
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.00014022069172092352,
      "loss": 1.0056,
      "step": 1001
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.00014011269525881636,
      "loss": 0.9647,
      "step": 1002
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.00014000464301226656,
      "loss": 1.0912,
      "step": 1003
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.00013989653513154165,
      "loss": 0.8811,
      "step": 1004
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.00013978837176698646,
      "loss": 1.0667,
      "step": 1005
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.000139680153069023,
      "loss": 1.0096,
      "step": 1006
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.00013957187918815032,
      "loss": 0.926,
      "step": 1007
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.0001394635502749441,
      "loss": 1.0814,
      "step": 1008
    },
    {
      "epoch": 1.48,
      "eval_loss": 1.0040607452392578,
      "eval_runtime": 2.6168,
      "eval_samples_per_second": 417.304,
      "eval_steps_per_second": 26.368,
      "step": 1008
    },
    {
      "epoch": 1.48,
      "learning_rate": 0.0001393551664800566,
      "loss": 1.0941,
      "step": 1009
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.00013924672795421637,
      "loss": 1.044,
      "step": 1010
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.00013913823484822815,
      "loss": 1.049,
      "step": 1011
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.00013902968731297255,
      "loss": 0.9891,
      "step": 1012
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.00013892108549940583,
      "loss": 0.9663,
      "step": 1013
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.00013881242955855974,
      "loss": 1.0298,
      "step": 1014
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.0001387037196415414,
      "loss": 1.0083,
      "step": 1015
    },
    {
      "epoch": 1.49,
      "learning_rate": 0.0001385949558995329,
      "loss": 0.9182,
      "step": 1016
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.00013848613848379114,
      "loss": 1.013,
      "step": 1017
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.00013837726754564785,
      "loss": 1.0022,
      "step": 1018
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.000138268343236509,
      "loss": 0.9423,
      "step": 1019
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.00013815936570785487,
      "loss": 1.058,
      "step": 1020
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.00013805033511123975,
      "loss": 0.931,
      "step": 1021
    },
    {
      "epoch": 1.5,
      "learning_rate": 0.00013794125159829172,
      "loss": 1.0137,
      "step": 1022
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.00013783211532071246,
      "loss": 1.0517,
      "step": 1023
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.000137722926430277,
      "loss": 1.0259,
      "step": 1024
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.00013761368507883359,
      "loss": 1.0263,
      "step": 1025
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.00013750439141830339,
      "loss": 1.0286,
      "step": 1026
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.00013739504560068033,
      "loss": 0.9749,
      "step": 1027
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.00013728564777803088,
      "loss": 0.9317,
      "step": 1028
    },
    {
      "epoch": 1.51,
      "learning_rate": 0.00013717619810249378,
      "loss": 1.0653,
      "step": 1029
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.00013706669672627997,
      "loss": 0.9623,
      "step": 1030
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.00013695714380167223,
      "loss": 0.9911,
      "step": 1031
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.000136847539481025,
      "loss": 0.9843,
      "step": 1032
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.0001367378839167643,
      "loss": 0.981,
      "step": 1033
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.00013662817726138728,
      "loss": 1.0651,
      "step": 1034
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.00013651841966746232,
      "loss": 1.0602,
      "step": 1035
    },
    {
      "epoch": 1.52,
      "learning_rate": 0.0001364086112876284,
      "loss": 0.9524,
      "step": 1036
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.00013629875227459532,
      "loss": 1.0264,
      "step": 1037
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.00013618884278114324,
      "loss": 1.0691,
      "step": 1038
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.00013607888296012259,
      "loss": 1.0527,
      "step": 1039
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.0001359688729644536,
      "loss": 0.9629,
      "step": 1040
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.0001358588129471264,
      "loss": 0.957,
      "step": 1041
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.00013574870306120077,
      "loss": 1.0976,
      "step": 1042
    },
    {
      "epoch": 1.54,
      "learning_rate": 0.00013563854345980569,
      "loss": 0.9317,
      "step": 1043
    },
    {
      "epoch": 1.54,
      "learning_rate": 0.00013552833429613938,
      "loss": 1.0359,
      "step": 1044
    },
    {
      "epoch": 1.54,
      "learning_rate": 0.0001354180757234689,
      "loss": 1.0642,
      "step": 1045
    },
    {
      "epoch": 1.54,
      "learning_rate": 0.0001353077678951301,
      "loss": 1.0526,
      "step": 1046
    },
    {
      "epoch": 1.54,
      "learning_rate": 0.00013519741096452726,
      "loss": 1.0276,
      "step": 1047
    },
    {
      "epoch": 1.54,
      "learning_rate": 0.00013508700508513307,
      "loss": 1.0471,
      "step": 1048
    },
    {
      "epoch": 1.54,
      "learning_rate": 0.0001349765504104881,
      "loss": 1.0353,
      "step": 1049
    },
    {
      "epoch": 1.55,
      "learning_rate": 0.00013486604709420102,
      "loss": 1.0025,
      "step": 1050
    },
    {
      "epoch": 1.55,
      "learning_rate": 0.00013475549528994786,
      "loss": 0.9019,
      "step": 1051
    },
    {
      "epoch": 1.55,
      "learning_rate": 0.00013464489515147238,
      "loss": 1.0453,
      "step": 1052
    },
    {
      "epoch": 1.55,
      "learning_rate": 0.00013453424683258528,
      "loss": 1.0395,
      "step": 1053
    },
    {
      "epoch": 1.55,
      "learning_rate": 0.0001344235504871645,
      "loss": 0.8939,
      "step": 1054
    },
    {
      "epoch": 1.55,
      "learning_rate": 0.00013431280626915467,
      "loss": 0.9198,
      "step": 1055
    },
    {
      "epoch": 1.55,
      "learning_rate": 0.00013420201433256689,
      "loss": 1.0046,
      "step": 1056
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.0001340911748314788,
      "loss": 0.9197,
      "step": 1057
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.00013398028792003413,
      "loss": 0.9547,
      "step": 1058
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.00013386935375244246,
      "loss": 0.968,
      "step": 1059
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.00013375837248297926,
      "loss": 0.9611,
      "step": 1060
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.00013364734426598527,
      "loss": 1.0125,
      "step": 1061
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.00013353626925586672,
      "loss": 1.0179,
      "step": 1062
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.00013342514760709485,
      "loss": 1.04,
      "step": 1063
    },
    {
      "epoch": 1.57,
      "learning_rate": 0.00013331397947420576,
      "loss": 0.9251,
      "step": 1064
    },
    {
      "epoch": 1.57,
      "learning_rate": 0.00013320276501180015,
      "loss": 1.0762,
      "step": 1065
    },
    {
      "epoch": 1.57,
      "learning_rate": 0.00013309150437454322,
      "loss": 1.0137,
      "step": 1066
    },
    {
      "epoch": 1.57,
      "learning_rate": 0.00013298019771716435,
      "loss": 0.9981,
      "step": 1067
    },
    {
      "epoch": 1.57,
      "learning_rate": 0.0001328688451944569,
      "loss": 1.003,
      "step": 1068
    },
    {
      "epoch": 1.57,
      "learning_rate": 0.00013275744696127805,
      "loss": 1.0307,
      "step": 1069
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.00013264600317254853,
      "loss": 1.0257,
      "step": 1070
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.00013253451398325249,
      "loss": 1.0426,
      "step": 1071
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.00013242297954843711,
      "loss": 1.0167,
      "step": 1072
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.00013231140002321253,
      "loss": 1.012,
      "step": 1073
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.00013219977556275163,
      "loss": 1.0649,
      "step": 1074
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.00013208810632228977,
      "loss": 0.9297,
      "step": 1075
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.00013197639245712454,
      "loss": 0.9772,
      "step": 1076
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.00013186463412261565,
      "loss": 1.0194,
      "step": 1077
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.00013175283147418465,
      "loss": 1.0596,
      "step": 1078
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.00013164098466731468,
      "loss": 0.9938,
      "step": 1079
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.00013152909385755025,
      "loss": 0.9405,
      "step": 1080
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.0001314171592004972,
      "loss": 1.0175,
      "step": 1081
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.00013130518085182225,
      "loss": 0.9994,
      "step": 1082
    },
    {
      "epoch": 1.59,
      "learning_rate": 0.00013119315896725287,
      "loss": 0.9524,
      "step": 1083
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.00013108109370257712,
      "loss": 0.9112,
      "step": 1084
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.00013096898521364338,
      "loss": 0.9339,
      "step": 1085
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.00013085683365636014,
      "loss": 0.9718,
      "step": 1086
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.0001307446391866958,
      "loss": 0.969,
      "step": 1087
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.00013063240196067836,
      "loss": 1.0255,
      "step": 1088
    },
    {
      "epoch": 1.6,
      "learning_rate": 0.00013052012213439536,
      "loss": 1.0119,
      "step": 1089
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.00013040779986399362,
      "loss": 1.0396,
      "step": 1090
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.00013029543530567884,
      "loss": 1.0202,
      "step": 1091
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.0001301830286157157,
      "loss": 0.9024,
      "step": 1092
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.00013007057995042732,
      "loss": 1.0079,
      "step": 1093
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.0001299580894661953,
      "loss": 0.9771,
      "step": 1094
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.0001298455573194594,
      "loss": 0.9942,
      "step": 1095
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.00012973298366671725,
      "loss": 0.9879,
      "step": 1096
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.00012962036866452422,
      "loss": 0.9365,
      "step": 1097
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.0001295077124694932,
      "loss": 1.0128,
      "step": 1098
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.00012939501523829444,
      "loss": 1.0707,
      "step": 1099
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.00012928227712765504,
      "loss": 0.9769,
      "step": 1100
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.00012916949829435922,
      "loss": 1.0208,
      "step": 1101
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.0001290566788952477,
      "loss": 1.0376,
      "step": 1102
    },
    {
      "epoch": 1.62,
      "learning_rate": 0.00012894381908721756,
      "loss": 1.0588,
      "step": 1103
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.0001288309190272222,
      "loss": 1.0217,
      "step": 1104
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.00012871797887227087,
      "loss": 0.9684,
      "step": 1105
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.00012860499877942875,
      "loss": 0.9753,
      "step": 1106
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.00012849197890581638,
      "loss": 1.0094,
      "step": 1107
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.00012837891940860972,
      "loss": 1.0346,
      "step": 1108
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.00012826582044503978,
      "loss": 0.8741,
      "step": 1109
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.00012815268217239252,
      "loss": 1.0223,
      "step": 1110
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.00012803950474800862,
      "loss": 0.8748,
      "step": 1111
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.00012792628832928302,
      "loss": 1.0296,
      "step": 1112
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.000127813033073665,
      "loss": 0.9993,
      "step": 1113
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.00012769973913865794,
      "loss": 1.0555,
      "step": 1114
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.00012758640668181882,
      "loss": 1.0245,
      "step": 1115
    },
    {
      "epoch": 1.64,
      "learning_rate": 0.0001274730358607583,
      "loss": 0.9502,
      "step": 1116
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.00012735962683314042,
      "loss": 1.0165,
      "step": 1117
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.0001272461797566823,
      "loss": 1.0669,
      "step": 1118
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.000127132694789154,
      "loss": 0.8676,
      "step": 1119
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.0001270191720883782,
      "loss": 0.9432,
      "step": 1120
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.00012690561181223024,
      "loss": 1.0614,
      "step": 1121
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.0001267920141186375,
      "loss": 0.9924,
      "step": 1122
    },
    {
      "epoch": 1.65,
      "learning_rate": 0.00012667837916557954,
      "loss": 1.139,
      "step": 1123
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.00012656470711108764,
      "loss": 1.0043,
      "step": 1124
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.00012645099811324476,
      "loss": 1.0747,
      "step": 1125
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.0001263372523301852,
      "loss": 0.9668,
      "step": 1126
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.00012622346992009447,
      "loss": 0.9931,
      "step": 1127
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.00012610965104120885,
      "loss": 0.9393,
      "step": 1128
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.00012599579585181552,
      "loss": 0.9918,
      "step": 1129
    },
    {
      "epoch": 1.66,
      "learning_rate": 0.00012588190451025207,
      "loss": 1.0172,
      "step": 1130
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.00012576797717490644,
      "loss": 1.0586,
      "step": 1131
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.00012565401400421651,
      "loss": 1.0482,
      "step": 1132
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.00012554001515667008,
      "loss": 1.0548,
      "step": 1133
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.00012542598079080456,
      "loss": 1.0092,
      "step": 1134
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.00012531191106520672,
      "loss": 1.0162,
      "step": 1135
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.00012519780613851254,
      "loss": 1.0387,
      "step": 1136
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0001250836661694069,
      "loss": 0.9607,
      "step": 1137
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.00012496949131662348,
      "loss": 1.0025,
      "step": 1138
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.00012485528173894448,
      "loss": 1.0014,
      "step": 1139
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.00012474103759520027,
      "loss": 0.9838,
      "step": 1140
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.0001246267590442694,
      "loss": 1.0384,
      "step": 1141
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.00012451244624507831,
      "loss": 0.9958,
      "step": 1142
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.00012439809935660095,
      "loss": 0.9927,
      "step": 1143
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.0001242837185378587,
      "loss": 1.0082,
      "step": 1144
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.00012416930394792026,
      "loss": 0.9729,
      "step": 1145
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.00012405485574590113,
      "loss": 1.0464,
      "step": 1146
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.00012394037409096357,
      "loss": 0.987,
      "step": 1147
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.0001238258591423165,
      "loss": 0.9402,
      "step": 1148
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.00012371131105921504,
      "loss": 1.0293,
      "step": 1149
    },
    {
      "epoch": 1.69,
      "learning_rate": 0.00012359673000096033,
      "loss": 0.9418,
      "step": 1150
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0001234821161268995,
      "loss": 0.964,
      "step": 1151
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.00012336746959642526,
      "loss": 0.9982,
      "step": 1152
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0001232527905689757,
      "loss": 0.9364,
      "step": 1153
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.00012313807920403419,
      "loss": 0.9399,
      "step": 1154
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.0001230233356611289,
      "loss": 1.015,
      "step": 1155
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.000122908560099833,
      "loss": 1.0214,
      "step": 1156
    },
    {
      "epoch": 1.7,
      "learning_rate": 0.00012279375267976398,
      "loss": 1.0262,
      "step": 1157
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.00012267891356058377,
      "loss": 1.0277,
      "step": 1158
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.00012256404290199825,
      "loss": 1.0095,
      "step": 1159
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.00012244914086375724,
      "loss": 1.0314,
      "step": 1160
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.00012233420760565428,
      "loss": 0.8282,
      "step": 1161
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.00012221924328752616,
      "loss": 0.9709,
      "step": 1162
    },
    {
      "epoch": 1.71,
      "learning_rate": 0.00012210424806925301,
      "loss": 0.941,
      "step": 1163
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.00012198922211075778,
      "loss": 0.9716,
      "step": 1164
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.00012187416557200633,
      "loss": 1.0125,
      "step": 1165
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.00012175907861300697,
      "loss": 1.0159,
      "step": 1166
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.00012164396139381029,
      "loss": 0.9306,
      "step": 1167
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.00012152881407450905,
      "loss": 1.1056,
      "step": 1168
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.00012141363681523776,
      "loss": 1.0113,
      "step": 1169
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.00012129842977617265,
      "loss": 0.9983,
      "step": 1170
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.00012118319311753137,
      "loss": 1.0076,
      "step": 1171
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.00012106792699957263,
      "loss": 1.1181,
      "step": 1172
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.00012095263158259631,
      "loss": 0.8759,
      "step": 1173
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.00012083730702694291,
      "loss": 0.9855,
      "step": 1174
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.00012072195349299345,
      "loss": 1.1361,
      "step": 1175
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.00012060657114116926,
      "loss": 1.0275,
      "step": 1176
    },
    {
      "epoch": 1.73,
      "eval_loss": 0.9928944110870361,
      "eval_runtime": 2.6469,
      "eval_samples_per_second": 412.56,
      "eval_steps_per_second": 26.068,
      "step": 1176
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.0001204911601319318,
      "loss": 1.0256,
      "step": 1177
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.00012037572062578238,
      "loss": 0.9218,
      "step": 1178
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.00012026025278326187,
      "loss": 1.0394,
      "step": 1179
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.00012014475676495052,
      "loss": 1.0318,
      "step": 1180
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.00012002923273146794,
      "loss": 1.0361,
      "step": 1181
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.00011991368084347252,
      "loss": 1.0093,
      "step": 1182
    },
    {
      "epoch": 1.74,
      "learning_rate": 0.00011979810126166151,
      "loss": 0.9527,
      "step": 1183
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.00011968249414677055,
      "loss": 1.0946,
      "step": 1184
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.00011956685965957368,
      "loss": 1.0124,
      "step": 1185
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.000119451197960883,
      "loss": 1.0074,
      "step": 1186
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.00011933550921154834,
      "loss": 1.0315,
      "step": 1187
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.0001192197935724573,
      "loss": 0.9915,
      "step": 1188
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.00011910405120453476,
      "loss": 0.9823,
      "step": 1189
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.00011898828226874284,
      "loss": 1.0294,
      "step": 1190
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.00011887248692608057,
      "loss": 1.0176,
      "step": 1191
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.00011875666533758372,
      "loss": 1.0486,
      "step": 1192
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.00011864081766432456,
      "loss": 1.0237,
      "step": 1193
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.00011852494406741165,
      "loss": 1.0469,
      "step": 1194
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.00011840904470798955,
      "loss": 0.9545,
      "step": 1195
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.00011829311974723867,
      "loss": 0.9812,
      "step": 1196
    },
    {
      "epoch": 1.76,
      "learning_rate": 0.00011817716934637509,
      "loss": 1.0503,
      "step": 1197
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.0001180611936666502,
      "loss": 1.0693,
      "step": 1198
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.00011794519286935055,
      "loss": 0.9627,
      "step": 1199
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.00011782916711579759,
      "loss": 0.9728,
      "step": 1200
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.0001177131165673476,
      "loss": 1.13,
      "step": 1201
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.0001175970413853912,
      "loss": 0.9756,
      "step": 1202
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.00011748094173135337,
      "loss": 1.0069,
      "step": 1203
    },
    {
      "epoch": 1.77,
      "learning_rate": 0.00011736481776669306,
      "loss": 1.033,
      "step": 1204
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.00011724866965290302,
      "loss": 0.9906,
      "step": 1205
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.00011713249755150965,
      "loss": 1.1008,
      "step": 1206
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.00011701630162407266,
      "loss": 0.9987,
      "step": 1207
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.00011690008203218493,
      "loss": 1.0122,
      "step": 1208
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.0001167838389374722,
      "loss": 1.0495,
      "step": 1209
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.0001166675725015929,
      "loss": 0.9875,
      "step": 1210
    },
    {
      "epoch": 1.79,
      "learning_rate": 0.00011655128288623802,
      "loss": 1.0231,
      "step": 1211
    },
    {
      "epoch": 1.79,
      "learning_rate": 0.00011643497025313061,
      "loss": 0.9342,
      "step": 1212
    },
    {
      "epoch": 1.79,
      "learning_rate": 0.00011631863476402594,
      "loss": 1.1006,
      "step": 1213
    },
    {
      "epoch": 1.79,
      "learning_rate": 0.00011620227658071087,
      "loss": 0.9264,
      "step": 1214
    },
    {
      "epoch": 1.79,
      "learning_rate": 0.00011608589586500391,
      "loss": 1.1099,
      "step": 1215
    },
    {
      "epoch": 1.79,
      "learning_rate": 0.00011596949277875495,
      "loss": 1.0326,
      "step": 1216
    },
    {
      "epoch": 1.79,
      "learning_rate": 0.0001158530674838449,
      "loss": 0.9235,
      "step": 1217
    },
    {
      "epoch": 1.8,
      "learning_rate": 0.00011573662014218564,
      "loss": 1.0227,
      "step": 1218
    },
    {
      "epoch": 1.8,
      "learning_rate": 0.00011562015091571963,
      "loss": 1.0028,
      "step": 1219
    },
    {
      "epoch": 1.8,
      "learning_rate": 0.00011550365996641979,
      "loss": 1.0744,
      "step": 1220
    },
    {
      "epoch": 1.8,
      "learning_rate": 0.00011538714745628931,
      "loss": 0.9521,
      "step": 1221
    },
    {
      "epoch": 1.8,
      "learning_rate": 0.00011527061354736129,
      "loss": 1.0171,
      "step": 1222
    },
    {
      "epoch": 1.8,
      "learning_rate": 0.00011515405840169861,
      "loss": 1.0481,
      "step": 1223
    },
    {
      "epoch": 1.8,
      "learning_rate": 0.00011503748218139369,
      "loss": 1.0034,
      "step": 1224
    },
    {
      "epoch": 1.81,
      "learning_rate": 0.00011492088504856826,
      "loss": 1.1384,
      "step": 1225
    },
    {
      "epoch": 1.81,
      "learning_rate": 0.00011480426716537315,
      "loss": 1.0268,
      "step": 1226
    },
    {
      "epoch": 1.81,
      "learning_rate": 0.00011468762869398802,
      "loss": 1.003,
      "step": 1227
    },
    {
      "epoch": 1.81,
      "learning_rate": 0.00011457096979662114,
      "loss": 1.1087,
      "step": 1228
    },
    {
      "epoch": 1.81,
      "learning_rate": 0.00011445429063550926,
      "loss": 1.0809,
      "step": 1229
    },
    {
      "epoch": 1.81,
      "learning_rate": 0.00011433759137291727,
      "loss": 1.0054,
      "step": 1230
    },
    {
      "epoch": 1.82,
      "learning_rate": 0.00011422087217113795,
      "loss": 0.9416,
      "step": 1231
    },
    {
      "epoch": 1.82,
      "learning_rate": 0.00011410413319249194,
      "loss": 1.0153,
      "step": 1232
    },
    {
      "epoch": 1.82,
      "learning_rate": 0.00011398737459932728,
      "loss": 1.0622,
      "step": 1233
    },
    {
      "epoch": 1.82,
      "learning_rate": 0.00011387059655401932,
      "loss": 1.0792,
      "step": 1234
    },
    {
      "epoch": 1.82,
      "learning_rate": 0.00011375379921897051,
      "loss": 0.9822,
      "step": 1235
    },
    {
      "epoch": 1.82,
      "learning_rate": 0.00011363698275661001,
      "loss": 0.9949,
      "step": 1236
    },
    {
      "epoch": 1.82,
      "learning_rate": 0.00011352014732939369,
      "loss": 0.9653,
      "step": 1237
    },
    {
      "epoch": 1.83,
      "learning_rate": 0.00011340329309980377,
      "loss": 1.0694,
      "step": 1238
    },
    {
      "epoch": 1.83,
      "learning_rate": 0.00011328642023034857,
      "loss": 0.9925,
      "step": 1239
    },
    {
      "epoch": 1.83,
      "learning_rate": 0.00011316952888356237,
      "loss": 0.9829,
      "step": 1240
    },
    {
      "epoch": 1.83,
      "learning_rate": 0.00011305261922200519,
      "loss": 0.9659,
      "step": 1241
    },
    {
      "epoch": 1.83,
      "learning_rate": 0.00011293569140826239,
      "loss": 1.109,
      "step": 1242
    },
    {
      "epoch": 1.83,
      "learning_rate": 0.00011281874560494472,
      "loss": 1.0614,
      "step": 1243
    },
    {
      "epoch": 1.83,
      "learning_rate": 0.00011270178197468789,
      "loss": 0.9013,
      "step": 1244
    },
    {
      "epoch": 1.84,
      "learning_rate": 0.00011258480068015235,
      "loss": 1.0049,
      "step": 1245
    },
    {
      "epoch": 1.84,
      "learning_rate": 0.00011246780188402322,
      "loss": 0.9746,
      "step": 1246
    },
    {
      "epoch": 1.84,
      "learning_rate": 0.00011235078574900984,
      "loss": 1.1433,
      "step": 1247
    },
    {
      "epoch": 1.84,
      "learning_rate": 0.00011223375243784573,
      "loss": 1.0196,
      "step": 1248
    },
    {
      "epoch": 1.84,
      "learning_rate": 0.00011211670211328833,
      "loss": 0.9859,
      "step": 1249
    },
    {
      "epoch": 1.84,
      "learning_rate": 0.0001119996349381187,
      "loss": 0.9037,
      "step": 1250
    },
    {
      "epoch": 1.85,
      "learning_rate": 0.0001118825510751413,
      "loss": 1.0481,
      "step": 1251
    },
    {
      "epoch": 1.85,
      "learning_rate": 0.00011176545068718385,
      "loss": 1.0324,
      "step": 1252
    },
    {
      "epoch": 1.85,
      "learning_rate": 0.00011164833393709706,
      "loss": 1.0155,
      "step": 1253
    },
    {
      "epoch": 1.85,
      "learning_rate": 0.00011153120098775434,
      "loss": 0.967,
      "step": 1254
    },
    {
      "epoch": 1.85,
      "learning_rate": 0.00011141405200205166,
      "loss": 0.9766,
      "step": 1255
    },
    {
      "epoch": 1.85,
      "learning_rate": 0.00011129688714290729,
      "loss": 1.0075,
      "step": 1256
    },
    {
      "epoch": 1.85,
      "learning_rate": 0.00011117970657326158,
      "loss": 0.9472,
      "step": 1257
    },
    {
      "epoch": 1.86,
      "learning_rate": 0.00011106251045607674,
      "loss": 0.9949,
      "step": 1258
    },
    {
      "epoch": 1.86,
      "learning_rate": 0.00011094529895433652,
      "loss": 1.0302,
      "step": 1259
    },
    {
      "epoch": 1.86,
      "learning_rate": 0.0001108280722310462,
      "loss": 1.0538,
      "step": 1260
    },
    {
      "epoch": 1.86,
      "learning_rate": 0.00011071083044923214,
      "loss": 1.0025,
      "step": 1261
    },
    {
      "epoch": 1.86,
      "learning_rate": 0.00011059357377194161,
      "loss": 1.024,
      "step": 1262
    },
    {
      "epoch": 1.86,
      "learning_rate": 0.00011047630236224271,
      "loss": 0.9452,
      "step": 1263
    },
    {
      "epoch": 1.86,
      "learning_rate": 0.00011035901638322392,
      "loss": 1.0055,
      "step": 1264
    },
    {
      "epoch": 1.87,
      "learning_rate": 0.00011024171599799409,
      "loss": 0.9875,
      "step": 1265
    },
    {
      "epoch": 1.87,
      "learning_rate": 0.00011012440136968196,
      "loss": 0.9582,
      "step": 1266
    },
    {
      "epoch": 1.87,
      "learning_rate": 0.00011000707266143617,
      "loss": 0.9986,
      "step": 1267
    },
    {
      "epoch": 1.87,
      "learning_rate": 0.00010988973003642499,
      "loss": 1.0328,
      "step": 1268
    },
    {
      "epoch": 1.87,
      "learning_rate": 0.0001097723736578359,
      "loss": 1.0108,
      "step": 1269
    },
    {
      "epoch": 1.87,
      "learning_rate": 0.00010965500368887567,
      "loss": 0.9941,
      "step": 1270
    },
    {
      "epoch": 1.87,
      "learning_rate": 0.00010953762029276982,
      "loss": 1.0842,
      "step": 1271
    },
    {
      "epoch": 1.88,
      "learning_rate": 0.00010942022363276264,
      "loss": 0.994,
      "step": 1272
    },
    {
      "epoch": 1.88,
      "learning_rate": 0.00010930281387211683,
      "loss": 1.0151,
      "step": 1273
    },
    {
      "epoch": 1.88,
      "learning_rate": 0.00010918539117411333,
      "loss": 1.0172,
      "step": 1274
    },
    {
      "epoch": 1.88,
      "learning_rate": 0.00010906795570205104,
      "loss": 1.0698,
      "step": 1275
    },
    {
      "epoch": 1.88,
      "learning_rate": 0.00010895050761924668,
      "loss": 0.9835,
      "step": 1276
    },
    {
      "epoch": 1.88,
      "learning_rate": 0.0001088330470890344,
      "loss": 0.9461,
      "step": 1277
    },
    {
      "epoch": 1.89,
      "learning_rate": 0.00010871557427476583,
      "loss": 1.0394,
      "step": 1278
    },
    {
      "epoch": 1.89,
      "learning_rate": 0.00010859808933980948,
      "loss": 0.9639,
      "step": 1279
    },
    {
      "epoch": 1.89,
      "learning_rate": 0.00010848059244755093,
      "loss": 0.9863,
      "step": 1280
    },
    {
      "epoch": 1.89,
      "learning_rate": 0.00010836308376139221,
      "loss": 1.0728,
      "step": 1281
    },
    {
      "epoch": 1.89,
      "learning_rate": 0.00010824556344475181,
      "loss": 0.9989,
      "step": 1282
    },
    {
      "epoch": 1.89,
      "learning_rate": 0.00010812803166106444,
      "loss": 0.962,
      "step": 1283
    },
    {
      "epoch": 1.89,
      "learning_rate": 0.00010801048857378071,
      "loss": 0.8658,
      "step": 1284
    },
    {
      "epoch": 1.9,
      "learning_rate": 0.00010789293434636698,
      "loss": 1.0488,
      "step": 1285
    },
    {
      "epoch": 1.9,
      "learning_rate": 0.00010777536914230508,
      "loss": 1.0183,
      "step": 1286
    },
    {
      "epoch": 1.9,
      "learning_rate": 0.00010765779312509208,
      "loss": 0.9535,
      "step": 1287
    },
    {
      "epoch": 1.9,
      "learning_rate": 0.00010754020645824017,
      "loss": 0.9978,
      "step": 1288
    },
    {
      "epoch": 1.9,
      "learning_rate": 0.00010742260930527625,
      "loss": 0.8929,
      "step": 1289
    },
    {
      "epoch": 1.9,
      "learning_rate": 0.0001073050018297419,
      "loss": 0.9762,
      "step": 1290
    },
    {
      "epoch": 1.9,
      "learning_rate": 0.00010718738419519297,
      "loss": 1.0559,
      "step": 1291
    },
    {
      "epoch": 1.91,
      "learning_rate": 0.00010706975656519946,
      "loss": 1.0327,
      "step": 1292
    },
    {
      "epoch": 1.91,
      "learning_rate": 0.00010695211910334537,
      "loss": 1.0322,
      "step": 1293
    },
    {
      "epoch": 1.91,
      "learning_rate": 0.00010683447197322817,
      "loss": 1.0542,
      "step": 1294
    },
    {
      "epoch": 1.91,
      "learning_rate": 0.00010671681533845899,
      "loss": 1.0521,
      "step": 1295
    },
    {
      "epoch": 1.91,
      "learning_rate": 0.00010659914936266206,
      "loss": 0.9967,
      "step": 1296
    },
    {
      "epoch": 1.91,
      "learning_rate": 0.00010648147420947461,
      "loss": 1.0491,
      "step": 1297
    },
    {
      "epoch": 1.92,
      "learning_rate": 0.00010636379004254664,
      "loss": 0.9035,
      "step": 1298
    },
    {
      "epoch": 1.92,
      "learning_rate": 0.00010624609702554069,
      "loss": 1.0704,
      "step": 1299
    },
    {
      "epoch": 1.92,
      "learning_rate": 0.00010612839532213164,
      "loss": 0.9533,
      "step": 1300
    },
    {
      "epoch": 1.92,
      "learning_rate": 0.00010601068509600642,
      "loss": 1.0396,
      "step": 1301
    },
    {
      "epoch": 1.92,
      "learning_rate": 0.00010589296651086376,
      "loss": 0.9543,
      "step": 1302
    },
    {
      "epoch": 1.92,
      "learning_rate": 0.0001057752397304141,
      "loss": 1.0591,
      "step": 1303
    },
    {
      "epoch": 1.92,
      "learning_rate": 0.00010565750491837925,
      "loss": 1.1191,
      "step": 1304
    },
    {
      "epoch": 1.93,
      "learning_rate": 0.00010553976223849218,
      "loss": 0.916,
      "step": 1305
    },
    {
      "epoch": 1.93,
      "learning_rate": 0.00010542201185449678,
      "loss": 0.9732,
      "step": 1306
    },
    {
      "epoch": 1.93,
      "learning_rate": 0.00010530425393014774,
      "loss": 1.01,
      "step": 1307
    },
    {
      "epoch": 1.93,
      "learning_rate": 0.00010518648862921012,
      "loss": 0.9849,
      "step": 1308
    },
    {
      "epoch": 1.93,
      "learning_rate": 0.0001050687161154593,
      "loss": 1.0519,
      "step": 1309
    },
    {
      "epoch": 1.93,
      "learning_rate": 0.00010495093655268071,
      "loss": 1.0539,
      "step": 1310
    },
    {
      "epoch": 1.93,
      "learning_rate": 0.00010483315010466952,
      "loss": 0.9922,
      "step": 1311
    },
    {
      "epoch": 1.94,
      "learning_rate": 0.00010471535693523057,
      "loss": 1.0048,
      "step": 1312
    },
    {
      "epoch": 1.94,
      "learning_rate": 0.00010459755720817797,
      "loss": 1.0576,
      "step": 1313
    },
    {
      "epoch": 1.94,
      "learning_rate": 0.00010447975108733492,
      "loss": 1.0268,
      "step": 1314
    },
    {
      "epoch": 1.94,
      "learning_rate": 0.00010436193873653361,
      "loss": 1.0566,
      "step": 1315
    },
    {
      "epoch": 1.94,
      "learning_rate": 0.00010424412031961484,
      "loss": 1.0294,
      "step": 1316
    },
    {
      "epoch": 1.94,
      "learning_rate": 0.00010412629600042785,
      "loss": 1.0808,
      "step": 1317
    },
    {
      "epoch": 1.94,
      "learning_rate": 0.00010400846594283012,
      "loss": 1.0487,
      "step": 1318
    },
    {
      "epoch": 1.95,
      "learning_rate": 0.00010389063031068698,
      "loss": 1.04,
      "step": 1319
    },
    {
      "epoch": 1.95,
      "learning_rate": 0.00010377278926787173,
      "loss": 1.033,
      "step": 1320
    },
    {
      "epoch": 1.95,
      "learning_rate": 0.000103654942978265,
      "loss": 0.9637,
      "step": 1321
    },
    {
      "epoch": 1.95,
      "learning_rate": 0.00010353709160575489,
      "loss": 0.9665,
      "step": 1322
    },
    {
      "epoch": 1.95,
      "learning_rate": 0.00010341923531423634,
      "loss": 1.0079,
      "step": 1323
    },
    {
      "epoch": 1.95,
      "learning_rate": 0.00010330137426761135,
      "loss": 0.9989,
      "step": 1324
    },
    {
      "epoch": 1.96,
      "learning_rate": 0.00010318350862978848,
      "loss": 1.0103,
      "step": 1325
    },
    {
      "epoch": 1.96,
      "learning_rate": 0.00010306563856468253,
      "loss": 0.9872,
      "step": 1326
    },
    {
      "epoch": 1.96,
      "learning_rate": 0.00010294776423621464,
      "loss": 0.9684,
      "step": 1327
    },
    {
      "epoch": 1.96,
      "learning_rate": 0.00010282988580831183,
      "loss": 0.9745,
      "step": 1328
    },
    {
      "epoch": 1.96,
      "learning_rate": 0.00010271200344490674,
      "loss": 1.0621,
      "step": 1329
    },
    {
      "epoch": 1.96,
      "learning_rate": 0.0001025941173099376,
      "loss": 1.0639,
      "step": 1330
    },
    {
      "epoch": 1.96,
      "learning_rate": 0.00010247622756734774,
      "loss": 0.914,
      "step": 1331
    },
    {
      "epoch": 1.97,
      "learning_rate": 0.00010235833438108571,
      "loss": 1.0135,
      "step": 1332
    },
    {
      "epoch": 1.97,
      "learning_rate": 0.00010224043791510465,
      "loss": 1.0132,
      "step": 1333
    },
    {
      "epoch": 1.97,
      "learning_rate": 0.00010212253833336237,
      "loss": 0.9912,
      "step": 1334
    },
    {
      "epoch": 1.97,
      "learning_rate": 0.00010200463579982098,
      "loss": 0.9869,
      "step": 1335
    },
    {
      "epoch": 1.97,
      "learning_rate": 0.0001018867304784467,
      "loss": 0.9784,
      "step": 1336
    },
    {
      "epoch": 1.97,
      "learning_rate": 0.00010176882253320967,
      "loss": 0.9837,
      "step": 1337
    },
    {
      "epoch": 1.97,
      "learning_rate": 0.0001016509121280836,
      "loss": 1.039,
      "step": 1338
    },
    {
      "epoch": 1.98,
      "learning_rate": 0.00010153299942704566,
      "loss": 0.9984,
      "step": 1339
    },
    {
      "epoch": 1.98,
      "learning_rate": 0.00010141508459407623,
      "loss": 1.0526,
      "step": 1340
    },
    {
      "epoch": 1.98,
      "learning_rate": 0.00010129716779315862,
      "loss": 1.0581,
      "step": 1341
    },
    {
      "epoch": 1.98,
      "learning_rate": 0.0001011792491882789,
      "loss": 1.0607,
      "step": 1342
    },
    {
      "epoch": 1.98,
      "learning_rate": 0.00010106132894342564,
      "loss": 0.9734,
      "step": 1343
    },
    {
      "epoch": 1.98,
      "learning_rate": 0.00010094340722258969,
      "loss": 0.8872,
      "step": 1344
    },
    {
      "epoch": 1.98,
      "eval_loss": 0.9883129000663757,
      "eval_runtime": 2.6191,
      "eval_samples_per_second": 416.937,
      "eval_steps_per_second": 26.345,
      "step": 1344
    },
    {
      "epoch": 1.99,
      "learning_rate": 0.00010082548418976399,
      "loss": 1.0538,
      "step": 1345
    },
    {
      "epoch": 1.99,
      "learning_rate": 0.00010070756000894321,
      "loss": 0.9638,
      "step": 1346
    },
    {
      "epoch": 1.99,
      "learning_rate": 0.00010058963484412372,
      "loss": 1.0556,
      "step": 1347
    },
    {
      "epoch": 1.99,
      "learning_rate": 0.00010047170885930324,
      "loss": 1.0462,
      "step": 1348
    },
    {
      "epoch": 1.99,
      "learning_rate": 0.00010035378221848053,
      "loss": 0.9999,
      "step": 1349
    },
    {
      "epoch": 1.99,
      "learning_rate": 0.00010023585508565538,
      "loss": 0.9644,
      "step": 1350
    },
    {
      "epoch": 1.99,
      "learning_rate": 0.00010011792762482826,
      "loss": 1.0601,
      "step": 1351
    },
    {
      "epoch": 2.0,
      "learning_rate": 0.0001,
      "loss": 0.9745,
      "step": 1352
    },
    {
      "epoch": 2.0,
      "learning_rate": 9.988207237517178e-05,
      "loss": 1.0911,
      "step": 1353
    },
    {
      "epoch": 2.0,
      "learning_rate": 9.976414491434463e-05,
      "loss": 0.9607,
      "step": 1354
    },
    {
      "epoch": 2.0,
      "learning_rate": 9.964621778151946e-05,
      "loss": 1.0151,
      "step": 1355
    },
    {
      "epoch": 2.0,
      "learning_rate": 9.95282911406968e-05,
      "loss": 1.036,
      "step": 1356
    },
    {
      "epoch": 2.0,
      "learning_rate": 9.94103651558763e-05,
      "loss": 1.0518,
      "step": 1357
    },
    {
      "epoch": 2.0,
      "learning_rate": 9.929243999105682e-05,
      "loss": 0.9251,
      "step": 1358
    },
    {
      "epoch": 2.01,
      "learning_rate": 9.917451581023607e-05,
      "loss": 0.9899,
      "step": 1359
    },
    {
      "epoch": 2.01,
      "learning_rate": 9.905659277741032e-05,
      "loss": 0.9737,
      "step": 1360
    },
    {
      "epoch": 2.01,
      "learning_rate": 9.89386710565744e-05,
      "loss": 1.0575,
      "step": 1361
    },
    {
      "epoch": 2.01,
      "learning_rate": 9.882075081172112e-05,
      "loss": 1.0074,
      "step": 1362
    },
    {
      "epoch": 2.01,
      "learning_rate": 9.870283220684142e-05,
      "loss": 0.9959,
      "step": 1363
    },
    {
      "epoch": 2.01,
      "learning_rate": 9.858491540592382e-05,
      "loss": 1.039,
      "step": 1364
    },
    {
      "epoch": 2.01,
      "learning_rate": 9.846700057295435e-05,
      "loss": 1.0228,
      "step": 1365
    },
    {
      "epoch": 2.02,
      "learning_rate": 9.834908787191642e-05,
      "loss": 1.014,
      "step": 1366
    },
    {
      "epoch": 2.02,
      "learning_rate": 9.823117746679034e-05,
      "loss": 0.9919,
      "step": 1367
    },
    {
      "epoch": 2.02,
      "learning_rate": 9.811326952155331e-05,
      "loss": 1.0975,
      "step": 1368
    },
    {
      "epoch": 2.02,
      "learning_rate": 9.799536420017906e-05,
      "loss": 0.934,
      "step": 1369
    },
    {
      "epoch": 2.0,
      "learning_rate": 9.787746166663764e-05,
      "loss": 0.9068,
      "step": 1370
    },
    {
      "epoch": 2.0,
      "learning_rate": 9.775956208489536e-05,
      "loss": 0.8948,
      "step": 1371
    },
    {
      "epoch": 2.0,
      "learning_rate": 9.764166561891432e-05,
      "loss": 0.9547,
      "step": 1372
    },
    {
      "epoch": 2.01,
      "learning_rate": 9.752377243265229e-05,
      "loss": 0.933,
      "step": 1373
    },
    {
      "epoch": 2.01,
      "learning_rate": 9.740588269006246e-05,
      "loss": 0.9899,
      "step": 1374
    },
    {
      "epoch": 2.01,
      "learning_rate": 9.728799655509327e-05,
      "loss": 0.9059,
      "step": 1375
    },
    {
      "epoch": 2.01,
      "learning_rate": 9.71701141916882e-05,
      "loss": 0.9062,
      "step": 1376
    },
    {
      "epoch": 2.01,
      "learning_rate": 9.705223576378539e-05,
      "loss": 0.9101,
      "step": 1377
    },
    {
      "epoch": 2.01,
      "learning_rate": 9.69343614353175e-05,
      "loss": 0.9635,
      "step": 1378
    },
    {
      "epoch": 2.01,
      "learning_rate": 9.681649137021158e-05,
      "loss": 0.8867,
      "step": 1379
    },
    {
      "epoch": 2.02,
      "learning_rate": 9.669862573238863e-05,
      "loss": 0.968,
      "step": 1380
    },
    {
      "epoch": 2.02,
      "learning_rate": 9.658076468576368e-05,
      "loss": 0.9432,
      "step": 1381
    },
    {
      "epoch": 2.02,
      "learning_rate": 9.646290839424515e-05,
      "loss": 0.9186,
      "step": 1382
    },
    {
      "epoch": 2.02,
      "learning_rate": 9.6345057021735e-05,
      "loss": 1.038,
      "step": 1383
    },
    {
      "epoch": 2.02,
      "learning_rate": 9.622721073212832e-05,
      "loss": 0.9284,
      "step": 1384
    },
    {
      "epoch": 2.02,
      "learning_rate": 9.610936968931303e-05,
      "loss": 0.9076,
      "step": 1385
    },
    {
      "epoch": 2.03,
      "learning_rate": 9.599153405716992e-05,
      "loss": 0.9196,
      "step": 1386
    },
    {
      "epoch": 2.03,
      "learning_rate": 9.587370399957217e-05,
      "loss": 0.9589,
      "step": 1387
    },
    {
      "epoch": 2.03,
      "learning_rate": 9.57558796803852e-05,
      "loss": 0.9658,
      "step": 1388
    },
    {
      "epoch": 2.03,
      "learning_rate": 9.563806126346642e-05,
      "loss": 0.8864,
      "step": 1389
    },
    {
      "epoch": 2.03,
      "learning_rate": 9.55202489126651e-05,
      "loss": 1.0115,
      "step": 1390
    },
    {
      "epoch": 2.03,
      "learning_rate": 9.540244279182205e-05,
      "loss": 0.9058,
      "step": 1391
    },
    {
      "epoch": 2.03,
      "learning_rate": 9.528464306476945e-05,
      "loss": 0.8686,
      "step": 1392
    },
    {
      "epoch": 2.04,
      "learning_rate": 9.516684989533051e-05,
      "loss": 0.9784,
      "step": 1393
    },
    {
      "epoch": 2.04,
      "learning_rate": 9.504906344731932e-05,
      "loss": 1.0002,
      "step": 1394
    },
    {
      "epoch": 2.04,
      "learning_rate": 9.49312838845407e-05,
      "loss": 0.9873,
      "step": 1395
    },
    {
      "epoch": 2.04,
      "learning_rate": 9.48135113707899e-05,
      "loss": 0.9026,
      "step": 1396
    },
    {
      "epoch": 2.04,
      "learning_rate": 9.46957460698523e-05,
      "loss": 0.9592,
      "step": 1397
    },
    {
      "epoch": 2.04,
      "learning_rate": 9.457798814550323e-05,
      "loss": 0.8996,
      "step": 1398
    },
    {
      "epoch": 2.04,
      "learning_rate": 9.446023776150787e-05,
      "loss": 1.0221,
      "step": 1399
    },
    {
      "epoch": 2.05,
      "learning_rate": 9.434249508162076e-05,
      "loss": 0.9343,
      "step": 1400
    },
    {
      "epoch": 2.05,
      "learning_rate": 9.422476026958593e-05,
      "loss": 0.9033,
      "step": 1401
    },
    {
      "epoch": 2.05,
      "learning_rate": 9.410703348913627e-05,
      "loss": 0.9904,
      "step": 1402
    },
    {
      "epoch": 2.05,
      "learning_rate": 9.398931490399363e-05,
      "loss": 0.8898,
      "step": 1403
    },
    {
      "epoch": 2.05,
      "learning_rate": 9.38716046778684e-05,
      "loss": 0.926,
      "step": 1404
    },
    {
      "epoch": 2.05,
      "learning_rate": 9.375390297445932e-05,
      "loss": 0.9515,
      "step": 1405
    },
    {
      "epoch": 2.06,
      "learning_rate": 9.363620995745337e-05,
      "loss": 0.9963,
      "step": 1406
    },
    {
      "epoch": 2.06,
      "learning_rate": 9.351852579052543e-05,
      "loss": 0.8657,
      "step": 1407
    },
    {
      "epoch": 2.06,
      "learning_rate": 9.340085063733797e-05,
      "loss": 0.9137,
      "step": 1408
    },
    {
      "epoch": 2.06,
      "learning_rate": 9.328318466154102e-05,
      "loss": 0.9354,
      "step": 1409
    },
    {
      "epoch": 2.06,
      "learning_rate": 9.316552802677184e-05,
      "loss": 0.8961,
      "step": 1410
    },
    {
      "epoch": 2.06,
      "learning_rate": 9.304788089665466e-05,
      "loss": 0.9796,
      "step": 1411
    },
    {
      "epoch": 2.06,
      "learning_rate": 9.293024343480055e-05,
      "loss": 0.8937,
      "step": 1412
    },
    {
      "epoch": 2.07,
      "learning_rate": 9.281261580480705e-05,
      "loss": 0.9589,
      "step": 1413
    },
    {
      "epoch": 2.07,
      "learning_rate": 9.269499817025814e-05,
      "loss": 1.025,
      "step": 1414
    },
    {
      "epoch": 2.07,
      "learning_rate": 9.257739069472374e-05,
      "loss": 0.9527,
      "step": 1415
    },
    {
      "epoch": 2.07,
      "learning_rate": 9.245979354175985e-05,
      "loss": 0.9236,
      "step": 1416
    },
    {
      "epoch": 2.07,
      "learning_rate": 9.234220687490794e-05,
      "loss": 0.8775,
      "step": 1417
    },
    {
      "epoch": 2.07,
      "learning_rate": 9.222463085769494e-05,
      "loss": 0.9171,
      "step": 1418
    },
    {
      "epoch": 2.07,
      "learning_rate": 9.210706565363305e-05,
      "loss": 0.8984,
      "step": 1419
    },
    {
      "epoch": 2.08,
      "learning_rate": 9.198951142621929e-05,
      "loss": 0.8724,
      "step": 1420
    },
    {
      "epoch": 2.08,
      "learning_rate": 9.187196833893558e-05,
      "loss": 0.918,
      "step": 1421
    },
    {
      "epoch": 2.08,
      "learning_rate": 9.175443655524821e-05,
      "loss": 0.9457,
      "step": 1422
    },
    {
      "epoch": 2.08,
      "learning_rate": 9.163691623860784e-05,
      "loss": 0.9508,
      "step": 1423
    },
    {
      "epoch": 2.08,
      "learning_rate": 9.151940755244912e-05,
      "loss": 0.8994,
      "step": 1424
    },
    {
      "epoch": 2.08,
      "learning_rate": 9.14019106601905e-05,
      "loss": 0.9118,
      "step": 1425
    },
    {
      "epoch": 2.08,
      "learning_rate": 9.128442572523417e-05,
      "loss": 0.8908,
      "step": 1426
    },
    {
      "epoch": 2.09,
      "learning_rate": 9.11669529109656e-05,
      "loss": 0.9755,
      "step": 1427
    },
    {
      "epoch": 2.09,
      "learning_rate": 9.104949238075336e-05,
      "loss": 0.9706,
      "step": 1428
    },
    {
      "epoch": 2.09,
      "learning_rate": 9.093204429794898e-05,
      "loss": 0.9156,
      "step": 1429
    },
    {
      "epoch": 2.09,
      "learning_rate": 9.081460882588667e-05,
      "loss": 0.9797,
      "step": 1430
    },
    {
      "epoch": 2.09,
      "learning_rate": 9.069718612788318e-05,
      "loss": 0.9402,
      "step": 1431
    },
    {
      "epoch": 2.09,
      "learning_rate": 9.057977636723739e-05,
      "loss": 0.9339,
      "step": 1432
    },
    {
      "epoch": 2.1,
      "learning_rate": 9.04623797072302e-05,
      "loss": 0.9026,
      "step": 1433
    },
    {
      "epoch": 2.1,
      "learning_rate": 9.034499631112437e-05,
      "loss": 0.9424,
      "step": 1434
    },
    {
      "epoch": 2.1,
      "learning_rate": 9.022762634216409e-05,
      "loss": 0.948,
      "step": 1435
    },
    {
      "epoch": 2.1,
      "learning_rate": 9.011026996357503e-05,
      "loss": 0.9323,
      "step": 1436
    },
    {
      "epoch": 2.1,
      "learning_rate": 8.999292733856385e-05,
      "loss": 1.0037,
      "step": 1437
    },
    {
      "epoch": 2.1,
      "learning_rate": 8.987559863031808e-05,
      "loss": 0.9549,
      "step": 1438
    },
    {
      "epoch": 2.1,
      "learning_rate": 8.975828400200592e-05,
      "loss": 0.915,
      "step": 1439
    },
    {
      "epoch": 2.11,
      "learning_rate": 8.964098361677607e-05,
      "loss": 0.9074,
      "step": 1440
    },
    {
      "epoch": 2.11,
      "learning_rate": 8.952369763775733e-05,
      "loss": 0.843,
      "step": 1441
    },
    {
      "epoch": 2.11,
      "learning_rate": 8.94064262280584e-05,
      "loss": 0.8654,
      "step": 1442
    },
    {
      "epoch": 2.11,
      "learning_rate": 8.928916955076792e-05,
      "loss": 0.9349,
      "step": 1443
    },
    {
      "epoch": 2.11,
      "learning_rate": 8.917192776895382e-05,
      "loss": 0.9353,
      "step": 1444
    },
    {
      "epoch": 2.11,
      "learning_rate": 8.90547010456635e-05,
      "loss": 0.9137,
      "step": 1445
    },
    {
      "epoch": 2.11,
      "learning_rate": 8.89374895439233e-05,
      "loss": 0.9528,
      "step": 1446
    },
    {
      "epoch": 2.12,
      "learning_rate": 8.882029342673844e-05,
      "loss": 0.8906,
      "step": 1447
    },
    {
      "epoch": 2.12,
      "learning_rate": 8.870311285709274e-05,
      "loss": 0.9351,
      "step": 1448
    },
    {
      "epoch": 2.12,
      "learning_rate": 8.858594799794835e-05,
      "loss": 1.0043,
      "step": 1449
    },
    {
      "epoch": 2.12,
      "learning_rate": 8.846879901224566e-05,
      "loss": 0.8462,
      "step": 1450
    },
    {
      "epoch": 2.12,
      "learning_rate": 8.835166606290295e-05,
      "loss": 1.01,
      "step": 1451
    },
    {
      "epoch": 2.12,
      "learning_rate": 8.823454931281616e-05,
      "loss": 0.8858,
      "step": 1452
    },
    {
      "epoch": 2.13,
      "learning_rate": 8.811744892485871e-05,
      "loss": 0.9954,
      "step": 1453
    },
    {
      "epoch": 2.13,
      "learning_rate": 8.800036506188129e-05,
      "loss": 0.9609,
      "step": 1454
    },
    {
      "epoch": 2.13,
      "learning_rate": 8.788329788671167e-05,
      "loss": 0.8891,
      "step": 1455
    },
    {
      "epoch": 2.13,
      "learning_rate": 8.776624756215429e-05,
      "loss": 0.9669,
      "step": 1456
    },
    {
      "epoch": 2.13,
      "learning_rate": 8.76492142509902e-05,
      "loss": 0.981,
      "step": 1457
    },
    {
      "epoch": 2.13,
      "learning_rate": 8.753219811597683e-05,
      "loss": 0.9607,
      "step": 1458
    },
    {
      "epoch": 2.13,
      "learning_rate": 8.741519931984766e-05,
      "loss": 0.8752,
      "step": 1459
    },
    {
      "epoch": 2.14,
      "learning_rate": 8.729821802531212e-05,
      "loss": 0.8649,
      "step": 1460
    },
    {
      "epoch": 2.14,
      "learning_rate": 8.71812543950553e-05,
      "loss": 0.9491,
      "step": 1461
    },
    {
      "epoch": 2.14,
      "learning_rate": 8.706430859173763e-05,
      "loss": 0.9335,
      "step": 1462
    },
    {
      "epoch": 2.14,
      "learning_rate": 8.694738077799488e-05,
      "loss": 0.9752,
      "step": 1463
    },
    {
      "epoch": 2.14,
      "learning_rate": 8.683047111643763e-05,
      "loss": 0.987,
      "step": 1464
    },
    {
      "epoch": 2.14,
      "learning_rate": 8.671357976965147e-05,
      "loss": 0.9728,
      "step": 1465
    },
    {
      "epoch": 2.14,
      "learning_rate": 8.659670690019625e-05,
      "loss": 0.8742,
      "step": 1466
    },
    {
      "epoch": 2.15,
      "learning_rate": 8.647985267060635e-05,
      "loss": 0.8658,
      "step": 1467
    },
    {
      "epoch": 2.15,
      "learning_rate": 8.636301724339004e-05,
      "loss": 1.0111,
      "step": 1468
    },
    {
      "epoch": 2.15,
      "learning_rate": 8.624620078102951e-05,
      "loss": 0.865,
      "step": 1469
    },
    {
      "epoch": 2.15,
      "learning_rate": 8.612940344598067e-05,
      "loss": 0.9406,
      "step": 1470
    },
    {
      "epoch": 2.15,
      "learning_rate": 8.601262540067274e-05,
      "loss": 0.9647,
      "step": 1471
    },
    {
      "epoch": 2.15,
      "learning_rate": 8.58958668075081e-05,
      "loss": 0.9346,
      "step": 1472
    },
    {
      "epoch": 2.15,
      "learning_rate": 8.577912782886206e-05,
      "loss": 0.9357,
      "step": 1473
    },
    {
      "epoch": 2.16,
      "learning_rate": 8.566240862708274e-05,
      "loss": 0.9226,
      "step": 1474
    },
    {
      "epoch": 2.16,
      "learning_rate": 8.554570936449074e-05,
      "loss": 0.8791,
      "step": 1475
    },
    {
      "epoch": 2.16,
      "learning_rate": 8.542903020337887e-05,
      "loss": 0.9111,
      "step": 1476
    },
    {
      "epoch": 2.16,
      "learning_rate": 8.531237130601199e-05,
      "loss": 0.928,
      "step": 1477
    },
    {
      "epoch": 2.16,
      "learning_rate": 8.519573283462687e-05,
      "loss": 0.9075,
      "step": 1478
    },
    {
      "epoch": 2.16,
      "learning_rate": 8.507911495143173e-05,
      "loss": 0.9911,
      "step": 1479
    },
    {
      "epoch": 2.17,
      "learning_rate": 8.496251781860633e-05,
      "loss": 0.9687,
      "step": 1480
    },
    {
      "epoch": 2.17,
      "learning_rate": 8.484594159830141e-05,
      "loss": 0.8915,
      "step": 1481
    },
    {
      "epoch": 2.17,
      "learning_rate": 8.472938645263875e-05,
      "loss": 0.9551,
      "step": 1482
    },
    {
      "epoch": 2.17,
      "learning_rate": 8.461285254371073e-05,
      "loss": 0.902,
      "step": 1483
    },
    {
      "epoch": 2.17,
      "learning_rate": 8.449634003358022e-05,
      "loss": 0.963,
      "step": 1484
    },
    {
      "epoch": 2.17,
      "learning_rate": 8.437984908428041e-05,
      "loss": 0.8784,
      "step": 1485
    },
    {
      "epoch": 2.17,
      "learning_rate": 8.426337985781438e-05,
      "loss": 0.9793,
      "step": 1486
    },
    {
      "epoch": 2.18,
      "learning_rate": 8.414693251615512e-05,
      "loss": 0.9435,
      "step": 1487
    },
    {
      "epoch": 2.18,
      "learning_rate": 8.403050722124509e-05,
      "loss": 0.9944,
      "step": 1488
    },
    {
      "epoch": 2.18,
      "learning_rate": 8.39141041349961e-05,
      "loss": 0.9912,
      "step": 1489
    },
    {
      "epoch": 2.18,
      "learning_rate": 8.379772341928915e-05,
      "loss": 0.8934,
      "step": 1490
    },
    {
      "epoch": 2.18,
      "learning_rate": 8.36813652359741e-05,
      "loss": 0.9868,
      "step": 1491
    },
    {
      "epoch": 2.18,
      "learning_rate": 8.356502974686941e-05,
      "loss": 0.8958,
      "step": 1492
    },
    {
      "epoch": 2.18,
      "learning_rate": 8.344871711376203e-05,
      "loss": 0.9075,
      "step": 1493
    },
    {
      "epoch": 2.19,
      "learning_rate": 8.33324274984071e-05,
      "loss": 0.954,
      "step": 1494
    },
    {
      "epoch": 2.19,
      "learning_rate": 8.321616106252783e-05,
      "loss": 0.9316,
      "step": 1495
    },
    {
      "epoch": 2.19,
      "learning_rate": 8.309991796781511e-05,
      "loss": 0.9949,
      "step": 1496
    },
    {
      "epoch": 2.19,
      "learning_rate": 8.298369837592735e-05,
      "loss": 1.0344,
      "step": 1497
    },
    {
      "epoch": 2.19,
      "learning_rate": 8.286750244849038e-05,
      "loss": 0.9034,
      "step": 1498
    },
    {
      "epoch": 2.19,
      "learning_rate": 8.275133034709699e-05,
      "loss": 0.9102,
      "step": 1499
    },
    {
      "epoch": 2.2,
      "learning_rate": 8.263518223330697e-05,
      "loss": 0.9013,
      "step": 1500
    },
    {
      "epoch": 2.2,
      "learning_rate": 8.251905826864665e-05,
      "loss": 1.0105,
      "step": 1501
    },
    {
      "epoch": 2.2,
      "learning_rate": 8.240295861460881e-05,
      "loss": 0.9215,
      "step": 1502
    },
    {
      "epoch": 2.2,
      "learning_rate": 8.228688343265242e-05,
      "loss": 0.8733,
      "step": 1503
    },
    {
      "epoch": 2.2,
      "learning_rate": 8.217083288420241e-05,
      "loss": 0.9214,
      "step": 1504
    },
    {
      "epoch": 2.2,
      "learning_rate": 8.205480713064946e-05,
      "loss": 0.9535,
      "step": 1505
    },
    {
      "epoch": 2.2,
      "learning_rate": 8.193880633334982e-05,
      "loss": 0.9155,
      "step": 1506
    },
    {
      "epoch": 2.21,
      "learning_rate": 8.182283065362493e-05,
      "loss": 0.9045,
      "step": 1507
    },
    {
      "epoch": 2.21,
      "learning_rate": 8.170688025276134e-05,
      "loss": 1.0051,
      "step": 1508
    },
    {
      "epoch": 2.21,
      "learning_rate": 8.159095529201049e-05,
      "loss": 0.9856,
      "step": 1509
    },
    {
      "epoch": 2.21,
      "learning_rate": 8.147505593258837e-05,
      "loss": 0.9189,
      "step": 1510
    },
    {
      "epoch": 2.21,
      "learning_rate": 8.135918233567545e-05,
      "loss": 0.9455,
      "step": 1511
    },
    {
      "epoch": 2.21,
      "learning_rate": 8.124333466241632e-05,
      "loss": 0.9351,
      "step": 1512
    },
    {
      "epoch": 2.21,
      "eval_loss": 0.9985308051109314,
      "eval_runtime": 2.6183,
      "eval_samples_per_second": 417.062,
      "eval_steps_per_second": 26.353,
      "step": 1512
    },
    {
      "epoch": 2.21,
      "learning_rate": 8.112751307391946e-05,
      "loss": 0.9408,
      "step": 1513
    },
    {
      "epoch": 2.22,
      "learning_rate": 8.101171773125716e-05,
      "loss": 0.8725,
      "step": 1514
    },
    {
      "epoch": 2.22,
      "learning_rate": 8.089594879546525e-05,
      "loss": 0.9437,
      "step": 1515
    },
    {
      "epoch": 2.22,
      "learning_rate": 8.078020642754274e-05,
      "loss": 0.904,
      "step": 1516
    },
    {
      "epoch": 2.22,
      "learning_rate": 8.066449078845168e-05,
      "loss": 0.93,
      "step": 1517
    },
    {
      "epoch": 2.22,
      "learning_rate": 8.054880203911705e-05,
      "loss": 1.0044,
      "step": 1518
    },
    {
      "epoch": 2.22,
      "learning_rate": 8.043314034042631e-05,
      "loss": 0.957,
      "step": 1519
    },
    {
      "epoch": 2.23,
      "learning_rate": 8.031750585322947e-05,
      "loss": 0.9136,
      "step": 1520
    },
    {
      "epoch": 2.23,
      "learning_rate": 8.020189873833852e-05,
      "loss": 0.9619,
      "step": 1521
    },
    {
      "epoch": 2.23,
      "learning_rate": 8.00863191565275e-05,
      "loss": 0.961,
      "step": 1522
    },
    {
      "epoch": 2.23,
      "learning_rate": 7.99707672685321e-05,
      "loss": 0.8426,
      "step": 1523
    },
    {
      "epoch": 2.23,
      "learning_rate": 7.985524323504948e-05,
      "loss": 0.9201,
      "step": 1524
    },
    {
      "epoch": 2.23,
      "learning_rate": 7.973974721673815e-05,
      "loss": 0.8758,
      "step": 1525
    },
    {
      "epoch": 2.23,
      "learning_rate": 7.962427937421763e-05,
      "loss": 0.8926,
      "step": 1526
    },
    {
      "epoch": 2.24,
      "learning_rate": 7.950883986806821e-05,
      "loss": 0.9891,
      "step": 1527
    },
    {
      "epoch": 2.24,
      "learning_rate": 7.939342885883076e-05,
      "loss": 0.9544,
      "step": 1528
    },
    {
      "epoch": 2.24,
      "learning_rate": 7.927804650700659e-05,
      "loss": 0.9669,
      "step": 1529
    },
    {
      "epoch": 2.24,
      "learning_rate": 7.916269297305712e-05,
      "loss": 0.9712,
      "step": 1530
    },
    {
      "epoch": 2.24,
      "learning_rate": 7.90473684174037e-05,
      "loss": 0.952,
      "step": 1531
    },
    {
      "epoch": 2.24,
      "learning_rate": 7.89320730004274e-05,
      "loss": 0.9196,
      "step": 1532
    },
    {
      "epoch": 2.24,
      "learning_rate": 7.881680688246869e-05,
      "loss": 1.0468,
      "step": 1533
    },
    {
      "epoch": 2.25,
      "learning_rate": 7.870157022382735e-05,
      "loss": 0.9278,
      "step": 1534
    },
    {
      "epoch": 2.25,
      "learning_rate": 7.858636318476226e-05,
      "loss": 0.9395,
      "step": 1535
    },
    {
      "epoch": 2.25,
      "learning_rate": 7.847118592549099e-05,
      "loss": 0.9261,
      "step": 1536
    },
    {
      "epoch": 2.25,
      "learning_rate": 7.835603860618972e-05,
      "loss": 0.8694,
      "step": 1537
    },
    {
      "epoch": 2.25,
      "learning_rate": 7.824092138699307e-05,
      "loss": 1.0352,
      "step": 1538
    },
    {
      "epoch": 2.25,
      "learning_rate": 7.812583442799368e-05,
      "loss": 0.8773,
      "step": 1539
    },
    {
      "epoch": 2.25,
      "learning_rate": 7.801077788924224e-05,
      "loss": 0.8136,
      "step": 1540
    },
    {
      "epoch": 2.26,
      "learning_rate": 7.789575193074704e-05,
      "loss": 0.9875,
      "step": 1541
    },
    {
      "epoch": 2.26,
      "learning_rate": 7.778075671247385e-05,
      "loss": 0.9006,
      "step": 1542
    },
    {
      "epoch": 2.26,
      "learning_rate": 7.766579239434575e-05,
      "loss": 0.9503,
      "step": 1543
    },
    {
      "epoch": 2.26,
      "learning_rate": 7.755085913624274e-05,
      "loss": 0.9282,
      "step": 1544
    },
    {
      "epoch": 2.26,
      "learning_rate": 7.743595709800176e-05,
      "loss": 0.9057,
      "step": 1545
    },
    {
      "epoch": 2.26,
      "learning_rate": 7.732108643941625e-05,
      "loss": 0.8577,
      "step": 1546
    },
    {
      "epoch": 2.27,
      "learning_rate": 7.720624732023603e-05,
      "loss": 1.0027,
      "step": 1547
    },
    {
      "epoch": 2.27,
      "learning_rate": 7.709143990016702e-05,
      "loss": 0.8771,
      "step": 1548
    },
    {
      "epoch": 2.27,
      "learning_rate": 7.697666433887108e-05,
      "loss": 0.9449,
      "step": 1549
    },
    {
      "epoch": 2.27,
      "learning_rate": 7.686192079596586e-05,
      "loss": 0.9362,
      "step": 1550
    },
    {
      "epoch": 2.27,
      "learning_rate": 7.674720943102432e-05,
      "loss": 0.9937,
      "step": 1551
    },
    {
      "epoch": 2.27,
      "learning_rate": 7.663253040357476e-05,
      "loss": 0.9218,
      "step": 1552
    },
    {
      "epoch": 2.27,
      "learning_rate": 7.651788387310052e-05,
      "loss": 0.9557,
      "step": 1553
    },
    {
      "epoch": 2.28,
      "learning_rate": 7.640326999903967e-05,
      "loss": 0.9602,
      "step": 1554
    },
    {
      "epoch": 2.28,
      "learning_rate": 7.628868894078501e-05,
      "loss": 0.9414,
      "step": 1555
    },
    {
      "epoch": 2.28,
      "learning_rate": 7.617414085768351e-05,
      "loss": 0.8552,
      "step": 1556
    },
    {
      "epoch": 2.28,
      "learning_rate": 7.605962590903643e-05,
      "loss": 0.9802,
      "step": 1557
    },
    {
      "epoch": 2.28,
      "learning_rate": 7.594514425409893e-05,
      "loss": 0.9204,
      "step": 1558
    },
    {
      "epoch": 2.28,
      "learning_rate": 7.583069605207975e-05,
      "loss": 0.9143,
      "step": 1559
    },
    {
      "epoch": 2.28,
      "learning_rate": 7.571628146214129e-05,
      "loss": 0.9383,
      "step": 1560
    },
    {
      "epoch": 2.29,
      "learning_rate": 7.560190064339908e-05,
      "loss": 0.9052,
      "step": 1561
    },
    {
      "epoch": 2.29,
      "learning_rate": 7.548755375492172e-05,
      "loss": 0.9063,
      "step": 1562
    },
    {
      "epoch": 2.29,
      "learning_rate": 7.537324095573064e-05,
      "loss": 0.9549,
      "step": 1563
    },
    {
      "epoch": 2.29,
      "learning_rate": 7.525896240479976e-05,
      "loss": 0.9499,
      "step": 1564
    },
    {
      "epoch": 2.29,
      "learning_rate": 7.514471826105556e-05,
      "loss": 0.8888,
      "step": 1565
    },
    {
      "epoch": 2.29,
      "learning_rate": 7.503050868337653e-05,
      "loss": 0.9178,
      "step": 1566
    },
    {
      "epoch": 2.3,
      "learning_rate": 7.491633383059313e-05,
      "loss": 0.9875,
      "step": 1567
    },
    {
      "epoch": 2.3,
      "learning_rate": 7.48021938614875e-05,
      "loss": 0.8715,
      "step": 1568
    },
    {
      "epoch": 2.3,
      "learning_rate": 7.468808893479327e-05,
      "loss": 0.9246,
      "step": 1569
    },
    {
      "epoch": 2.3,
      "learning_rate": 7.457401920919544e-05,
      "loss": 0.8635,
      "step": 1570
    },
    {
      "epoch": 2.3,
      "learning_rate": 7.445998484332993e-05,
      "loss": 0.95,
      "step": 1571
    },
    {
      "epoch": 2.3,
      "learning_rate": 7.434598599578351e-05,
      "loss": 0.9175,
      "step": 1572
    },
    {
      "epoch": 2.3,
      "learning_rate": 7.42320228250936e-05,
      "loss": 0.9439,
      "step": 1573
    },
    {
      "epoch": 2.31,
      "learning_rate": 7.411809548974792e-05,
      "loss": 0.9417,
      "step": 1574
    },
    {
      "epoch": 2.31,
      "learning_rate": 7.400420414818451e-05,
      "loss": 0.9001,
      "step": 1575
    },
    {
      "epoch": 2.31,
      "learning_rate": 7.389034895879118e-05,
      "loss": 0.8844,
      "step": 1576
    },
    {
      "epoch": 2.31,
      "learning_rate": 7.37765300799056e-05,
      "loss": 0.9377,
      "step": 1577
    },
    {
      "epoch": 2.31,
      "learning_rate": 7.366274766981483e-05,
      "loss": 0.9851,
      "step": 1578
    },
    {
      "epoch": 2.31,
      "learning_rate": 7.354900188675525e-05,
      "loss": 0.965,
      "step": 1579
    },
    {
      "epoch": 2.31,
      "learning_rate": 7.343529288891239e-05,
      "loss": 0.9737,
      "step": 1580
    },
    {
      "epoch": 2.32,
      "learning_rate": 7.332162083442049e-05,
      "loss": 0.9686,
      "step": 1581
    },
    {
      "epoch": 2.32,
      "learning_rate": 7.320798588136253e-05,
      "loss": 0.968,
      "step": 1582
    },
    {
      "epoch": 2.32,
      "learning_rate": 7.309438818776981e-05,
      "loss": 0.8647,
      "step": 1583
    },
    {
      "epoch": 2.32,
      "learning_rate": 7.29808279116218e-05,
      "loss": 0.8572,
      "step": 1584
    },
    {
      "epoch": 2.32,
      "learning_rate": 7.286730521084602e-05,
      "loss": 0.9264,
      "step": 1585
    },
    {
      "epoch": 2.32,
      "learning_rate": 7.275382024331772e-05,
      "loss": 0.9601,
      "step": 1586
    },
    {
      "epoch": 2.32,
      "learning_rate": 7.264037316685962e-05,
      "loss": 0.9996,
      "step": 1587
    },
    {
      "epoch": 2.33,
      "learning_rate": 7.252696413924174e-05,
      "loss": 0.9598,
      "step": 1588
    },
    {
      "epoch": 2.33,
      "learning_rate": 7.24135933181812e-05,
      "loss": 0.8675,
      "step": 1589
    },
    {
      "epoch": 2.33,
      "learning_rate": 7.23002608613421e-05,
      "loss": 0.9298,
      "step": 1590
    },
    {
      "epoch": 2.33,
      "learning_rate": 7.218696692633501e-05,
      "loss": 1.0448,
      "step": 1591
    },
    {
      "epoch": 2.33,
      "learning_rate": 7.2073711670717e-05,
      "loss": 0.9525,
      "step": 1592
    },
    {
      "epoch": 2.33,
      "learning_rate": 7.196049525199142e-05,
      "loss": 0.9621,
      "step": 1593
    },
    {
      "epoch": 2.34,
      "learning_rate": 7.184731782760746e-05,
      "loss": 0.933,
      "step": 1594
    },
    {
      "epoch": 2.34,
      "learning_rate": 7.173417955496024e-05,
      "loss": 1.0498,
      "step": 1595
    },
    {
      "epoch": 2.34,
      "learning_rate": 7.162108059139032e-05,
      "loss": 0.9508,
      "step": 1596
    },
    {
      "epoch": 2.34,
      "learning_rate": 7.150802109418366e-05,
      "loss": 0.9854,
      "step": 1597
    },
    {
      "epoch": 2.34,
      "learning_rate": 7.13950012205713e-05,
      "loss": 0.971,
      "step": 1598
    },
    {
      "epoch": 2.34,
      "learning_rate": 7.128202112772912e-05,
      "loss": 0.9346,
      "step": 1599
    },
    {
      "epoch": 2.34,
      "learning_rate": 7.116908097277781e-05,
      "loss": 1.0434,
      "step": 1600
    },
    {
      "epoch": 2.35,
      "learning_rate": 7.105618091278245e-05,
      "loss": 0.9354,
      "step": 1601
    },
    {
      "epoch": 2.35,
      "learning_rate": 7.094332110475234e-05,
      "loss": 0.9009,
      "step": 1602
    },
    {
      "epoch": 2.35,
      "learning_rate": 7.083050170564077e-05,
      "loss": 0.9379,
      "step": 1603
    },
    {
      "epoch": 2.35,
      "learning_rate": 7.071772287234497e-05,
      "loss": 0.9488,
      "step": 1604
    },
    {
      "epoch": 2.35,
      "learning_rate": 7.06049847617056e-05,
      "loss": 0.8778,
      "step": 1605
    },
    {
      "epoch": 2.35,
      "learning_rate": 7.049228753050681e-05,
      "loss": 0.8452,
      "step": 1606
    },
    {
      "epoch": 2.35,
      "learning_rate": 7.037963133547583e-05,
      "loss": 0.8394,
      "step": 1607
    },
    {
      "epoch": 2.36,
      "learning_rate": 7.026701633328276e-05,
      "loss": 0.8895,
      "step": 1608
    },
    {
      "epoch": 2.36,
      "learning_rate": 7.015444268054059e-05,
      "loss": 0.9667,
      "step": 1609
    },
    {
      "epoch": 2.36,
      "learning_rate": 7.004191053380469e-05,
      "loss": 0.9573,
      "step": 1610
    },
    {
      "epoch": 2.36,
      "learning_rate": 6.992942004957271e-05,
      "loss": 1.0102,
      "step": 1611
    },
    {
      "epoch": 2.36,
      "learning_rate": 6.981697138428434e-05,
      "loss": 0.9507,
      "step": 1612
    },
    {
      "epoch": 2.36,
      "learning_rate": 6.970456469432117e-05,
      "loss": 0.9597,
      "step": 1613
    },
    {
      "epoch": 2.37,
      "learning_rate": 6.959220013600641e-05,
      "loss": 0.8432,
      "step": 1614
    },
    {
      "epoch": 2.37,
      "learning_rate": 6.947987786560466e-05,
      "loss": 0.9855,
      "step": 1615
    },
    {
      "epoch": 2.37,
      "learning_rate": 6.936759803932167e-05,
      "loss": 0.922,
      "step": 1616
    },
    {
      "epoch": 2.37,
      "learning_rate": 6.925536081330424e-05,
      "loss": 0.9261,
      "step": 1617
    },
    {
      "epoch": 2.37,
      "learning_rate": 6.914316634363984e-05,
      "loss": 0.9046,
      "step": 1618
    },
    {
      "epoch": 2.37,
      "learning_rate": 6.903101478635662e-05,
      "loss": 0.93,
      "step": 1619
    },
    {
      "epoch": 2.37,
      "learning_rate": 6.891890629742288e-05,
      "loss": 0.8668,
      "step": 1620
    },
    {
      "epoch": 2.38,
      "learning_rate": 6.880684103274715e-05,
      "loss": 0.9852,
      "step": 1621
    },
    {
      "epoch": 2.38,
      "learning_rate": 6.869481914817779e-05,
      "loss": 0.976,
      "step": 1622
    },
    {
      "epoch": 2.38,
      "learning_rate": 6.85828407995028e-05,
      "loss": 0.9287,
      "step": 1623
    },
    {
      "epoch": 2.38,
      "learning_rate": 6.847090614244977e-05,
      "loss": 0.9192,
      "step": 1624
    },
    {
      "epoch": 2.38,
      "learning_rate": 6.835901533268536e-05,
      "loss": 0.9999,
      "step": 1625
    },
    {
      "epoch": 2.38,
      "learning_rate": 6.824716852581539e-05,
      "loss": 0.8986,
      "step": 1626
    },
    {
      "epoch": 2.38,
      "learning_rate": 6.813536587738436e-05,
      "loss": 0.9818,
      "step": 1627
    },
    {
      "epoch": 2.39,
      "learning_rate": 6.802360754287547e-05,
      "loss": 0.9658,
      "step": 1628
    },
    {
      "epoch": 2.39,
      "learning_rate": 6.791189367771025e-05,
      "loss": 0.8793,
      "step": 1629
    },
    {
      "epoch": 2.39,
      "learning_rate": 6.780022443724839e-05,
      "loss": 0.9363,
      "step": 1630
    },
    {
      "epoch": 2.39,
      "learning_rate": 6.768859997678751e-05,
      "loss": 0.9108,
      "step": 1631
    },
    {
      "epoch": 2.39,
      "learning_rate": 6.757702045156292e-05,
      "loss": 0.9561,
      "step": 1632
    },
    {
      "epoch": 2.39,
      "learning_rate": 6.74654860167475e-05,
      "loss": 0.8761,
      "step": 1633
    },
    {
      "epoch": 2.39,
      "learning_rate": 6.735399682745145e-05,
      "loss": 0.9307,
      "step": 1634
    },
    {
      "epoch": 2.4,
      "learning_rate": 6.724255303872197e-05,
      "loss": 0.8416,
      "step": 1635
    },
    {
      "epoch": 2.4,
      "learning_rate": 6.713115480554313e-05,
      "loss": 0.908,
      "step": 1636
    },
    {
      "epoch": 2.4,
      "learning_rate": 6.701980228283568e-05,
      "loss": 0.9147,
      "step": 1637
    },
    {
      "epoch": 2.4,
      "learning_rate": 6.690849562545678e-05,
      "loss": 0.9877,
      "step": 1638
    },
    {
      "epoch": 2.4,
      "learning_rate": 6.679723498819986e-05,
      "loss": 0.9294,
      "step": 1639
    },
    {
      "epoch": 2.4,
      "learning_rate": 6.668602052579424e-05,
      "loss": 0.929,
      "step": 1640
    },
    {
      "epoch": 2.41,
      "learning_rate": 6.657485239290515e-05,
      "loss": 0.8996,
      "step": 1641
    },
    {
      "epoch": 2.41,
      "learning_rate": 6.64637307441333e-05,
      "loss": 0.8994,
      "step": 1642
    },
    {
      "epoch": 2.41,
      "learning_rate": 6.635265573401474e-05,
      "loss": 0.9253,
      "step": 1643
    },
    {
      "epoch": 2.41,
      "learning_rate": 6.624162751702076e-05,
      "loss": 0.8867,
      "step": 1644
    },
    {
      "epoch": 2.41,
      "learning_rate": 6.613064624755753e-05,
      "loss": 0.987,
      "step": 1645
    },
    {
      "epoch": 2.41,
      "learning_rate": 6.601971207996591e-05,
      "loss": 0.9079,
      "step": 1646
    },
    {
      "epoch": 2.41,
      "learning_rate": 6.590882516852122e-05,
      "loss": 0.9696,
      "step": 1647
    },
    {
      "epoch": 2.42,
      "learning_rate": 6.579798566743314e-05,
      "loss": 0.964,
      "step": 1648
    },
    {
      "epoch": 2.42,
      "learning_rate": 6.568719373084538e-05,
      "loss": 0.9736,
      "step": 1649
    },
    {
      "epoch": 2.42,
      "learning_rate": 6.557644951283551e-05,
      "loss": 0.9806,
      "step": 1650
    },
    {
      "epoch": 2.42,
      "learning_rate": 6.546575316741474e-05,
      "loss": 1.0062,
      "step": 1651
    },
    {
      "epoch": 2.42,
      "learning_rate": 6.535510484852767e-05,
      "loss": 0.8794,
      "step": 1652
    },
    {
      "epoch": 2.42,
      "learning_rate": 6.524450471005213e-05,
      "loss": 0.8359,
      "step": 1653
    },
    {
      "epoch": 2.42,
      "learning_rate": 6.513395290579901e-05,
      "loss": 0.9442,
      "step": 1654
    },
    {
      "epoch": 2.43,
      "learning_rate": 6.50234495895119e-05,
      "loss": 0.8903,
      "step": 1655
    },
    {
      "epoch": 2.43,
      "learning_rate": 6.491299491486695e-05,
      "loss": 0.9625,
      "step": 1656
    },
    {
      "epoch": 2.43,
      "learning_rate": 6.480258903547276e-05,
      "loss": 0.929,
      "step": 1657
    },
    {
      "epoch": 2.43,
      "learning_rate": 6.469223210486992e-05,
      "loss": 0.9751,
      "step": 1658
    },
    {
      "epoch": 2.43,
      "learning_rate": 6.458192427653112e-05,
      "loss": 0.982,
      "step": 1659
    },
    {
      "epoch": 2.43,
      "learning_rate": 6.447166570386063e-05,
      "loss": 0.8681,
      "step": 1660
    },
    {
      "epoch": 2.44,
      "learning_rate": 6.436145654019432e-05,
      "loss": 0.9149,
      "step": 1661
    },
    {
      "epoch": 2.44,
      "learning_rate": 6.425129693879925e-05,
      "loss": 0.9411,
      "step": 1662
    },
    {
      "epoch": 2.44,
      "learning_rate": 6.414118705287359e-05,
      "loss": 0.9072,
      "step": 1663
    },
    {
      "epoch": 2.44,
      "learning_rate": 6.403112703554643e-05,
      "loss": 0.9784,
      "step": 1664
    },
    {
      "epoch": 2.44,
      "learning_rate": 6.392111703987744e-05,
      "loss": 0.8754,
      "step": 1665
    },
    {
      "epoch": 2.44,
      "learning_rate": 6.381115721885675e-05,
      "loss": 1.0216,
      "step": 1666
    },
    {
      "epoch": 2.44,
      "learning_rate": 6.370124772540469e-05,
      "loss": 0.8493,
      "step": 1667
    },
    {
      "epoch": 2.45,
      "learning_rate": 6.35913887123716e-05,
      "loss": 0.9215,
      "step": 1668
    },
    {
      "epoch": 2.45,
      "learning_rate": 6.348158033253773e-05,
      "loss": 0.919,
      "step": 1669
    },
    {
      "epoch": 2.45,
      "learning_rate": 6.337182273861273e-05,
      "loss": 0.9128,
      "step": 1670
    },
    {
      "epoch": 2.45,
      "learning_rate": 6.326211608323573e-05,
      "loss": 0.9469,
      "step": 1671
    },
    {
      "epoch": 2.45,
      "learning_rate": 6.315246051897503e-05,
      "loss": 0.8582,
      "step": 1672
    },
    {
      "epoch": 2.45,
      "learning_rate": 6.30428561983278e-05,
      "loss": 0.8487,
      "step": 1673
    },
    {
      "epoch": 2.45,
      "learning_rate": 6.293330327372005e-05,
      "loss": 0.8687,
      "step": 1674
    },
    {
      "epoch": 2.46,
      "learning_rate": 6.282380189750625e-05,
      "loss": 0.9028,
      "step": 1675
    },
    {
      "epoch": 2.46,
      "learning_rate": 6.271435222196916e-05,
      "loss": 0.9105,
      "step": 1676
    },
    {
      "epoch": 2.46,
      "learning_rate": 6.26049543993197e-05,
      "loss": 0.9384,
      "step": 1677
    },
    {
      "epoch": 2.46,
      "learning_rate": 6.249560858169661e-05,
      "loss": 0.9658,
      "step": 1678
    },
    {
      "epoch": 2.46,
      "learning_rate": 6.238631492116644e-05,
      "loss": 0.9193,
      "step": 1679
    },
    {
      "epoch": 2.46,
      "learning_rate": 6.227707356972301e-05,
      "loss": 0.9077,
      "step": 1680
    },
    {
      "epoch": 2.46,
      "eval_loss": 0.9968231916427612,
      "eval_runtime": 2.6101,
      "eval_samples_per_second": 418.377,
      "eval_steps_per_second": 26.436,
      "step": 1680
    },
    {
      "epoch": 2.46,
      "learning_rate": 6.216788467928758e-05,
      "loss": 0.9083,
      "step": 1681
    },
    {
      "epoch": 2.47,
      "learning_rate": 6.205874840170833e-05,
      "loss": 0.9397,
      "step": 1682
    },
    {
      "epoch": 2.47,
      "learning_rate": 6.194966488876027e-05,
      "loss": 0.9631,
      "step": 1683
    },
    {
      "epoch": 2.47,
      "learning_rate": 6.184063429214515e-05,
      "loss": 0.9214,
      "step": 1684
    },
    {
      "epoch": 2.47,
      "learning_rate": 6.173165676349103e-05,
      "loss": 0.9544,
      "step": 1685
    },
    {
      "epoch": 2.47,
      "learning_rate": 6.162273245435219e-05,
      "loss": 0.9303,
      "step": 1686
    },
    {
      "epoch": 2.47,
      "learning_rate": 6.151386151620887e-05,
      "loss": 0.8605,
      "step": 1687
    },
    {
      "epoch": 2.48,
      "learning_rate": 6.140504410046712e-05,
      "loss": 0.9125,
      "step": 1688
    },
    {
      "epoch": 2.48,
      "learning_rate": 6.129628035845861e-05,
      "loss": 0.9202,
      "step": 1689
    },
    {
      "epoch": 2.48,
      "learning_rate": 6.118757044144025e-05,
      "loss": 0.937,
      "step": 1690
    },
    {
      "epoch": 2.48,
      "learning_rate": 6.107891450059419e-05,
      "loss": 0.8941,
      "step": 1691
    },
    {
      "epoch": 2.48,
      "learning_rate": 6.097031268702746e-05,
      "loss": 0.9205,
      "step": 1692
    },
    {
      "epoch": 2.48,
      "learning_rate": 6.086176515177182e-05,
      "loss": 0.9164,
      "step": 1693
    },
    {
      "epoch": 2.48,
      "learning_rate": 6.0753272045783625e-05,
      "loss": 0.9261,
      "step": 1694
    },
    {
      "epoch": 2.49,
      "learning_rate": 6.0644833519943425e-05,
      "loss": 0.8642,
      "step": 1695
    },
    {
      "epoch": 2.49,
      "learning_rate": 6.053644972505593e-05,
      "loss": 0.9648,
      "step": 1696
    },
    {
      "epoch": 2.49,
      "learning_rate": 6.042812081184972e-05,
      "loss": 0.9213,
      "step": 1697
    },
    {
      "epoch": 2.49,
      "learning_rate": 6.0319846930977e-05,
      "loss": 0.9314,
      "step": 1698
    },
    {
      "epoch": 2.49,
      "learning_rate": 6.021162823301358e-05,
      "loss": 0.9055,
      "step": 1699
    },
    {
      "epoch": 2.49,
      "learning_rate": 6.010346486845837e-05,
      "loss": 0.9335,
      "step": 1700
    },
    {
      "epoch": 2.49,
      "learning_rate": 5.9995356987733466e-05,
      "loss": 0.9365,
      "step": 1701
    },
    {
      "epoch": 2.5,
      "learning_rate": 5.988730474118367e-05,
      "loss": 0.9817,
      "step": 1702
    },
    {
      "epoch": 2.5,
      "learning_rate": 5.977930827907649e-05,
      "loss": 0.9145,
      "step": 1703
    },
    {
      "epoch": 2.5,
      "learning_rate": 5.967136775160187e-05,
      "loss": 0.9974,
      "step": 1704
    },
    {
      "epoch": 2.5,
      "learning_rate": 5.956348330887196e-05,
      "loss": 0.9577,
      "step": 1705
    },
    {
      "epoch": 2.5,
      "learning_rate": 5.945565510092086e-05,
      "loss": 0.8716,
      "step": 1706
    },
    {
      "epoch": 2.5,
      "learning_rate": 5.93478832777045e-05,
      "loss": 0.9541,
      "step": 1707
    },
    {
      "epoch": 2.51,
      "learning_rate": 5.924016798910037e-05,
      "loss": 0.9464,
      "step": 1708
    },
    {
      "epoch": 2.51,
      "learning_rate": 5.913250938490744e-05,
      "loss": 0.9774,
      "step": 1709
    },
    {
      "epoch": 2.51,
      "learning_rate": 5.9024907614845716e-05,
      "loss": 0.8654,
      "step": 1710
    },
    {
      "epoch": 2.51,
      "learning_rate": 5.891736282855622e-05,
      "loss": 0.8559,
      "step": 1711
    },
    {
      "epoch": 2.51,
      "learning_rate": 5.880987517560075e-05,
      "loss": 1.0032,
      "step": 1712
    },
    {
      "epoch": 2.51,
      "learning_rate": 5.870244480546159e-05,
      "loss": 0.9373,
      "step": 1713
    },
    {
      "epoch": 2.51,
      "learning_rate": 5.859507186754146e-05,
      "loss": 0.8892,
      "step": 1714
    },
    {
      "epoch": 2.52,
      "learning_rate": 5.848775651116309e-05,
      "loss": 0.8809,
      "step": 1715
    },
    {
      "epoch": 2.52,
      "learning_rate": 5.838049888556925e-05,
      "loss": 0.9596,
      "step": 1716
    },
    {
      "epoch": 2.52,
      "learning_rate": 5.827329913992232e-05,
      "loss": 0.95,
      "step": 1717
    },
    {
      "epoch": 2.52,
      "learning_rate": 5.81661574233042e-05,
      "loss": 0.9694,
      "step": 1718
    },
    {
      "epoch": 2.52,
      "learning_rate": 5.80590738847162e-05,
      "loss": 0.9333,
      "step": 1719
    },
    {
      "epoch": 2.52,
      "learning_rate": 5.79520486730786e-05,
      "loss": 0.943,
      "step": 1720
    },
    {
      "epoch": 2.52,
      "learning_rate": 5.784508193723057e-05,
      "loss": 0.906,
      "step": 1721
    },
    {
      "epoch": 2.53,
      "learning_rate": 5.773817382593008e-05,
      "loss": 0.9575,
      "step": 1722
    },
    {
      "epoch": 2.53,
      "learning_rate": 5.763132448785339e-05,
      "loss": 0.8638,
      "step": 1723
    },
    {
      "epoch": 2.53,
      "learning_rate": 5.752453407159522e-05,
      "loss": 0.9667,
      "step": 1724
    },
    {
      "epoch": 2.53,
      "learning_rate": 5.741780272566821e-05,
      "loss": 1.0119,
      "step": 1725
    },
    {
      "epoch": 2.53,
      "learning_rate": 5.7311130598502885e-05,
      "loss": 0.8563,
      "step": 1726
    },
    {
      "epoch": 2.53,
      "learning_rate": 5.7204517838447405e-05,
      "loss": 0.9852,
      "step": 1727
    },
    {
      "epoch": 2.54,
      "learning_rate": 5.7097964593767375e-05,
      "loss": 0.8659,
      "step": 1728
    },
    {
      "epoch": 2.54,
      "learning_rate": 5.699147101264566e-05,
      "loss": 0.9579,
      "step": 1729
    },
    {
      "epoch": 2.54,
      "learning_rate": 5.688503724318217e-05,
      "loss": 0.9226,
      "step": 1730
    },
    {
      "epoch": 2.54,
      "learning_rate": 5.6778663433393574e-05,
      "loss": 0.9771,
      "step": 1731
    },
    {
      "epoch": 2.54,
      "learning_rate": 5.667234973121317e-05,
      "loss": 0.8307,
      "step": 1732
    },
    {
      "epoch": 2.54,
      "learning_rate": 5.6566096284490635e-05,
      "loss": 0.9012,
      "step": 1733
    },
    {
      "epoch": 2.54,
      "learning_rate": 5.645990324099197e-05,
      "loss": 0.8566,
      "step": 1734
    },
    {
      "epoch": 2.55,
      "learning_rate": 5.635377074839907e-05,
      "loss": 0.941,
      "step": 1735
    },
    {
      "epoch": 2.55,
      "learning_rate": 5.624769895430961e-05,
      "loss": 0.8925,
      "step": 1736
    },
    {
      "epoch": 2.55,
      "learning_rate": 5.614168800623687e-05,
      "loss": 0.9387,
      "step": 1737
    },
    {
      "epoch": 2.55,
      "learning_rate": 5.6035738051609555e-05,
      "loss": 0.9025,
      "step": 1738
    },
    {
      "epoch": 2.55,
      "learning_rate": 5.5929849237771556e-05,
      "loss": 0.9394,
      "step": 1739
    },
    {
      "epoch": 2.55,
      "learning_rate": 5.5824021711981686e-05,
      "loss": 0.9743,
      "step": 1740
    },
    {
      "epoch": 2.55,
      "learning_rate": 5.5718255621413526e-05,
      "loss": 0.9604,
      "step": 1741
    },
    {
      "epoch": 2.56,
      "learning_rate": 5.561255111315524e-05,
      "loss": 0.9223,
      "step": 1742
    },
    {
      "epoch": 2.56,
      "learning_rate": 5.550690833420928e-05,
      "loss": 0.8748,
      "step": 1743
    },
    {
      "epoch": 2.56,
      "learning_rate": 5.540132743149242e-05,
      "loss": 1.0009,
      "step": 1744
    },
    {
      "epoch": 2.56,
      "learning_rate": 5.5295808551835184e-05,
      "loss": 0.9533,
      "step": 1745
    },
    {
      "epoch": 2.56,
      "learning_rate": 5.5190351841982014e-05,
      "loss": 0.9026,
      "step": 1746
    },
    {
      "epoch": 2.56,
      "learning_rate": 5.508495744859077e-05,
      "loss": 0.9514,
      "step": 1747
    },
    {
      "epoch": 2.56,
      "learning_rate": 5.497962551823266e-05,
      "loss": 0.9328,
      "step": 1748
    },
    {
      "epoch": 2.57,
      "learning_rate": 5.487435619739214e-05,
      "loss": 0.9765,
      "step": 1749
    },
    {
      "epoch": 2.57,
      "learning_rate": 5.476914963246647e-05,
      "loss": 0.8417,
      "step": 1750
    },
    {
      "epoch": 2.57,
      "learning_rate": 5.4664005969765674e-05,
      "loss": 0.93,
      "step": 1751
    },
    {
      "epoch": 2.57,
      "learning_rate": 5.4558925355512256e-05,
      "loss": 0.8757,
      "step": 1752
    },
    {
      "epoch": 2.57,
      "learning_rate": 5.445390793584115e-05,
      "loss": 0.9529,
      "step": 1753
    },
    {
      "epoch": 2.57,
      "learning_rate": 5.434895385679937e-05,
      "loss": 0.8961,
      "step": 1754
    },
    {
      "epoch": 2.58,
      "learning_rate": 5.4244063264345745e-05,
      "loss": 0.844,
      "step": 1755
    },
    {
      "epoch": 2.58,
      "learning_rate": 5.4139236304350935e-05,
      "loss": 0.9071,
      "step": 1756
    },
    {
      "epoch": 2.58,
      "learning_rate": 5.403447312259702e-05,
      "loss": 0.8767,
      "step": 1757
    },
    {
      "epoch": 2.58,
      "learning_rate": 5.392977386477738e-05,
      "loss": 0.9043,
      "step": 1758
    },
    {
      "epoch": 2.58,
      "learning_rate": 5.382513867649663e-05,
      "loss": 1.0033,
      "step": 1759
    },
    {
      "epoch": 2.58,
      "learning_rate": 5.372056770327013e-05,
      "loss": 0.9496,
      "step": 1760
    },
    {
      "epoch": 2.58,
      "learning_rate": 5.361606109052397e-05,
      "loss": 0.9717,
      "step": 1761
    },
    {
      "epoch": 2.59,
      "learning_rate": 5.3511618983594845e-05,
      "loss": 0.9332,
      "step": 1762
    },
    {
      "epoch": 2.59,
      "learning_rate": 5.340724152772956e-05,
      "loss": 0.8972,
      "step": 1763
    },
    {
      "epoch": 2.59,
      "learning_rate": 5.33029288680852e-05,
      "loss": 1.0025,
      "step": 1764
    },
    {
      "epoch": 2.59,
      "learning_rate": 5.31986811497286e-05,
      "loss": 0.9639,
      "step": 1765
    },
    {
      "epoch": 2.59,
      "learning_rate": 5.309449851763633e-05,
      "loss": 0.9181,
      "step": 1766
    },
    {
      "epoch": 2.59,
      "learning_rate": 5.299038111669444e-05,
      "loss": 0.9608,
      "step": 1767
    },
    {
      "epoch": 2.59,
      "learning_rate": 5.288632909169823e-05,
      "loss": 0.9961,
      "step": 1768
    },
    {
      "epoch": 2.6,
      "learning_rate": 5.2782342587352154e-05,
      "loss": 0.9709,
      "step": 1769
    },
    {
      "epoch": 2.6,
      "learning_rate": 5.267842174826955e-05,
      "loss": 1.0208,
      "step": 1770
    },
    {
      "epoch": 2.6,
      "learning_rate": 5.2574566718972364e-05,
      "loss": 0.8965,
      "step": 1771
    },
    {
      "epoch": 2.6,
      "learning_rate": 5.247077764389099e-05,
      "loss": 0.9577,
      "step": 1772
    },
    {
      "epoch": 2.6,
      "learning_rate": 5.236705466736428e-05,
      "loss": 0.9513,
      "step": 1773
    },
    {
      "epoch": 2.6,
      "learning_rate": 5.226339793363898e-05,
      "loss": 0.995,
      "step": 1774
    },
    {
      "epoch": 2.61,
      "learning_rate": 5.215980758686978e-05,
      "loss": 0.8956,
      "step": 1775
    },
    {
      "epoch": 2.61,
      "learning_rate": 5.205628377111902e-05,
      "loss": 0.9213,
      "step": 1776
    },
    {
      "epoch": 2.61,
      "learning_rate": 5.195282663035661e-05,
      "loss": 0.9413,
      "step": 1777
    },
    {
      "epoch": 2.61,
      "learning_rate": 5.18494363084596e-05,
      "loss": 0.9791,
      "step": 1778
    },
    {
      "epoch": 2.61,
      "learning_rate": 5.174611294921224e-05,
      "loss": 0.9651,
      "step": 1779
    },
    {
      "epoch": 2.61,
      "learning_rate": 5.1642856696305575e-05,
      "loss": 0.9441,
      "step": 1780
    },
    {
      "epoch": 2.61,
      "learning_rate": 5.1539667693337335e-05,
      "loss": 0.9206,
      "step": 1781
    },
    {
      "epoch": 2.62,
      "learning_rate": 5.143654608381172e-05,
      "loss": 0.8989,
      "step": 1782
    },
    {
      "epoch": 2.62,
      "learning_rate": 5.133349201113929e-05,
      "loss": 0.9114,
      "step": 1783
    },
    {
      "epoch": 2.62,
      "learning_rate": 5.123050561863657e-05,
      "loss": 0.8995,
      "step": 1784
    },
    {
      "epoch": 2.62,
      "learning_rate": 5.112758704952598e-05,
      "loss": 0.952,
      "step": 1785
    },
    {
      "epoch": 2.62,
      "learning_rate": 5.1024736446935754e-05,
      "loss": 0.8444,
      "step": 1786
    },
    {
      "epoch": 2.62,
      "learning_rate": 5.092195395389937e-05,
      "loss": 0.9613,
      "step": 1787
    },
    {
      "epoch": 2.62,
      "learning_rate": 5.081923971335582e-05,
      "loss": 0.9385,
      "step": 1788
    },
    {
      "epoch": 2.63,
      "learning_rate": 5.071659386814907e-05,
      "loss": 0.8304,
      "step": 1789
    },
    {
      "epoch": 2.63,
      "learning_rate": 5.061401656102791e-05,
      "loss": 0.9589,
      "step": 1790
    },
    {
      "epoch": 2.63,
      "learning_rate": 5.051150793464592e-05,
      "loss": 0.9628,
      "step": 1791
    },
    {
      "epoch": 2.63,
      "learning_rate": 5.0409068131561067e-05,
      "loss": 0.9054,
      "step": 1792
    },
    {
      "epoch": 2.63,
      "learning_rate": 5.0306697294235714e-05,
      "loss": 0.8631,
      "step": 1793
    },
    {
      "epoch": 2.63,
      "learning_rate": 5.020439556503629e-05,
      "loss": 0.8949,
      "step": 1794
    },
    {
      "epoch": 2.63,
      "learning_rate": 5.0102163086233065e-05,
      "loss": 1.0095,
      "step": 1795
    },
    {
      "epoch": 2.64,
      "learning_rate": 5.000000000000002e-05,
      "loss": 1.0535,
      "step": 1796
    },
    {
      "epoch": 2.64,
      "learning_rate": 4.98979064484146e-05,
      "loss": 0.9593,
      "step": 1797
    },
    {
      "epoch": 2.64,
      "learning_rate": 4.979588257345766e-05,
      "loss": 0.9763,
      "step": 1798
    },
    {
      "epoch": 2.64,
      "learning_rate": 4.969392851701305e-05,
      "loss": 0.9788,
      "step": 1799
    },
    {
      "epoch": 2.64,
      "learning_rate": 4.959204442086753e-05,
      "loss": 0.9516,
      "step": 1800
    },
    {
      "epoch": 2.64,
      "learning_rate": 4.949023042671066e-05,
      "loss": 0.9403,
      "step": 1801
    },
    {
      "epoch": 2.65,
      "learning_rate": 4.938848667613436e-05,
      "loss": 0.9843,
      "step": 1802
    },
    {
      "epoch": 2.65,
      "learning_rate": 4.928681331063304e-05,
      "loss": 1.0064,
      "step": 1803
    },
    {
      "epoch": 2.65,
      "learning_rate": 4.918521047160308e-05,
      "loss": 0.9603,
      "step": 1804
    },
    {
      "epoch": 2.65,
      "learning_rate": 4.908367830034284e-05,
      "loss": 0.8546,
      "step": 1805
    },
    {
      "epoch": 2.65,
      "learning_rate": 4.8982216938052394e-05,
      "loss": 0.9444,
      "step": 1806
    },
    {
      "epoch": 2.65,
      "learning_rate": 4.888082652583331e-05,
      "loss": 0.9918,
      "step": 1807
    },
    {
      "epoch": 2.65,
      "learning_rate": 4.877950720468859e-05,
      "loss": 0.8489,
      "step": 1808
    },
    {
      "epoch": 2.66,
      "learning_rate": 4.8678259115522215e-05,
      "loss": 1.0195,
      "step": 1809
    },
    {
      "epoch": 2.66,
      "learning_rate": 4.8577082399139296e-05,
      "loss": 0.9962,
      "step": 1810
    },
    {
      "epoch": 2.66,
      "learning_rate": 4.8475977196245504e-05,
      "loss": 0.9836,
      "step": 1811
    },
    {
      "epoch": 2.66,
      "learning_rate": 4.837494364744711e-05,
      "loss": 0.8597,
      "step": 1812
    },
    {
      "epoch": 2.66,
      "learning_rate": 4.827398189325085e-05,
      "loss": 1.002,
      "step": 1813
    },
    {
      "epoch": 2.66,
      "learning_rate": 4.817309207406346e-05,
      "loss": 0.9228,
      "step": 1814
    },
    {
      "epoch": 2.66,
      "learning_rate": 4.8072274330191725e-05,
      "loss": 0.9954,
      "step": 1815
    },
    {
      "epoch": 2.67,
      "learning_rate": 4.7971528801842116e-05,
      "loss": 0.9124,
      "step": 1816
    },
    {
      "epoch": 2.67,
      "learning_rate": 4.787085562912076e-05,
      "loss": 0.925,
      "step": 1817
    },
    {
      "epoch": 2.67,
      "learning_rate": 4.777025495203319e-05,
      "loss": 0.9245,
      "step": 1818
    },
    {
      "epoch": 2.67,
      "learning_rate": 4.7669726910484e-05,
      "loss": 0.8668,
      "step": 1819
    },
    {
      "epoch": 2.67,
      "learning_rate": 4.756927164427685e-05,
      "loss": 1.0186,
      "step": 1820
    },
    {
      "epoch": 2.67,
      "learning_rate": 4.746888929311415e-05,
      "loss": 0.95,
      "step": 1821
    },
    {
      "epoch": 2.68,
      "learning_rate": 4.7368579996596904e-05,
      "loss": 0.9373,
      "step": 1822
    },
    {
      "epoch": 2.68,
      "learning_rate": 4.726834389422461e-05,
      "loss": 0.9235,
      "step": 1823
    },
    {
      "epoch": 2.68,
      "learning_rate": 4.716818112539485e-05,
      "loss": 0.8682,
      "step": 1824
    },
    {
      "epoch": 2.68,
      "learning_rate": 4.706809182940334e-05,
      "loss": 0.9166,
      "step": 1825
    },
    {
      "epoch": 2.68,
      "learning_rate": 4.6968076145443515e-05,
      "loss": 0.9526,
      "step": 1826
    },
    {
      "epoch": 2.68,
      "learning_rate": 4.686813421260646e-05,
      "loss": 0.8961,
      "step": 1827
    },
    {
      "epoch": 2.68,
      "learning_rate": 4.6768266169880804e-05,
      "loss": 0.9258,
      "step": 1828
    },
    {
      "epoch": 2.69,
      "learning_rate": 4.666847215615226e-05,
      "loss": 0.9526,
      "step": 1829
    },
    {
      "epoch": 2.69,
      "learning_rate": 4.656875231020368e-05,
      "loss": 0.9606,
      "step": 1830
    },
    {
      "epoch": 2.69,
      "learning_rate": 4.6469106770714745e-05,
      "loss": 0.9352,
      "step": 1831
    },
    {
      "epoch": 2.69,
      "learning_rate": 4.636953567626177e-05,
      "loss": 0.9786,
      "step": 1832
    },
    {
      "epoch": 2.69,
      "learning_rate": 4.6270039165317605e-05,
      "loss": 0.8797,
      "step": 1833
    },
    {
      "epoch": 2.69,
      "learning_rate": 4.617061737625139e-05,
      "loss": 0.9609,
      "step": 1834
    },
    {
      "epoch": 2.69,
      "learning_rate": 4.6071270447328276e-05,
      "loss": 0.8972,
      "step": 1835
    },
    {
      "epoch": 2.7,
      "learning_rate": 4.597199851670932e-05,
      "loss": 0.9233,
      "step": 1836
    },
    {
      "epoch": 2.7,
      "learning_rate": 4.587280172245129e-05,
      "loss": 0.9455,
      "step": 1837
    },
    {
      "epoch": 2.7,
      "learning_rate": 4.57736802025065e-05,
      "loss": 0.8729,
      "step": 1838
    },
    {
      "epoch": 2.7,
      "learning_rate": 4.567463409472255e-05,
      "loss": 0.8894,
      "step": 1839
    },
    {
      "epoch": 2.7,
      "learning_rate": 4.557566353684209e-05,
      "loss": 0.9784,
      "step": 1840
    },
    {
      "epoch": 2.7,
      "learning_rate": 4.547676866650289e-05,
      "loss": 0.9586,
      "step": 1841
    },
    {
      "epoch": 2.7,
      "learning_rate": 4.537794962123726e-05,
      "loss": 0.978,
      "step": 1842
    },
    {
      "epoch": 2.71,
      "learning_rate": 4.527920653847221e-05,
      "loss": 0.9431,
      "step": 1843
    },
    {
      "epoch": 2.71,
      "learning_rate": 4.518053955552903e-05,
      "loss": 0.8657,
      "step": 1844
    },
    {
      "epoch": 2.71,
      "learning_rate": 4.50819488096232e-05,
      "loss": 0.7781,
      "step": 1845
    },
    {
      "epoch": 2.71,
      "learning_rate": 4.498343443786416e-05,
      "loss": 0.9086,
      "step": 1846
    },
    {
      "epoch": 2.71,
      "learning_rate": 4.488499657725511e-05,
      "loss": 0.897,
      "step": 1847
    },
    {
      "epoch": 2.71,
      "learning_rate": 4.478663536469296e-05,
      "loss": 0.9494,
      "step": 1848
    },
    {
      "epoch": 2.71,
      "eval_loss": 0.9907075762748718,
      "eval_runtime": 2.6095,
      "eval_samples_per_second": 418.473,
      "eval_steps_per_second": 26.442,
      "step": 1848
    },
    {
      "epoch": 2.72,
      "learning_rate": 4.468835093696796e-05,
      "loss": 0.9201,
      "step": 1849
    },
    {
      "epoch": 2.72,
      "learning_rate": 4.4590143430763555e-05,
      "loss": 0.989,
      "step": 1850
    },
    {
      "epoch": 2.72,
      "learning_rate": 4.449201298265622e-05,
      "loss": 0.8615,
      "step": 1851
    },
    {
      "epoch": 2.72,
      "learning_rate": 4.4393959729115244e-05,
      "loss": 0.9388,
      "step": 1852
    },
    {
      "epoch": 2.72,
      "learning_rate": 4.429598380650266e-05,
      "loss": 0.9179,
      "step": 1853
    },
    {
      "epoch": 2.72,
      "learning_rate": 4.419808535107287e-05,
      "loss": 0.9471,
      "step": 1854
    },
    {
      "epoch": 2.72,
      "learning_rate": 4.4100264498972564e-05,
      "loss": 0.9584,
      "step": 1855
    },
    {
      "epoch": 2.73,
      "learning_rate": 4.4002521386240466e-05,
      "loss": 0.9077,
      "step": 1856
    },
    {
      "epoch": 2.73,
      "learning_rate": 4.3904856148807284e-05,
      "loss": 0.876,
      "step": 1857
    },
    {
      "epoch": 2.73,
      "learning_rate": 4.3807268922495406e-05,
      "loss": 0.9785,
      "step": 1858
    },
    {
      "epoch": 2.73,
      "learning_rate": 4.370975984301866e-05,
      "loss": 0.8762,
      "step": 1859
    },
    {
      "epoch": 2.73,
      "learning_rate": 4.3612329045982236e-05,
      "loss": 0.8797,
      "step": 1860
    },
    {
      "epoch": 2.73,
      "learning_rate": 4.351497666688246e-05,
      "loss": 1.05,
      "step": 1861
    },
    {
      "epoch": 2.73,
      "learning_rate": 4.341770284110656e-05,
      "loss": 0.9844,
      "step": 1862
    },
    {
      "epoch": 2.74,
      "learning_rate": 4.332050770393263e-05,
      "loss": 0.9814,
      "step": 1863
    },
    {
      "epoch": 2.74,
      "learning_rate": 4.322339139052921e-05,
      "loss": 0.7994,
      "step": 1864
    },
    {
      "epoch": 2.74,
      "learning_rate": 4.312635403595532e-05,
      "loss": 0.8213,
      "step": 1865
    },
    {
      "epoch": 2.74,
      "learning_rate": 4.3029395775160106e-05,
      "loss": 0.9036,
      "step": 1866
    },
    {
      "epoch": 2.74,
      "learning_rate": 4.293251674298269e-05,
      "loss": 0.9382,
      "step": 1867
    },
    {
      "epoch": 2.74,
      "learning_rate": 4.283571707415214e-05,
      "loss": 0.9933,
      "step": 1868
    },
    {
      "epoch": 2.75,
      "learning_rate": 4.273899690328702e-05,
      "loss": 0.897,
      "step": 1869
    },
    {
      "epoch": 2.75,
      "learning_rate": 4.264235636489542e-05,
      "loss": 0.9293,
      "step": 1870
    },
    {
      "epoch": 2.75,
      "learning_rate": 4.2545795593374594e-05,
      "loss": 0.9417,
      "step": 1871
    },
    {
      "epoch": 2.75,
      "learning_rate": 4.244931472301098e-05,
      "loss": 0.8857,
      "step": 1872
    },
    {
      "epoch": 2.75,
      "learning_rate": 4.235291388797986e-05,
      "loss": 0.8865,
      "step": 1873
    },
    {
      "epoch": 2.75,
      "learning_rate": 4.2256593222345185e-05,
      "loss": 0.8899,
      "step": 1874
    },
    {
      "epoch": 2.75,
      "learning_rate": 4.216035286005942e-05,
      "loss": 0.9032,
      "step": 1875
    },
    {
      "epoch": 2.76,
      "learning_rate": 4.206419293496333e-05,
      "loss": 0.9283,
      "step": 1876
    },
    {
      "epoch": 2.76,
      "learning_rate": 4.196811358078585e-05,
      "loss": 0.852,
      "step": 1877
    },
    {
      "epoch": 2.76,
      "learning_rate": 4.18721149311439e-05,
      "loss": 0.9111,
      "step": 1878
    },
    {
      "epoch": 2.76,
      "learning_rate": 4.177619711954211e-05,
      "loss": 0.9561,
      "step": 1879
    },
    {
      "epoch": 2.76,
      "learning_rate": 4.168036027937267e-05,
      "loss": 0.8661,
      "step": 1880
    },
    {
      "epoch": 2.76,
      "learning_rate": 4.1584604543915254e-05,
      "loss": 0.9675,
      "step": 1881
    },
    {
      "epoch": 2.76,
      "learning_rate": 4.148893004633663e-05,
      "loss": 0.9231,
      "step": 1882
    },
    {
      "epoch": 2.77,
      "learning_rate": 4.139333691969071e-05,
      "loss": 0.8708,
      "step": 1883
    },
    {
      "epoch": 2.77,
      "learning_rate": 4.129782529691815e-05,
      "loss": 0.9466,
      "step": 1884
    },
    {
      "epoch": 2.77,
      "learning_rate": 4.1202395310846296e-05,
      "loss": 0.8492,
      "step": 1885
    },
    {
      "epoch": 2.77,
      "learning_rate": 4.1107047094188946e-05,
      "loss": 0.8113,
      "step": 1886
    },
    {
      "epoch": 2.77,
      "learning_rate": 4.101178077954617e-05,
      "loss": 0.8231,
      "step": 1887
    },
    {
      "epoch": 2.77,
      "learning_rate": 4.091659649940419e-05,
      "loss": 1.025,
      "step": 1888
    },
    {
      "epoch": 2.77,
      "learning_rate": 4.082149438613514e-05,
      "loss": 0.9734,
      "step": 1889
    },
    {
      "epoch": 2.78,
      "learning_rate": 4.072647457199684e-05,
      "loss": 0.9829,
      "step": 1890
    },
    {
      "epoch": 2.78,
      "learning_rate": 4.063153718913267e-05,
      "loss": 0.9899,
      "step": 1891
    },
    {
      "epoch": 2.78,
      "learning_rate": 4.053668236957134e-05,
      "loss": 0.9136,
      "step": 1892
    },
    {
      "epoch": 2.78,
      "learning_rate": 4.044191024522686e-05,
      "loss": 0.955,
      "step": 1893
    },
    {
      "epoch": 2.78,
      "learning_rate": 4.034722094789809e-05,
      "loss": 0.9592,
      "step": 1894
    },
    {
      "epoch": 2.78,
      "learning_rate": 4.0252614609268766e-05,
      "loss": 0.9513,
      "step": 1895
    },
    {
      "epoch": 2.79,
      "learning_rate": 4.015809136090732e-05,
      "loss": 0.8921,
      "step": 1896
    },
    {
      "epoch": 2.79,
      "learning_rate": 4.0063651334266496e-05,
      "loss": 0.9401,
      "step": 1897
    },
    {
      "epoch": 2.79,
      "learning_rate": 3.996929466068344e-05,
      "loss": 0.9658,
      "step": 1898
    },
    {
      "epoch": 2.79,
      "learning_rate": 3.987502147137928e-05,
      "loss": 1.0133,
      "step": 1899
    },
    {
      "epoch": 2.79,
      "learning_rate": 3.978083189745907e-05,
      "loss": 0.8799,
      "step": 1900
    },
    {
      "epoch": 2.79,
      "learning_rate": 3.96867260699116e-05,
      "loss": 0.9336,
      "step": 1901
    },
    {
      "epoch": 2.79,
      "learning_rate": 3.9592704119609125e-05,
      "loss": 0.9089,
      "step": 1902
    },
    {
      "epoch": 2.8,
      "learning_rate": 3.9498766177307403e-05,
      "loss": 0.9544,
      "step": 1903
    },
    {
      "epoch": 2.8,
      "learning_rate": 3.9404912373645185e-05,
      "loss": 0.877,
      "step": 1904
    },
    {
      "epoch": 2.8,
      "learning_rate": 3.9311142839144365e-05,
      "loss": 0.9766,
      "step": 1905
    },
    {
      "epoch": 2.8,
      "learning_rate": 3.9217457704209536e-05,
      "loss": 0.9153,
      "step": 1906
    },
    {
      "epoch": 2.8,
      "learning_rate": 3.9123857099127936e-05,
      "loss": 0.8915,
      "step": 1907
    },
    {
      "epoch": 2.8,
      "learning_rate": 3.903034115406931e-05,
      "loss": 0.8892,
      "step": 1908
    },
    {
      "epoch": 2.8,
      "learning_rate": 3.893690999908562e-05,
      "loss": 0.9365,
      "step": 1909
    },
    {
      "epoch": 2.81,
      "learning_rate": 3.884356376411089e-05,
      "loss": 0.9644,
      "step": 1910
    },
    {
      "epoch": 2.81,
      "learning_rate": 3.875030257896105e-05,
      "loss": 0.8724,
      "step": 1911
    },
    {
      "epoch": 2.81,
      "learning_rate": 3.8657126573333804e-05,
      "loss": 0.9454,
      "step": 1912
    },
    {
      "epoch": 2.81,
      "learning_rate": 3.85640358768084e-05,
      "loss": 0.9356,
      "step": 1913
    },
    {
      "epoch": 2.81,
      "learning_rate": 3.8471030618845375e-05,
      "loss": 0.9073,
      "step": 1914
    },
    {
      "epoch": 2.81,
      "learning_rate": 3.837811092878649e-05,
      "loss": 0.9627,
      "step": 1915
    },
    {
      "epoch": 2.82,
      "learning_rate": 3.828527693585451e-05,
      "loss": 0.9601,
      "step": 1916
    },
    {
      "epoch": 2.82,
      "learning_rate": 3.819252876915297e-05,
      "loss": 0.9001,
      "step": 1917
    },
    {
      "epoch": 2.82,
      "learning_rate": 3.809986655766616e-05,
      "loss": 0.9256,
      "step": 1918
    },
    {
      "epoch": 2.82,
      "learning_rate": 3.800729043025871e-05,
      "loss": 0.9052,
      "step": 1919
    },
    {
      "epoch": 2.82,
      "learning_rate": 3.791480051567564e-05,
      "loss": 1.0201,
      "step": 1920
    },
    {
      "epoch": 2.82,
      "learning_rate": 3.7822396942542005e-05,
      "loss": 0.9417,
      "step": 1921
    },
    {
      "epoch": 2.82,
      "learning_rate": 3.7730079839362755e-05,
      "loss": 0.8914,
      "step": 1922
    },
    {
      "epoch": 2.83,
      "learning_rate": 3.76378493345227e-05,
      "loss": 0.9132,
      "step": 1923
    },
    {
      "epoch": 2.83,
      "learning_rate": 3.7545705556286126e-05,
      "loss": 0.8873,
      "step": 1924
    },
    {
      "epoch": 2.83,
      "learning_rate": 3.745364863279675e-05,
      "loss": 0.9416,
      "step": 1925
    },
    {
      "epoch": 2.83,
      "learning_rate": 3.7361678692077416e-05,
      "loss": 0.9625,
      "step": 1926
    },
    {
      "epoch": 2.83,
      "learning_rate": 3.726979586203019e-05,
      "loss": 0.9371,
      "step": 1927
    },
    {
      "epoch": 2.83,
      "learning_rate": 3.717800027043576e-05,
      "loss": 0.9022,
      "step": 1928
    },
    {
      "epoch": 2.83,
      "learning_rate": 3.708629204495371e-05,
      "loss": 0.9038,
      "step": 1929
    },
    {
      "epoch": 2.84,
      "learning_rate": 3.699467131312197e-05,
      "loss": 0.9053,
      "step": 1930
    },
    {
      "epoch": 2.84,
      "learning_rate": 3.6903138202356855e-05,
      "loss": 1.0199,
      "step": 1931
    },
    {
      "epoch": 2.84,
      "learning_rate": 3.681169283995279e-05,
      "loss": 0.9524,
      "step": 1932
    },
    {
      "epoch": 2.84,
      "learning_rate": 3.6720335353082246e-05,
      "loss": 0.8613,
      "step": 1933
    },
    {
      "epoch": 2.84,
      "learning_rate": 3.662906586879542e-05,
      "loss": 0.9004,
      "step": 1934
    },
    {
      "epoch": 2.84,
      "learning_rate": 3.653788451402009e-05,
      "loss": 0.9132,
      "step": 1935
    },
    {
      "epoch": 2.85,
      "learning_rate": 3.6446791415561574e-05,
      "loss": 0.866,
      "step": 1936
    },
    {
      "epoch": 2.85,
      "learning_rate": 3.635578670010242e-05,
      "loss": 0.975,
      "step": 1937
    },
    {
      "epoch": 2.85,
      "learning_rate": 3.626487049420223e-05,
      "loss": 0.9129,
      "step": 1938
    },
    {
      "epoch": 2.85,
      "learning_rate": 3.61740429242975e-05,
      "loss": 0.9127,
      "step": 1939
    },
    {
      "epoch": 2.85,
      "learning_rate": 3.6083304116701535e-05,
      "loss": 0.9235,
      "step": 1940
    },
    {
      "epoch": 2.85,
      "learning_rate": 3.599265419760408e-05,
      "loss": 0.9644,
      "step": 1941
    },
    {
      "epoch": 2.85,
      "learning_rate": 3.5902093293071425e-05,
      "loss": 0.9893,
      "step": 1942
    },
    {
      "epoch": 2.86,
      "learning_rate": 3.581162152904592e-05,
      "loss": 0.967,
      "step": 1943
    },
    {
      "epoch": 2.86,
      "learning_rate": 3.5721239031346066e-05,
      "loss": 0.8631,
      "step": 1944
    },
    {
      "epoch": 2.86,
      "learning_rate": 3.5630945925666134e-05,
      "loss": 0.9066,
      "step": 1945
    },
    {
      "epoch": 2.86,
      "learning_rate": 3.554074233757608e-05,
      "loss": 1.0079,
      "step": 1946
    },
    {
      "epoch": 2.86,
      "learning_rate": 3.545062839252147e-05,
      "loss": 0.8998,
      "step": 1947
    },
    {
      "epoch": 2.86,
      "learning_rate": 3.536060421582309e-05,
      "loss": 0.9351,
      "step": 1948
    },
    {
      "epoch": 2.86,
      "learning_rate": 3.5270669932676926e-05,
      "loss": 1.0081,
      "step": 1949
    },
    {
      "epoch": 2.87,
      "learning_rate": 3.518082566815396e-05,
      "loss": 0.9089,
      "step": 1950
    },
    {
      "epoch": 2.87,
      "learning_rate": 3.509107154719994e-05,
      "loss": 0.9823,
      "step": 1951
    },
    {
      "epoch": 2.87,
      "learning_rate": 3.500140769463533e-05,
      "loss": 0.9625,
      "step": 1952
    },
    {
      "epoch": 2.87,
      "learning_rate": 3.491183423515503e-05,
      "loss": 0.8762,
      "step": 1953
    },
    {
      "epoch": 2.87,
      "learning_rate": 3.48223512933282e-05,
      "loss": 0.9355,
      "step": 1954
    },
    {
      "epoch": 2.87,
      "learning_rate": 3.4732958993598154e-05,
      "loss": 0.9384,
      "step": 1955
    },
    {
      "epoch": 2.87,
      "learning_rate": 3.464365746028208e-05,
      "loss": 0.9222,
      "step": 1956
    },
    {
      "epoch": 2.88,
      "learning_rate": 3.455444681757105e-05,
      "loss": 0.9472,
      "step": 1957
    },
    {
      "epoch": 2.88,
      "learning_rate": 3.4465327189529664e-05,
      "loss": 0.9375,
      "step": 1958
    },
    {
      "epoch": 2.88,
      "learning_rate": 3.437629870009591e-05,
      "loss": 0.8795,
      "step": 1959
    },
    {
      "epoch": 2.88,
      "learning_rate": 3.428736147308115e-05,
      "loss": 0.8276,
      "step": 1960
    },
    {
      "epoch": 2.88,
      "learning_rate": 3.41985156321697e-05,
      "loss": 1.0247,
      "step": 1961
    },
    {
      "epoch": 2.88,
      "learning_rate": 3.410976130091892e-05,
      "loss": 0.9808,
      "step": 1962
    },
    {
      "epoch": 2.89,
      "learning_rate": 3.402109860275877e-05,
      "loss": 0.9254,
      "step": 1963
    },
    {
      "epoch": 2.89,
      "learning_rate": 3.393252766099187e-05,
      "loss": 0.8803,
      "step": 1964
    },
    {
      "epoch": 2.89,
      "learning_rate": 3.38440485987932e-05,
      "loss": 0.9022,
      "step": 1965
    },
    {
      "epoch": 2.89,
      "learning_rate": 3.375566153920992e-05,
      "loss": 0.8599,
      "step": 1966
    },
    {
      "epoch": 2.89,
      "learning_rate": 3.366736660516132e-05,
      "loss": 0.9316,
      "step": 1967
    },
    {
      "epoch": 2.89,
      "learning_rate": 3.3579163919438595e-05,
      "loss": 0.993,
      "step": 1968
    },
    {
      "epoch": 2.89,
      "learning_rate": 3.349105360470456e-05,
      "loss": 0.984,
      "step": 1969
    },
    {
      "epoch": 2.9,
      "learning_rate": 3.3403035783493605e-05,
      "loss": 0.9208,
      "step": 1970
    },
    {
      "epoch": 2.9,
      "learning_rate": 3.331511057821146e-05,
      "loss": 0.9139,
      "step": 1971
    },
    {
      "epoch": 2.9,
      "learning_rate": 3.322727811113516e-05,
      "loss": 0.9367,
      "step": 1972
    },
    {
      "epoch": 2.9,
      "learning_rate": 3.313953850441266e-05,
      "loss": 0.9278,
      "step": 1973
    },
    {
      "epoch": 2.9,
      "learning_rate": 3.305189188006281e-05,
      "loss": 0.9714,
      "step": 1974
    },
    {
      "epoch": 2.9,
      "learning_rate": 3.2964338359975134e-05,
      "loss": 0.8818,
      "step": 1975
    },
    {
      "epoch": 2.9,
      "learning_rate": 3.287687806590971e-05,
      "loss": 1.0362,
      "step": 1976
    },
    {
      "epoch": 2.91,
      "learning_rate": 3.2789511119496994e-05,
      "loss": 0.9065,
      "step": 1977
    },
    {
      "epoch": 2.91,
      "learning_rate": 3.270223764223755e-05,
      "loss": 0.9571,
      "step": 1978
    },
    {
      "epoch": 2.91,
      "learning_rate": 3.2615057755502e-05,
      "loss": 0.9128,
      "step": 1979
    },
    {
      "epoch": 2.91,
      "learning_rate": 3.252797158053077e-05,
      "loss": 0.8249,
      "step": 1980
    },
    {
      "epoch": 2.91,
      "learning_rate": 3.244097923843398e-05,
      "loss": 1.012,
      "step": 1981
    },
    {
      "epoch": 2.91,
      "learning_rate": 3.2354080850191324e-05,
      "loss": 0.854,
      "step": 1982
    },
    {
      "epoch": 2.92,
      "learning_rate": 3.226727653665171e-05,
      "loss": 0.8708,
      "step": 1983
    },
    {
      "epoch": 2.92,
      "learning_rate": 3.218056641853337e-05,
      "loss": 0.8986,
      "step": 1984
    },
    {
      "epoch": 2.92,
      "learning_rate": 3.2093950616423394e-05,
      "loss": 0.9326,
      "step": 1985
    },
    {
      "epoch": 2.92,
      "learning_rate": 3.200742925077775e-05,
      "loss": 0.941,
      "step": 1986
    },
    {
      "epoch": 2.92,
      "learning_rate": 3.192100244192116e-05,
      "loss": 1.0446,
      "step": 1987
    },
    {
      "epoch": 2.92,
      "learning_rate": 3.1834670310046734e-05,
      "loss": 0.971,
      "step": 1988
    },
    {
      "epoch": 2.92,
      "learning_rate": 3.174843297521596e-05,
      "loss": 0.9548,
      "step": 1989
    },
    {
      "epoch": 2.93,
      "learning_rate": 3.166229055735848e-05,
      "loss": 0.8474,
      "step": 1990
    },
    {
      "epoch": 2.93,
      "learning_rate": 3.157624317627195e-05,
      "loss": 0.8806,
      "step": 1991
    },
    {
      "epoch": 2.93,
      "learning_rate": 3.1490290951621904e-05,
      "loss": 0.9161,
      "step": 1992
    },
    {
      "epoch": 2.93,
      "learning_rate": 3.140443400294146e-05,
      "loss": 0.9025,
      "step": 1993
    },
    {
      "epoch": 2.93,
      "learning_rate": 3.1318672449631284e-05,
      "loss": 0.9566,
      "step": 1994
    },
    {
      "epoch": 2.93,
      "learning_rate": 3.123300641095935e-05,
      "loss": 0.8958,
      "step": 1995
    },
    {
      "epoch": 2.93,
      "learning_rate": 3.114743600606078e-05,
      "loss": 0.9001,
      "step": 1996
    },
    {
      "epoch": 2.94,
      "learning_rate": 3.106196135393782e-05,
      "loss": 0.993,
      "step": 1997
    },
    {
      "epoch": 2.94,
      "learning_rate": 3.09765825734594e-05,
      "loss": 0.932,
      "step": 1998
    },
    {
      "epoch": 2.94,
      "learning_rate": 3.089129978336118e-05,
      "loss": 0.9425,
      "step": 1999
    },
    {
      "epoch": 2.94,
      "learning_rate": 3.080611310224539e-05,
      "loss": 0.8835,
      "step": 2000
    },
    {
      "epoch": 2.94,
      "learning_rate": 3.0721022648580486e-05,
      "loss": 0.9171,
      "step": 2001
    },
    {
      "epoch": 2.94,
      "learning_rate": 3.063602854070123e-05,
      "loss": 0.8977,
      "step": 2002
    },
    {
      "epoch": 2.94,
      "learning_rate": 3.055113089680829e-05,
      "loss": 1.0178,
      "step": 2003
    },
    {
      "epoch": 2.95,
      "learning_rate": 3.0466329834968233e-05,
      "loss": 0.906,
      "step": 2004
    },
    {
      "epoch": 2.95,
      "learning_rate": 3.0381625473113284e-05,
      "loss": 0.9512,
      "step": 2005
    },
    {
      "epoch": 2.95,
      "learning_rate": 3.029701792904117e-05,
      "loss": 0.9653,
      "step": 2006
    },
    {
      "epoch": 2.95,
      "learning_rate": 3.0212507320415052e-05,
      "loss": 0.9445,
      "step": 2007
    },
    {
      "epoch": 2.95,
      "learning_rate": 3.0128093764763254e-05,
      "loss": 0.9406,
      "step": 2008
    },
    {
      "epoch": 2.95,
      "learning_rate": 3.0043777379479098e-05,
      "loss": 0.9888,
      "step": 2009
    },
    {
      "epoch": 2.96,
      "learning_rate": 2.9959558281820766e-05,
      "loss": 0.8496,
      "step": 2010
    },
    {
      "epoch": 2.96,
      "learning_rate": 2.9875436588911153e-05,
      "loss": 0.9583,
      "step": 2011
    },
    {
      "epoch": 2.96,
      "learning_rate": 2.979141241773775e-05,
      "loss": 0.9488,
      "step": 2012
    },
    {
      "epoch": 2.96,
      "learning_rate": 2.9707485885152363e-05,
      "loss": 0.9339,
      "step": 2013
    },
    {
      "epoch": 2.96,
      "learning_rate": 2.9623657107870996e-05,
      "loss": 0.998,
      "step": 2014
    },
    {
      "epoch": 2.96,
      "learning_rate": 2.953992620247379e-05,
      "loss": 0.9076,
      "step": 2015
    },
    {
      "epoch": 2.96,
      "learning_rate": 2.945629328540468e-05,
      "loss": 0.9596,
      "step": 2016
    },
    {
      "epoch": 2.96,
      "eval_loss": 0.9915859699249268,
      "eval_runtime": 2.6122,
      "eval_samples_per_second": 418.031,
      "eval_steps_per_second": 26.414,
      "step": 2016
    },
    {
      "epoch": 2.97,
      "learning_rate": 2.9372758472971417e-05,
      "loss": 0.9104,
      "step": 2017
    },
    {
      "epoch": 2.97,
      "learning_rate": 2.9289321881345254e-05,
      "loss": 1.0079,
      "step": 2018
    },
    {
      "epoch": 2.97,
      "learning_rate": 2.9205983626560874e-05,
      "loss": 0.9927,
      "step": 2019
    },
    {
      "epoch": 2.97,
      "learning_rate": 2.9122743824516195e-05,
      "loss": 0.9863,
      "step": 2020
    },
    {
      "epoch": 2.97,
      "learning_rate": 2.9039602590972205e-05,
      "loss": 0.8651,
      "step": 2021
    },
    {
      "epoch": 2.97,
      "learning_rate": 2.8956560041552882e-05,
      "loss": 0.9449,
      "step": 2022
    },
    {
      "epoch": 2.97,
      "learning_rate": 2.8873616291744843e-05,
      "loss": 0.9486,
      "step": 2023
    },
    {
      "epoch": 2.98,
      "learning_rate": 2.879077145689746e-05,
      "loss": 0.9713,
      "step": 2024
    },
    {
      "epoch": 2.98,
      "learning_rate": 2.8708025652222415e-05,
      "loss": 0.8496,
      "step": 2025
    },
    {
      "epoch": 2.98,
      "learning_rate": 2.8625378992793696e-05,
      "loss": 0.9048,
      "step": 2026
    },
    {
      "epoch": 2.98,
      "learning_rate": 2.854283159354748e-05,
      "loss": 0.9396,
      "step": 2027
    },
    {
      "epoch": 2.98,
      "learning_rate": 2.8460383569281824e-05,
      "loss": 0.9327,
      "step": 2028
    },
    {
      "epoch": 2.98,
      "learning_rate": 2.8378035034656625e-05,
      "loss": 1.0179,
      "step": 2029
    },
    {
      "epoch": 2.99,
      "learning_rate": 2.8295786104193366e-05,
      "loss": 1.0093,
      "step": 2030
    },
    {
      "epoch": 2.99,
      "learning_rate": 2.8213636892275087e-05,
      "loss": 1.0036,
      "step": 2031
    },
    {
      "epoch": 2.99,
      "learning_rate": 2.8131587513146164e-05,
      "loss": 0.9032,
      "step": 2032
    },
    {
      "epoch": 2.99,
      "learning_rate": 2.8049638080912044e-05,
      "loss": 0.9796,
      "step": 2033
    },
    {
      "epoch": 2.99,
      "learning_rate": 2.7967788709539233e-05,
      "loss": 0.9627,
      "step": 2034
    },
    {
      "epoch": 2.99,
      "learning_rate": 2.7886039512855065e-05,
      "loss": 0.8332,
      "step": 2035
    },
    {
      "epoch": 2.99,
      "learning_rate": 2.7804390604547557e-05,
      "loss": 0.9075,
      "step": 2036
    },
    {
      "epoch": 3.0,
      "learning_rate": 2.7722842098165325e-05,
      "loss": 0.8997,
      "step": 2037
    },
    {
      "epoch": 3.0,
      "learning_rate": 2.7641394107117226e-05,
      "loss": 1.0088,
      "step": 2038
    },
    {
      "epoch": 3.0,
      "learning_rate": 2.7560046744672495e-05,
      "loss": 0.9538,
      "step": 2039
    },
    {
      "epoch": 3.0,
      "learning_rate": 2.747880012396028e-05,
      "loss": 0.9061,
      "step": 2040
    },
    {
      "epoch": 3.0,
      "learning_rate": 2.739765435796967e-05,
      "loss": 0.9441,
      "step": 2041
    },
    {
      "epoch": 3.0,
      "learning_rate": 2.7316609559549565e-05,
      "loss": 0.9793,
      "step": 2042
    },
    {
      "epoch": 3.0,
      "learning_rate": 2.7235665841408366e-05,
      "loss": 0.9078,
      "step": 2043
    },
    {
      "epoch": 3.01,
      "learning_rate": 2.7154823316113932e-05,
      "loss": 0.9274,
      "step": 2044
    },
    {
      "epoch": 3.01,
      "learning_rate": 2.707408209609339e-05,
      "loss": 0.9687,
      "step": 2045
    },
    {
      "epoch": 3.01,
      "learning_rate": 2.6993442293632963e-05,
      "loss": 0.985,
      "step": 2046
    },
    {
      "epoch": 3.01,
      "learning_rate": 2.691290402087787e-05,
      "loss": 0.9312,
      "step": 2047
    },
    {
      "epoch": 3.01,
      "learning_rate": 2.6832467389832173e-05,
      "loss": 0.8891,
      "step": 2048
    },
    {
      "epoch": 3.01,
      "learning_rate": 2.6752132512358475e-05,
      "loss": 0.8605,
      "step": 2049
    },
    {
      "epoch": 3.01,
      "learning_rate": 2.667189950017793e-05,
      "loss": 0.9902,
      "step": 2050
    },
    {
      "epoch": 3.02,
      "learning_rate": 2.6591768464870016e-05,
      "loss": 0.9078,
      "step": 2051
    },
    {
      "epoch": 3.02,
      "learning_rate": 2.6511739517872426e-05,
      "loss": 0.956,
      "step": 2052
    },
    {
      "epoch": 3.02,
      "learning_rate": 2.6431812770480836e-05,
      "loss": 0.8834,
      "step": 2053
    },
    {
      "epoch": 3.02,
      "learning_rate": 2.6351988333848788e-05,
      "loss": 0.8702,
      "step": 2054
    },
    {
      "epoch": 3.0,
      "learning_rate": 2.6272266318987603e-05,
      "loss": 0.9522,
      "step": 2055
    },
    {
      "epoch": 3.0,
      "learning_rate": 2.6192646836766088e-05,
      "loss": 0.9338,
      "step": 2056
    },
    {
      "epoch": 3.0,
      "learning_rate": 2.6113129997910547e-05,
      "loss": 0.8851,
      "step": 2057
    },
    {
      "epoch": 3.01,
      "learning_rate": 2.6033715913004463e-05,
      "loss": 0.7981,
      "step": 2058
    },
    {
      "epoch": 3.01,
      "learning_rate": 2.5954404692488433e-05,
      "loss": 0.799,
      "step": 2059
    },
    {
      "epoch": 3.01,
      "learning_rate": 2.587519644666001e-05,
      "loss": 0.8267,
      "step": 2060
    },
    {
      "epoch": 3.01,
      "learning_rate": 2.579609128567353e-05,
      "loss": 0.846,
      "step": 2061
    },
    {
      "epoch": 3.01,
      "learning_rate": 2.571708931954e-05,
      "loss": 0.8217,
      "step": 2062
    },
    {
      "epoch": 3.01,
      "learning_rate": 2.5638190658126938e-05,
      "loss": 0.905,
      "step": 2063
    },
    {
      "epoch": 3.01,
      "learning_rate": 2.5559395411158115e-05,
      "loss": 0.8105,
      "step": 2064
    },
    {
      "epoch": 3.02,
      "learning_rate": 2.5480703688213524e-05,
      "loss": 0.8824,
      "step": 2065
    },
    {
      "epoch": 3.02,
      "learning_rate": 2.5402115598729182e-05,
      "loss": 0.9279,
      "step": 2066
    },
    {
      "epoch": 3.02,
      "learning_rate": 2.5323631251997026e-05,
      "loss": 0.9431,
      "step": 2067
    },
    {
      "epoch": 3.02,
      "learning_rate": 2.5245250757164663e-05,
      "loss": 0.8822,
      "step": 2068
    },
    {
      "epoch": 3.02,
      "learning_rate": 2.5166974223235296e-05,
      "loss": 0.8182,
      "step": 2069
    },
    {
      "epoch": 3.02,
      "learning_rate": 2.5088801759067528e-05,
      "loss": 0.9177,
      "step": 2070
    },
    {
      "epoch": 3.03,
      "learning_rate": 2.501073347337526e-05,
      "loss": 0.8085,
      "step": 2071
    },
    {
      "epoch": 3.03,
      "learning_rate": 2.493276947472756e-05,
      "loss": 0.9009,
      "step": 2072
    },
    {
      "epoch": 3.03,
      "learning_rate": 2.485490987154837e-05,
      "loss": 0.9113,
      "step": 2073
    },
    {
      "epoch": 3.03,
      "learning_rate": 2.4777154772116496e-05,
      "loss": 0.8268,
      "step": 2074
    },
    {
      "epoch": 3.03,
      "learning_rate": 2.4699504284565412e-05,
      "loss": 0.9038,
      "step": 2075
    },
    {
      "epoch": 3.03,
      "learning_rate": 2.462195851688306e-05,
      "loss": 0.9268,
      "step": 2076
    },
    {
      "epoch": 3.03,
      "learning_rate": 2.4544517576911862e-05,
      "loss": 0.7817,
      "step": 2077
    },
    {
      "epoch": 3.04,
      "learning_rate": 2.4467181572348318e-05,
      "loss": 0.9028,
      "step": 2078
    },
    {
      "epoch": 3.04,
      "learning_rate": 2.438995061074314e-05,
      "loss": 0.8584,
      "step": 2079
    },
    {
      "epoch": 3.04,
      "learning_rate": 2.4312824799500844e-05,
      "loss": 0.9867,
      "step": 2080
    },
    {
      "epoch": 3.04,
      "learning_rate": 2.4235804245879723e-05,
      "loss": 0.9181,
      "step": 2081
    },
    {
      "epoch": 3.04,
      "learning_rate": 2.4158889056991775e-05,
      "loss": 0.8883,
      "step": 2082
    },
    {
      "epoch": 3.04,
      "learning_rate": 2.4082079339802378e-05,
      "loss": 0.9067,
      "step": 2083
    },
    {
      "epoch": 3.04,
      "learning_rate": 2.4005375201130274e-05,
      "loss": 0.8907,
      "step": 2084
    },
    {
      "epoch": 3.05,
      "learning_rate": 2.392877674764734e-05,
      "loss": 0.9214,
      "step": 2085
    },
    {
      "epoch": 3.05,
      "learning_rate": 2.3852284085878517e-05,
      "loss": 0.8258,
      "step": 2086
    },
    {
      "epoch": 3.05,
      "learning_rate": 2.377589732220167e-05,
      "loss": 0.8416,
      "step": 2087
    },
    {
      "epoch": 3.05,
      "learning_rate": 2.369961656284727e-05,
      "loss": 0.8781,
      "step": 2088
    },
    {
      "epoch": 3.05,
      "learning_rate": 2.362344191389846e-05,
      "loss": 0.8864,
      "step": 2089
    },
    {
      "epoch": 3.05,
      "learning_rate": 2.354737348129077e-05,
      "loss": 0.9058,
      "step": 2090
    },
    {
      "epoch": 3.06,
      "learning_rate": 2.3471411370812014e-05,
      "loss": 0.87,
      "step": 2091
    },
    {
      "epoch": 3.06,
      "learning_rate": 2.339555568810221e-05,
      "loss": 0.966,
      "step": 2092
    },
    {
      "epoch": 3.06,
      "learning_rate": 2.33198065386533e-05,
      "loss": 0.9634,
      "step": 2093
    },
    {
      "epoch": 3.06,
      "learning_rate": 2.324416402780907e-05,
      "loss": 0.9022,
      "step": 2094
    },
    {
      "epoch": 3.06,
      "learning_rate": 2.316862826076507e-05,
      "loss": 0.875,
      "step": 2095
    },
    {
      "epoch": 3.06,
      "learning_rate": 2.3093199342568318e-05,
      "loss": 0.9033,
      "step": 2096
    },
    {
      "epoch": 3.06,
      "learning_rate": 2.3017877378117326e-05,
      "loss": 0.9128,
      "step": 2097
    },
    {
      "epoch": 3.07,
      "learning_rate": 2.294266247216178e-05,
      "loss": 0.8851,
      "step": 2098
    },
    {
      "epoch": 3.07,
      "learning_rate": 2.2867554729302542e-05,
      "loss": 0.8791,
      "step": 2099
    },
    {
      "epoch": 3.07,
      "learning_rate": 2.2792554253991415e-05,
      "loss": 0.978,
      "step": 2100
    },
    {
      "epoch": 3.07,
      "learning_rate": 2.271766115053099e-05,
      "loss": 0.8423,
      "step": 2101
    },
    {
      "epoch": 3.07,
      "learning_rate": 2.2642875523074613e-05,
      "loss": 0.9984,
      "step": 2102
    },
    {
      "epoch": 3.07,
      "learning_rate": 2.2568197475626175e-05,
      "loss": 0.8152,
      "step": 2103
    },
    {
      "epoch": 3.07,
      "learning_rate": 2.249362711203985e-05,
      "loss": 0.9389,
      "step": 2104
    },
    {
      "epoch": 3.08,
      "learning_rate": 2.241916453602011e-05,
      "loss": 0.9301,
      "step": 2105
    },
    {
      "epoch": 3.08,
      "learning_rate": 2.2344809851121583e-05,
      "loss": 0.8169,
      "step": 2106
    },
    {
      "epoch": 3.08,
      "learning_rate": 2.2270563160748793e-05,
      "loss": 0.9564,
      "step": 2107
    },
    {
      "epoch": 3.08,
      "learning_rate": 2.2196424568156073e-05,
      "loss": 0.9204,
      "step": 2108
    },
    {
      "epoch": 3.08,
      "learning_rate": 2.2122394176447416e-05,
      "loss": 0.9038,
      "step": 2109
    },
    {
      "epoch": 3.08,
      "learning_rate": 2.204847208857642e-05,
      "loss": 0.8507,
      "step": 2110
    },
    {
      "epoch": 3.08,
      "learning_rate": 2.197465840734596e-05,
      "loss": 0.8936,
      "step": 2111
    },
    {
      "epoch": 3.09,
      "learning_rate": 2.190095323540825e-05,
      "loss": 0.9433,
      "step": 2112
    },
    {
      "epoch": 3.09,
      "learning_rate": 2.1827356675264542e-05,
      "loss": 0.899,
      "step": 2113
    },
    {
      "epoch": 3.09,
      "learning_rate": 2.1753868829265046e-05,
      "loss": 0.7804,
      "step": 2114
    },
    {
      "epoch": 3.09,
      "learning_rate": 2.1680489799608762e-05,
      "loss": 0.8703,
      "step": 2115
    },
    {
      "epoch": 3.09,
      "learning_rate": 2.160721968834344e-05,
      "loss": 0.8523,
      "step": 2116
    },
    {
      "epoch": 3.09,
      "learning_rate": 2.153405859736528e-05,
      "loss": 0.9138,
      "step": 2117
    },
    {
      "epoch": 3.1,
      "learning_rate": 2.146100662841889e-05,
      "loss": 0.8954,
      "step": 2118
    },
    {
      "epoch": 3.1,
      "learning_rate": 2.1388063883097152e-05,
      "loss": 0.8585,
      "step": 2119
    },
    {
      "epoch": 3.1,
      "learning_rate": 2.1315230462840985e-05,
      "loss": 0.9287,
      "step": 2120
    },
    {
      "epoch": 3.1,
      "learning_rate": 2.1242506468939383e-05,
      "loss": 0.8709,
      "step": 2121
    },
    {
      "epoch": 3.1,
      "learning_rate": 2.1169892002529047e-05,
      "loss": 0.9009,
      "step": 2122
    },
    {
      "epoch": 3.1,
      "learning_rate": 2.1097387164594406e-05,
      "loss": 0.9042,
      "step": 2123
    },
    {
      "epoch": 3.1,
      "learning_rate": 2.102499205596743e-05,
      "loss": 0.8884,
      "step": 2124
    },
    {
      "epoch": 3.11,
      "learning_rate": 2.095270677732746e-05,
      "loss": 0.965,
      "step": 2125
    },
    {
      "epoch": 3.11,
      "learning_rate": 2.0880531429201145e-05,
      "loss": 0.8886,
      "step": 2126
    },
    {
      "epoch": 3.11,
      "learning_rate": 2.0808466111962264e-05,
      "loss": 0.8881,
      "step": 2127
    },
    {
      "epoch": 3.11,
      "learning_rate": 2.0736510925831532e-05,
      "loss": 0.8371,
      "step": 2128
    },
    {
      "epoch": 3.11,
      "learning_rate": 2.0664665970876496e-05,
      "loss": 0.9097,
      "step": 2129
    },
    {
      "epoch": 3.11,
      "learning_rate": 2.0592931347011423e-05,
      "loss": 0.9366,
      "step": 2130
    },
    {
      "epoch": 3.11,
      "learning_rate": 2.0521307153997182e-05,
      "loss": 0.8485,
      "step": 2131
    },
    {
      "epoch": 3.12,
      "learning_rate": 2.0449793491441028e-05,
      "loss": 0.8103,
      "step": 2132
    },
    {
      "epoch": 3.12,
      "learning_rate": 2.037839045879646e-05,
      "loss": 0.9271,
      "step": 2133
    },
    {
      "epoch": 3.12,
      "learning_rate": 2.0307098155363236e-05,
      "loss": 0.8905,
      "step": 2134
    },
    {
      "epoch": 3.12,
      "learning_rate": 2.0235916680287015e-05,
      "loss": 0.8801,
      "step": 2135
    },
    {
      "epoch": 3.12,
      "learning_rate": 2.0164846132559402e-05,
      "loss": 0.8613,
      "step": 2136
    },
    {
      "epoch": 3.12,
      "learning_rate": 2.009388661101771e-05,
      "loss": 0.8611,
      "step": 2137
    },
    {
      "epoch": 3.13,
      "learning_rate": 2.0023038214344823e-05,
      "loss": 0.9265,
      "step": 2138
    },
    {
      "epoch": 3.13,
      "learning_rate": 1.9952301041069122e-05,
      "loss": 0.8643,
      "step": 2139
    },
    {
      "epoch": 3.13,
      "learning_rate": 1.9881675189564254e-05,
      "loss": 0.8263,
      "step": 2140
    },
    {
      "epoch": 3.13,
      "learning_rate": 1.9811160758049163e-05,
      "loss": 0.8731,
      "step": 2141
    },
    {
      "epoch": 3.13,
      "learning_rate": 1.974075784458771e-05,
      "loss": 0.8522,
      "step": 2142
    },
    {
      "epoch": 3.13,
      "learning_rate": 1.9670466547088773e-05,
      "loss": 0.848,
      "step": 2143
    },
    {
      "epoch": 3.13,
      "learning_rate": 1.9600286963305957e-05,
      "loss": 0.9803,
      "step": 2144
    },
    {
      "epoch": 3.14,
      "learning_rate": 1.9530219190837485e-05,
      "loss": 0.8974,
      "step": 2145
    },
    {
      "epoch": 3.14,
      "learning_rate": 1.946026332712615e-05,
      "loss": 0.8892,
      "step": 2146
    },
    {
      "epoch": 3.14,
      "learning_rate": 1.9390419469459065e-05,
      "loss": 0.8785,
      "step": 2147
    },
    {
      "epoch": 3.14,
      "learning_rate": 1.93206877149676e-05,
      "loss": 0.9063,
      "step": 2148
    },
    {
      "epoch": 3.14,
      "learning_rate": 1.9251068160627173e-05,
      "loss": 0.8974,
      "step": 2149
    },
    {
      "epoch": 3.14,
      "learning_rate": 1.9181560903257232e-05,
      "loss": 0.8693,
      "step": 2150
    },
    {
      "epoch": 3.14,
      "learning_rate": 1.9112166039521073e-05,
      "loss": 0.8972,
      "step": 2151
    },
    {
      "epoch": 3.15,
      "learning_rate": 1.9042883665925604e-05,
      "loss": 0.9402,
      "step": 2152
    },
    {
      "epoch": 3.15,
      "learning_rate": 1.897371387882134e-05,
      "loss": 0.8868,
      "step": 2153
    },
    {
      "epoch": 3.15,
      "learning_rate": 1.8904656774402208e-05,
      "loss": 0.9122,
      "step": 2154
    },
    {
      "epoch": 3.15,
      "learning_rate": 1.8835712448705424e-05,
      "loss": 0.8585,
      "step": 2155
    },
    {
      "epoch": 3.15,
      "learning_rate": 1.8766880997611424e-05,
      "loss": 0.8969,
      "step": 2156
    },
    {
      "epoch": 3.15,
      "learning_rate": 1.8698162516843552e-05,
      "loss": 0.8787,
      "step": 2157
    },
    {
      "epoch": 3.15,
      "learning_rate": 1.86295571019682e-05,
      "loss": 0.8505,
      "step": 2158
    },
    {
      "epoch": 3.16,
      "learning_rate": 1.8561064848394382e-05,
      "loss": 0.869,
      "step": 2159
    },
    {
      "epoch": 3.16,
      "learning_rate": 1.849268585137377e-05,
      "loss": 0.8601,
      "step": 2160
    },
    {
      "epoch": 3.16,
      "learning_rate": 1.8424420206000616e-05,
      "loss": 0.8734,
      "step": 2161
    },
    {
      "epoch": 3.16,
      "learning_rate": 1.835626800721144e-05,
      "loss": 0.9073,
      "step": 2162
    },
    {
      "epoch": 3.16,
      "learning_rate": 1.8288229349785025e-05,
      "loss": 0.8833,
      "step": 2163
    },
    {
      "epoch": 3.16,
      "learning_rate": 1.8220304328342252e-05,
      "loss": 0.8833,
      "step": 2164
    },
    {
      "epoch": 3.17,
      "learning_rate": 1.815249303734594e-05,
      "loss": 0.7653,
      "step": 2165
    },
    {
      "epoch": 3.17,
      "learning_rate": 1.808479557110081e-05,
      "loss": 0.9249,
      "step": 2166
    },
    {
      "epoch": 3.17,
      "learning_rate": 1.801721202375327e-05,
      "loss": 0.9382,
      "step": 2167
    },
    {
      "epoch": 3.17,
      "learning_rate": 1.7949742489291255e-05,
      "loss": 0.8937,
      "step": 2168
    },
    {
      "epoch": 3.17,
      "learning_rate": 1.7882387061544182e-05,
      "loss": 0.9076,
      "step": 2169
    },
    {
      "epoch": 3.17,
      "learning_rate": 1.7815145834182734e-05,
      "loss": 0.8711,
      "step": 2170
    },
    {
      "epoch": 3.17,
      "learning_rate": 1.7748018900718854e-05,
      "loss": 0.8599,
      "step": 2171
    },
    {
      "epoch": 3.18,
      "learning_rate": 1.7681006354505493e-05,
      "loss": 0.9874,
      "step": 2172
    },
    {
      "epoch": 3.18,
      "learning_rate": 1.7614108288736486e-05,
      "loss": 0.8984,
      "step": 2173
    },
    {
      "epoch": 3.18,
      "learning_rate": 1.754732479644655e-05,
      "loss": 0.8941,
      "step": 2174
    },
    {
      "epoch": 3.18,
      "learning_rate": 1.7480655970510985e-05,
      "loss": 0.9727,
      "step": 2175
    },
    {
      "epoch": 3.18,
      "learning_rate": 1.7414101903645684e-05,
      "loss": 0.8093,
      "step": 2176
    },
    {
      "epoch": 3.18,
      "learning_rate": 1.7347662688406908e-05,
      "loss": 0.8109,
      "step": 2177
    },
    {
      "epoch": 3.18,
      "learning_rate": 1.7281338417191205e-05,
      "loss": 0.9079,
      "step": 2178
    },
    {
      "epoch": 3.19,
      "learning_rate": 1.721512918223527e-05,
      "loss": 0.9269,
      "step": 2179
    },
    {
      "epoch": 3.19,
      "learning_rate": 1.7149035075615794e-05,
      "loss": 0.8874,
      "step": 2180
    },
    {
      "epoch": 3.19,
      "learning_rate": 1.7083056189249412e-05,
      "loss": 0.8941,
      "step": 2181
    },
    {
      "epoch": 3.19,
      "learning_rate": 1.7017192614892508e-05,
      "loss": 0.8195,
      "step": 2182
    },
    {
      "epoch": 3.19,
      "learning_rate": 1.6951444444141084e-05,
      "loss": 0.9034,
      "step": 2183
    },
    {
      "epoch": 3.19,
      "learning_rate": 1.688581176843066e-05,
      "loss": 0.8771,
      "step": 2184
    },
    {
      "epoch": 3.19,
      "eval_loss": 1.001231074333191,
      "eval_runtime": 2.6132,
      "eval_samples_per_second": 417.878,
      "eval_steps_per_second": 26.404,
      "step": 2184
    },
    {
      "epoch": 3.2,
      "learning_rate": 1.6820294679036087e-05,
      "loss": 0.9257,
      "step": 2185
    },
    {
      "epoch": 3.2,
      "learning_rate": 1.675489326707159e-05,
      "loss": 0.8702,
      "step": 2186
    },
    {
      "epoch": 3.2,
      "learning_rate": 1.668960762349042e-05,
      "loss": 0.8393,
      "step": 2187
    },
    {
      "epoch": 3.2,
      "learning_rate": 1.6624437839084862e-05,
      "loss": 0.891,
      "step": 2188
    },
    {
      "epoch": 3.2,
      "learning_rate": 1.6559384004486055e-05,
      "loss": 0.8837,
      "step": 2189
    },
    {
      "epoch": 3.2,
      "learning_rate": 1.6494446210163926e-05,
      "loss": 0.8943,
      "step": 2190
    },
    {
      "epoch": 3.2,
      "learning_rate": 1.6429624546427034e-05,
      "loss": 0.9046,
      "step": 2191
    },
    {
      "epoch": 3.21,
      "learning_rate": 1.6364919103422393e-05,
      "loss": 0.7765,
      "step": 2192
    },
    {
      "epoch": 3.21,
      "learning_rate": 1.6300329971135397e-05,
      "loss": 0.8723,
      "step": 2193
    },
    {
      "epoch": 3.21,
      "learning_rate": 1.6235857239389696e-05,
      "loss": 0.8941,
      "step": 2194
    },
    {
      "epoch": 3.21,
      "learning_rate": 1.6171500997847054e-05,
      "loss": 0.8349,
      "step": 2195
    },
    {
      "epoch": 3.21,
      "learning_rate": 1.6107261336007285e-05,
      "loss": 0.7731,
      "step": 2196
    },
    {
      "epoch": 3.21,
      "learning_rate": 1.6043138343207975e-05,
      "loss": 0.936,
      "step": 2197
    },
    {
      "epoch": 3.21,
      "learning_rate": 1.5979132108624574e-05,
      "loss": 0.8672,
      "step": 2198
    },
    {
      "epoch": 3.22,
      "learning_rate": 1.5915242721270074e-05,
      "loss": 0.8693,
      "step": 2199
    },
    {
      "epoch": 3.22,
      "learning_rate": 1.5851470269994984e-05,
      "loss": 0.8971,
      "step": 2200
    },
    {
      "epoch": 3.22,
      "learning_rate": 1.5787814843487226e-05,
      "loss": 0.9144,
      "step": 2201
    },
    {
      "epoch": 3.22,
      "learning_rate": 1.5724276530271965e-05,
      "loss": 0.9183,
      "step": 2202
    },
    {
      "epoch": 3.22,
      "learning_rate": 1.566085541871145e-05,
      "loss": 0.8262,
      "step": 2203
    },
    {
      "epoch": 3.22,
      "learning_rate": 1.5597551597004966e-05,
      "loss": 0.8459,
      "step": 2204
    },
    {
      "epoch": 3.23,
      "learning_rate": 1.5534365153188723e-05,
      "loss": 0.9236,
      "step": 2205
    },
    {
      "epoch": 3.23,
      "learning_rate": 1.5471296175135674e-05,
      "loss": 0.9021,
      "step": 2206
    },
    {
      "epoch": 3.23,
      "learning_rate": 1.5408344750555383e-05,
      "loss": 0.8881,
      "step": 2207
    },
    {
      "epoch": 3.23,
      "learning_rate": 1.534551096699396e-05,
      "loss": 0.8462,
      "step": 2208
    },
    {
      "epoch": 3.23,
      "learning_rate": 1.5282794911833887e-05,
      "loss": 0.9467,
      "step": 2209
    },
    {
      "epoch": 3.23,
      "learning_rate": 1.5220196672293929e-05,
      "loss": 0.8647,
      "step": 2210
    },
    {
      "epoch": 3.23,
      "learning_rate": 1.5157716335429061e-05,
      "loss": 0.9548,
      "step": 2211
    },
    {
      "epoch": 3.24,
      "learning_rate": 1.5095353988130235e-05,
      "loss": 0.858,
      "step": 2212
    },
    {
      "epoch": 3.24,
      "learning_rate": 1.5033109717124283e-05,
      "loss": 0.883,
      "step": 2213
    },
    {
      "epoch": 3.24,
      "learning_rate": 1.4970983608973942e-05,
      "loss": 0.9509,
      "step": 2214
    },
    {
      "epoch": 3.24,
      "learning_rate": 1.4908975750077491e-05,
      "loss": 0.8861,
      "step": 2215
    },
    {
      "epoch": 3.24,
      "learning_rate": 1.4847086226668872e-05,
      "loss": 0.8819,
      "step": 2216
    },
    {
      "epoch": 3.24,
      "learning_rate": 1.478531512481739e-05,
      "loss": 0.8183,
      "step": 2217
    },
    {
      "epoch": 3.24,
      "learning_rate": 1.4723662530427673e-05,
      "loss": 0.8612,
      "step": 2218
    },
    {
      "epoch": 3.25,
      "learning_rate": 1.4662128529239572e-05,
      "loss": 0.8367,
      "step": 2219
    },
    {
      "epoch": 3.25,
      "learning_rate": 1.4600713206827932e-05,
      "loss": 0.9016,
      "step": 2220
    },
    {
      "epoch": 3.25,
      "learning_rate": 1.4539416648602656e-05,
      "loss": 0.85,
      "step": 2221
    },
    {
      "epoch": 3.25,
      "learning_rate": 1.4478238939808453e-05,
      "loss": 0.8941,
      "step": 2222
    },
    {
      "epoch": 3.25,
      "learning_rate": 1.4417180165524714e-05,
      "loss": 0.8326,
      "step": 2223
    },
    {
      "epoch": 3.25,
      "learning_rate": 1.4356240410665433e-05,
      "loss": 0.8711,
      "step": 2224
    },
    {
      "epoch": 3.25,
      "learning_rate": 1.429541975997908e-05,
      "loss": 0.8943,
      "step": 2225
    },
    {
      "epoch": 3.26,
      "learning_rate": 1.4234718298048555e-05,
      "loss": 0.9355,
      "step": 2226
    },
    {
      "epoch": 3.26,
      "learning_rate": 1.4174136109290925e-05,
      "loss": 0.9106,
      "step": 2227
    },
    {
      "epoch": 3.26,
      "learning_rate": 1.4113673277957395e-05,
      "loss": 0.9101,
      "step": 2228
    },
    {
      "epoch": 3.26,
      "learning_rate": 1.4053329888133238e-05,
      "loss": 0.9257,
      "step": 2229
    },
    {
      "epoch": 3.26,
      "learning_rate": 1.3993106023737546e-05,
      "loss": 0.9317,
      "step": 2230
    },
    {
      "epoch": 3.26,
      "learning_rate": 1.3933001768523269e-05,
      "loss": 0.8653,
      "step": 2231
    },
    {
      "epoch": 3.27,
      "learning_rate": 1.3873017206076933e-05,
      "loss": 0.8959,
      "step": 2232
    },
    {
      "epoch": 3.27,
      "learning_rate": 1.3813152419818654e-05,
      "loss": 1.0166,
      "step": 2233
    },
    {
      "epoch": 3.27,
      "learning_rate": 1.3753407493001968e-05,
      "loss": 0.8364,
      "step": 2234
    },
    {
      "epoch": 3.27,
      "learning_rate": 1.3693782508713704e-05,
      "loss": 0.9402,
      "step": 2235
    },
    {
      "epoch": 3.27,
      "learning_rate": 1.3634277549873953e-05,
      "loss": 0.9119,
      "step": 2236
    },
    {
      "epoch": 3.27,
      "learning_rate": 1.3574892699235797e-05,
      "loss": 0.9017,
      "step": 2237
    },
    {
      "epoch": 3.27,
      "learning_rate": 1.3515628039385365e-05,
      "loss": 0.9208,
      "step": 2238
    },
    {
      "epoch": 3.28,
      "learning_rate": 1.3456483652741591e-05,
      "loss": 0.9155,
      "step": 2239
    },
    {
      "epoch": 3.28,
      "learning_rate": 1.339745962155613e-05,
      "loss": 0.915,
      "step": 2240
    },
    {
      "epoch": 3.28,
      "learning_rate": 1.333855602791333e-05,
      "loss": 0.8611,
      "step": 2241
    },
    {
      "epoch": 3.28,
      "learning_rate": 1.3279772953729986e-05,
      "loss": 0.8849,
      "step": 2242
    },
    {
      "epoch": 3.28,
      "learning_rate": 1.3221110480755305e-05,
      "loss": 0.8616,
      "step": 2243
    },
    {
      "epoch": 3.28,
      "learning_rate": 1.3162568690570743e-05,
      "loss": 0.8999,
      "step": 2244
    },
    {
      "epoch": 3.28,
      "learning_rate": 1.3104147664589973e-05,
      "loss": 0.9156,
      "step": 2245
    },
    {
      "epoch": 3.29,
      "learning_rate": 1.3045847484058748e-05,
      "loss": 0.9132,
      "step": 2246
    },
    {
      "epoch": 3.29,
      "learning_rate": 1.2987668230054684e-05,
      "loss": 0.862,
      "step": 2247
    },
    {
      "epoch": 3.29,
      "learning_rate": 1.2929609983487257e-05,
      "loss": 0.8309,
      "step": 2248
    },
    {
      "epoch": 3.29,
      "learning_rate": 1.287167282509767e-05,
      "loss": 0.814,
      "step": 2249
    },
    {
      "epoch": 3.29,
      "learning_rate": 1.2813856835458682e-05,
      "loss": 0.8193,
      "step": 2250
    },
    {
      "epoch": 3.29,
      "learning_rate": 1.2756162094974644e-05,
      "loss": 0.8668,
      "step": 2251
    },
    {
      "epoch": 3.3,
      "learning_rate": 1.2698588683881186e-05,
      "loss": 0.8173,
      "step": 2252
    },
    {
      "epoch": 3.3,
      "learning_rate": 1.2641136682245257e-05,
      "loss": 0.8263,
      "step": 2253
    },
    {
      "epoch": 3.3,
      "learning_rate": 1.2583806169964961e-05,
      "loss": 0.865,
      "step": 2254
    },
    {
      "epoch": 3.3,
      "learning_rate": 1.252659722676941e-05,
      "loss": 0.8536,
      "step": 2255
    },
    {
      "epoch": 3.3,
      "learning_rate": 1.2469509932218703e-05,
      "loss": 0.8811,
      "step": 2256
    },
    {
      "epoch": 3.3,
      "learning_rate": 1.2412544365703738e-05,
      "loss": 0.8152,
      "step": 2257
    },
    {
      "epoch": 3.3,
      "learning_rate": 1.235570060644612e-05,
      "loss": 0.8511,
      "step": 2258
    },
    {
      "epoch": 3.31,
      "learning_rate": 1.2298978733498035e-05,
      "loss": 0.9096,
      "step": 2259
    },
    {
      "epoch": 3.31,
      "learning_rate": 1.224237882574224e-05,
      "loss": 0.8595,
      "step": 2260
    },
    {
      "epoch": 3.31,
      "learning_rate": 1.2185900961891794e-05,
      "loss": 0.9332,
      "step": 2261
    },
    {
      "epoch": 3.31,
      "learning_rate": 1.2129545220490102e-05,
      "loss": 0.9013,
      "step": 2262
    },
    {
      "epoch": 3.31,
      "learning_rate": 1.207331167991066e-05,
      "loss": 0.9224,
      "step": 2263
    },
    {
      "epoch": 3.31,
      "learning_rate": 1.2017200418357078e-05,
      "loss": 0.9001,
      "step": 2264
    },
    {
      "epoch": 3.31,
      "learning_rate": 1.1961211513862858e-05,
      "loss": 0.8989,
      "step": 2265
    },
    {
      "epoch": 3.32,
      "learning_rate": 1.1905345044291426e-05,
      "loss": 0.8607,
      "step": 2266
    },
    {
      "epoch": 3.32,
      "learning_rate": 1.184960108733586e-05,
      "loss": 0.872,
      "step": 2267
    },
    {
      "epoch": 3.32,
      "learning_rate": 1.1793979720518866e-05,
      "loss": 0.848,
      "step": 2268
    },
    {
      "epoch": 3.32,
      "learning_rate": 1.1738481021192704e-05,
      "loss": 0.875,
      "step": 2269
    },
    {
      "epoch": 3.32,
      "learning_rate": 1.1683105066539068e-05,
      "loss": 0.8266,
      "step": 2270
    },
    {
      "epoch": 3.32,
      "learning_rate": 1.1627851933568856e-05,
      "loss": 0.9271,
      "step": 2271
    },
    {
      "epoch": 3.32,
      "learning_rate": 1.1572721699122236e-05,
      "loss": 0.9028,
      "step": 2272
    },
    {
      "epoch": 3.33,
      "learning_rate": 1.151771443986842e-05,
      "loss": 0.8882,
      "step": 2273
    },
    {
      "epoch": 3.33,
      "learning_rate": 1.14628302323056e-05,
      "loss": 0.9062,
      "step": 2274
    },
    {
      "epoch": 3.33,
      "learning_rate": 1.1408069152760903e-05,
      "loss": 0.8219,
      "step": 2275
    },
    {
      "epoch": 3.33,
      "learning_rate": 1.1353431277390126e-05,
      "loss": 0.8061,
      "step": 2276
    },
    {
      "epoch": 3.33,
      "learning_rate": 1.129891668217783e-05,
      "loss": 0.8517,
      "step": 2277
    },
    {
      "epoch": 3.33,
      "learning_rate": 1.1244525442937049e-05,
      "loss": 0.8283,
      "step": 2278
    },
    {
      "epoch": 3.34,
      "learning_rate": 1.1190257635309275e-05,
      "loss": 0.9096,
      "step": 2279
    },
    {
      "epoch": 3.34,
      "learning_rate": 1.113611333476442e-05,
      "loss": 0.9243,
      "step": 2280
    },
    {
      "epoch": 3.34,
      "learning_rate": 1.1082092616600538e-05,
      "loss": 0.9085,
      "step": 2281
    },
    {
      "epoch": 3.34,
      "learning_rate": 1.1028195555943877e-05,
      "loss": 0.8516,
      "step": 2282
    },
    {
      "epoch": 3.34,
      "learning_rate": 1.0974422227748704e-05,
      "loss": 0.8565,
      "step": 2283
    },
    {
      "epoch": 3.34,
      "learning_rate": 1.0920772706797167e-05,
      "loss": 0.8599,
      "step": 2284
    },
    {
      "epoch": 3.34,
      "learning_rate": 1.0867247067699315e-05,
      "loss": 0.9778,
      "step": 2285
    },
    {
      "epoch": 3.35,
      "learning_rate": 1.0813845384892896e-05,
      "loss": 0.886,
      "step": 2286
    },
    {
      "epoch": 3.35,
      "learning_rate": 1.0760567732643223e-05,
      "loss": 0.9232,
      "step": 2287
    },
    {
      "epoch": 3.35,
      "learning_rate": 1.0707414185043163e-05,
      "loss": 0.7584,
      "step": 2288
    },
    {
      "epoch": 3.35,
      "learning_rate": 1.0654384816012953e-05,
      "loss": 0.8672,
      "step": 2289
    },
    {
      "epoch": 3.35,
      "learning_rate": 1.0601479699300209e-05,
      "loss": 0.8809,
      "step": 2290
    },
    {
      "epoch": 3.35,
      "learning_rate": 1.054869890847967e-05,
      "loss": 0.8126,
      "step": 2291
    },
    {
      "epoch": 3.35,
      "learning_rate": 1.0496042516953209e-05,
      "loss": 0.9004,
      "step": 2292
    },
    {
      "epoch": 3.36,
      "learning_rate": 1.0443510597949724e-05,
      "loss": 0.8827,
      "step": 2293
    },
    {
      "epoch": 3.36,
      "learning_rate": 1.0391103224524956e-05,
      "loss": 0.8887,
      "step": 2294
    },
    {
      "epoch": 3.36,
      "learning_rate": 1.0338820469561494e-05,
      "loss": 0.8627,
      "step": 2295
    },
    {
      "epoch": 3.36,
      "learning_rate": 1.02866624057686e-05,
      "loss": 0.8772,
      "step": 2296
    },
    {
      "epoch": 3.36,
      "learning_rate": 1.0234629105682103e-05,
      "loss": 0.9564,
      "step": 2297
    },
    {
      "epoch": 3.36,
      "learning_rate": 1.0182720641664356e-05,
      "loss": 0.885,
      "step": 2298
    },
    {
      "epoch": 3.37,
      "learning_rate": 1.013093708590408e-05,
      "loss": 0.8678,
      "step": 2299
    },
    {
      "epoch": 3.37,
      "learning_rate": 1.0079278510416313e-05,
      "loss": 0.9017,
      "step": 2300
    },
    {
      "epoch": 3.37,
      "learning_rate": 1.0027744987042299e-05,
      "loss": 0.9267,
      "step": 2301
    },
    {
      "epoch": 3.37,
      "learning_rate": 9.976336587449309e-06,
      "loss": 0.8748,
      "step": 2302
    },
    {
      "epoch": 3.37,
      "learning_rate": 9.925053383130667e-06,
      "loss": 0.8395,
      "step": 2303
    },
    {
      "epoch": 3.37,
      "learning_rate": 9.873895445405523e-06,
      "loss": 0.9574,
      "step": 2304
    },
    {
      "epoch": 3.37,
      "learning_rate": 9.822862845418912e-06,
      "loss": 0.8472,
      "step": 2305
    },
    {
      "epoch": 3.38,
      "learning_rate": 9.771955654141496e-06,
      "loss": 0.904,
      "step": 2306
    },
    {
      "epoch": 3.38,
      "learning_rate": 9.721173942369521e-06,
      "loss": 0.7936,
      "step": 2307
    },
    {
      "epoch": 3.38,
      "learning_rate": 9.670517780724775e-06,
      "loss": 0.7711,
      "step": 2308
    },
    {
      "epoch": 3.38,
      "learning_rate": 9.619987239654405e-06,
      "loss": 0.818,
      "step": 2309
    },
    {
      "epoch": 3.38,
      "learning_rate": 9.56958238943093e-06,
      "loss": 0.8908,
      "step": 2310
    },
    {
      "epoch": 3.38,
      "learning_rate": 9.519303300151983e-06,
      "loss": 0.9463,
      "step": 2311
    },
    {
      "epoch": 3.38,
      "learning_rate": 9.469150041740338e-06,
      "loss": 0.8929,
      "step": 2312
    },
    {
      "epoch": 3.39,
      "learning_rate": 9.419122683943793e-06,
      "loss": 0.8494,
      "step": 2313
    },
    {
      "epoch": 3.39,
      "learning_rate": 9.369221296335006e-06,
      "loss": 0.862,
      "step": 2314
    },
    {
      "epoch": 3.39,
      "learning_rate": 9.319445948311534e-06,
      "loss": 0.8727,
      "step": 2315
    },
    {
      "epoch": 3.39,
      "learning_rate": 9.269796709095558e-06,
      "loss": 0.8559,
      "step": 2316
    },
    {
      "epoch": 3.39,
      "learning_rate": 9.220273647733969e-06,
      "loss": 0.8936,
      "step": 2317
    },
    {
      "epoch": 3.39,
      "learning_rate": 9.170876833098118e-06,
      "loss": 0.8938,
      "step": 2318
    },
    {
      "epoch": 3.39,
      "learning_rate": 9.121606333883792e-06,
      "loss": 0.8864,
      "step": 2319
    },
    {
      "epoch": 3.4,
      "learning_rate": 9.072462218611155e-06,
      "loss": 0.865,
      "step": 2320
    },
    {
      "epoch": 3.4,
      "learning_rate": 9.023444555624571e-06,
      "loss": 0.9247,
      "step": 2321
    },
    {
      "epoch": 3.4,
      "learning_rate": 8.974553413092556e-06,
      "loss": 0.93,
      "step": 2322
    },
    {
      "epoch": 3.4,
      "learning_rate": 8.925788859007656e-06,
      "loss": 0.9428,
      "step": 2323
    },
    {
      "epoch": 3.4,
      "learning_rate": 8.87715096118642e-06,
      "loss": 0.8987,
      "step": 2324
    },
    {
      "epoch": 3.4,
      "learning_rate": 8.828639787269244e-06,
      "loss": 0.8987,
      "step": 2325
    },
    {
      "epoch": 3.41,
      "learning_rate": 8.78025540472025e-06,
      "loss": 0.8056,
      "step": 2326
    },
    {
      "epoch": 3.41,
      "learning_rate": 8.731997880827258e-06,
      "loss": 0.9573,
      "step": 2327
    },
    {
      "epoch": 3.41,
      "learning_rate": 8.683867282701686e-06,
      "loss": 0.8489,
      "step": 2328
    },
    {
      "epoch": 3.41,
      "learning_rate": 8.635863677278378e-06,
      "loss": 0.7916,
      "step": 2329
    },
    {
      "epoch": 3.41,
      "learning_rate": 8.587987131315656e-06,
      "loss": 0.8372,
      "step": 2330
    },
    {
      "epoch": 3.41,
      "learning_rate": 8.540237711395072e-06,
      "loss": 0.903,
      "step": 2331
    },
    {
      "epoch": 3.41,
      "learning_rate": 8.492615483921395e-06,
      "loss": 0.9401,
      "step": 2332
    },
    {
      "epoch": 3.42,
      "learning_rate": 8.445120515122551e-06,
      "loss": 0.8648,
      "step": 2333
    },
    {
      "epoch": 3.42,
      "learning_rate": 8.397752871049436e-06,
      "loss": 0.9071,
      "step": 2334
    },
    {
      "epoch": 3.42,
      "learning_rate": 8.350512617575912e-06,
      "loss": 0.9137,
      "step": 2335
    },
    {
      "epoch": 3.42,
      "learning_rate": 8.303399820398672e-06,
      "loss": 0.8674,
      "step": 2336
    },
    {
      "epoch": 3.42,
      "learning_rate": 8.256414545037127e-06,
      "loss": 0.8954,
      "step": 2337
    },
    {
      "epoch": 3.42,
      "learning_rate": 8.2095568568334e-06,
      "loss": 0.8513,
      "step": 2338
    },
    {
      "epoch": 3.42,
      "learning_rate": 8.162826820952097e-06,
      "loss": 0.877,
      "step": 2339
    },
    {
      "epoch": 3.43,
      "learning_rate": 8.116224502380387e-06,
      "loss": 0.8524,
      "step": 2340
    },
    {
      "epoch": 3.43,
      "learning_rate": 8.069749965927808e-06,
      "loss": 0.8995,
      "step": 2341
    },
    {
      "epoch": 3.43,
      "learning_rate": 8.023403276226126e-06,
      "loss": 0.8799,
      "step": 2342
    },
    {
      "epoch": 3.43,
      "learning_rate": 7.977184497729384e-06,
      "loss": 0.8406,
      "step": 2343
    },
    {
      "epoch": 3.43,
      "learning_rate": 7.931093694713687e-06,
      "loss": 0.8906,
      "step": 2344
    },
    {
      "epoch": 3.43,
      "learning_rate": 7.885130931277218e-06,
      "loss": 0.8706,
      "step": 2345
    },
    {
      "epoch": 3.44,
      "learning_rate": 7.839296271340058e-06,
      "loss": 0.8608,
      "step": 2346
    },
    {
      "epoch": 3.44,
      "learning_rate": 7.793589778644116e-06,
      "loss": 0.9749,
      "step": 2347
    },
    {
      "epoch": 3.44,
      "learning_rate": 7.74801151675314e-06,
      "loss": 0.8547,
      "step": 2348
    },
    {
      "epoch": 3.44,
      "learning_rate": 7.702561549052445e-06,
      "loss": 0.8711,
      "step": 2349
    },
    {
      "epoch": 3.44,
      "learning_rate": 7.65723993874904e-06,
      "loss": 0.8758,
      "step": 2350
    },
    {
      "epoch": 3.44,
      "learning_rate": 7.612046748871327e-06,
      "loss": 0.8865,
      "step": 2351
    },
    {
      "epoch": 3.44,
      "learning_rate": 7.566982042269177e-06,
      "loss": 0.8912,
      "step": 2352
    },
    {
      "epoch": 3.44,
      "eval_loss": 1.004106879234314,
      "eval_runtime": 2.6237,
      "eval_samples_per_second": 416.206,
      "eval_steps_per_second": 26.299,
      "step": 2352
    },
    {
      "epoch": 3.45,
      "learning_rate": 7.522045881613737e-06,
      "loss": 0.932,
      "step": 2353
    },
    {
      "epoch": 3.45,
      "learning_rate": 7.477238329397418e-06,
      "loss": 0.9275,
      "step": 2354
    },
    {
      "epoch": 3.45,
      "learning_rate": 7.432559447933785e-06,
      "loss": 0.8213,
      "step": 2355
    },
    {
      "epoch": 3.45,
      "learning_rate": 7.3880092993574125e-06,
      "loss": 0.9243,
      "step": 2356
    },
    {
      "epoch": 3.45,
      "learning_rate": 7.3435879456239085e-06,
      "loss": 0.894,
      "step": 2357
    },
    {
      "epoch": 3.45,
      "learning_rate": 7.299295448509724e-06,
      "loss": 0.8915,
      "step": 2358
    },
    {
      "epoch": 3.45,
      "learning_rate": 7.255131869612108e-06,
      "loss": 0.8819,
      "step": 2359
    },
    {
      "epoch": 3.46,
      "learning_rate": 7.211097270349066e-06,
      "loss": 0.9231,
      "step": 2360
    },
    {
      "epoch": 3.46,
      "learning_rate": 7.167191711959198e-06,
      "loss": 0.8868,
      "step": 2361
    },
    {
      "epoch": 3.46,
      "learning_rate": 7.123415255501653e-06,
      "loss": 0.9077,
      "step": 2362
    },
    {
      "epoch": 3.46,
      "learning_rate": 7.0797679618560095e-06,
      "loss": 0.8692,
      "step": 2363
    },
    {
      "epoch": 3.46,
      "learning_rate": 7.03624989172228e-06,
      "loss": 0.8744,
      "step": 2364
    },
    {
      "epoch": 3.46,
      "learning_rate": 6.992861105620752e-06,
      "loss": 0.8121,
      "step": 2365
    },
    {
      "epoch": 3.46,
      "learning_rate": 6.949601663891891e-06,
      "loss": 0.8156,
      "step": 2366
    },
    {
      "epoch": 3.47,
      "learning_rate": 6.906471626696287e-06,
      "loss": 0.8274,
      "step": 2367
    },
    {
      "epoch": 3.47,
      "learning_rate": 6.863471054014592e-06,
      "loss": 0.9709,
      "step": 2368
    },
    {
      "epoch": 3.47,
      "learning_rate": 6.820600005647382e-06,
      "loss": 0.9252,
      "step": 2369
    },
    {
      "epoch": 3.47,
      "learning_rate": 6.777858541215143e-06,
      "loss": 0.9281,
      "step": 2370
    },
    {
      "epoch": 3.47,
      "learning_rate": 6.735246720158117e-06,
      "loss": 0.9156,
      "step": 2371
    },
    {
      "epoch": 3.47,
      "learning_rate": 6.692764601736268e-06,
      "loss": 0.855,
      "step": 2372
    },
    {
      "epoch": 3.48,
      "learning_rate": 6.6504122450291804e-06,
      "loss": 0.8655,
      "step": 2373
    },
    {
      "epoch": 3.48,
      "learning_rate": 6.608189708935964e-06,
      "loss": 0.8569,
      "step": 2374
    },
    {
      "epoch": 3.48,
      "learning_rate": 6.566097052175213e-06,
      "loss": 0.8501,
      "step": 2375
    },
    {
      "epoch": 3.48,
      "learning_rate": 6.524134333284904e-06,
      "loss": 0.9269,
      "step": 2376
    },
    {
      "epoch": 3.48,
      "learning_rate": 6.4823016106222654e-06,
      "loss": 0.9151,
      "step": 2377
    },
    {
      "epoch": 3.48,
      "learning_rate": 6.440598942363796e-06,
      "loss": 0.8744,
      "step": 2378
    },
    {
      "epoch": 3.48,
      "learning_rate": 6.3990263865050695e-06,
      "loss": 0.9239,
      "step": 2379
    },
    {
      "epoch": 3.49,
      "learning_rate": 6.357584000860761e-06,
      "loss": 0.9291,
      "step": 2380
    },
    {
      "epoch": 3.49,
      "learning_rate": 6.316271843064536e-06,
      "loss": 0.9394,
      "step": 2381
    },
    {
      "epoch": 3.49,
      "learning_rate": 6.275089970568882e-06,
      "loss": 0.8364,
      "step": 2382
    },
    {
      "epoch": 3.49,
      "learning_rate": 6.234038440645163e-06,
      "loss": 0.9144,
      "step": 2383
    },
    {
      "epoch": 3.49,
      "learning_rate": 6.1931173103834115e-06,
      "loss": 0.8727,
      "step": 2384
    },
    {
      "epoch": 3.49,
      "learning_rate": 6.152326636692418e-06,
      "loss": 0.8763,
      "step": 2385
    },
    {
      "epoch": 3.49,
      "learning_rate": 6.111666476299438e-06,
      "loss": 0.8333,
      "step": 2386
    },
    {
      "epoch": 3.5,
      "learning_rate": 6.071136885750272e-06,
      "loss": 0.8994,
      "step": 2387
    },
    {
      "epoch": 3.5,
      "learning_rate": 6.030737921409169e-06,
      "loss": 0.9433,
      "step": 2388
    },
    {
      "epoch": 3.5,
      "learning_rate": 5.9904696394586405e-06,
      "loss": 0.9362,
      "step": 2389
    },
    {
      "epoch": 3.5,
      "learning_rate": 5.950332095899547e-06,
      "loss": 0.948,
      "step": 2390
    },
    {
      "epoch": 3.5,
      "learning_rate": 5.9103253465508605e-06,
      "loss": 0.8242,
      "step": 2391
    },
    {
      "epoch": 3.5,
      "learning_rate": 5.870449447049686e-06,
      "loss": 0.918,
      "step": 2392
    },
    {
      "epoch": 3.51,
      "learning_rate": 5.830704452851166e-06,
      "loss": 0.8843,
      "step": 2393
    },
    {
      "epoch": 3.51,
      "learning_rate": 5.791090419228351e-06,
      "loss": 0.8805,
      "step": 2394
    },
    {
      "epoch": 3.51,
      "learning_rate": 5.751607401272241e-06,
      "loss": 0.9051,
      "step": 2395
    },
    {
      "epoch": 3.51,
      "learning_rate": 5.71225545389158e-06,
      "loss": 0.9524,
      "step": 2396
    },
    {
      "epoch": 3.51,
      "learning_rate": 5.6730346318128455e-06,
      "loss": 1.0244,
      "step": 2397
    },
    {
      "epoch": 3.51,
      "learning_rate": 5.633944989580153e-06,
      "loss": 0.8581,
      "step": 2398
    },
    {
      "epoch": 3.51,
      "learning_rate": 5.594986581555173e-06,
      "loss": 0.8754,
      "step": 2399
    },
    {
      "epoch": 3.52,
      "learning_rate": 5.556159461917121e-06,
      "loss": 0.9386,
      "step": 2400
    },
    {
      "epoch": 3.52,
      "learning_rate": 5.517463684662582e-06,
      "loss": 0.8832,
      "step": 2401
    },
    {
      "epoch": 3.52,
      "learning_rate": 5.4788993036055115e-06,
      "loss": 0.8422,
      "step": 2402
    },
    {
      "epoch": 3.52,
      "learning_rate": 5.440466372377095e-06,
      "loss": 0.8945,
      "step": 2403
    },
    {
      "epoch": 3.52,
      "learning_rate": 5.402164944425758e-06,
      "loss": 0.8111,
      "step": 2404
    },
    {
      "epoch": 3.52,
      "learning_rate": 5.363995073017047e-06,
      "loss": 0.8646,
      "step": 2405
    },
    {
      "epoch": 3.52,
      "learning_rate": 5.325956811233512e-06,
      "loss": 0.849,
      "step": 2406
    },
    {
      "epoch": 3.53,
      "learning_rate": 5.288050211974694e-06,
      "loss": 0.8714,
      "step": 2407
    },
    {
      "epoch": 3.53,
      "learning_rate": 5.250275327957032e-06,
      "loss": 0.8775,
      "step": 2408
    },
    {
      "epoch": 3.53,
      "learning_rate": 5.212632211713797e-06,
      "loss": 0.8481,
      "step": 2409
    },
    {
      "epoch": 3.53,
      "learning_rate": 5.1751209155949995e-06,
      "loss": 0.9742,
      "step": 2410
    },
    {
      "epoch": 3.53,
      "learning_rate": 5.137741491767345e-06,
      "loss": 0.8831,
      "step": 2411
    },
    {
      "epoch": 3.53,
      "learning_rate": 5.100493992214128e-06,
      "loss": 0.8925,
      "step": 2412
    },
    {
      "epoch": 3.54,
      "learning_rate": 5.0633784687351915e-06,
      "loss": 0.9058,
      "step": 2413
    },
    {
      "epoch": 3.54,
      "learning_rate": 5.026394972946813e-06,
      "loss": 0.9003,
      "step": 2414
    },
    {
      "epoch": 3.54,
      "learning_rate": 4.989543556281695e-06,
      "loss": 0.8197,
      "step": 2415
    },
    {
      "epoch": 3.54,
      "learning_rate": 4.9528242699888535e-06,
      "loss": 0.8675,
      "step": 2416
    },
    {
      "epoch": 3.54,
      "learning_rate": 4.916237165133519e-06,
      "loss": 0.9537,
      "step": 2417
    },
    {
      "epoch": 3.54,
      "learning_rate": 4.8797822925971235e-06,
      "loss": 0.9662,
      "step": 2418
    },
    {
      "epoch": 3.54,
      "learning_rate": 4.843459703077202e-06,
      "loss": 0.9332,
      "step": 2419
    },
    {
      "epoch": 3.55,
      "learning_rate": 4.807269447087348e-06,
      "loss": 0.828,
      "step": 2420
    },
    {
      "epoch": 3.55,
      "learning_rate": 4.7712115749570685e-06,
      "loss": 0.8799,
      "step": 2421
    },
    {
      "epoch": 3.55,
      "learning_rate": 4.735286136831807e-06,
      "loss": 0.9163,
      "step": 2422
    },
    {
      "epoch": 3.55,
      "learning_rate": 4.69949318267281e-06,
      "loss": 0.8946,
      "step": 2423
    },
    {
      "epoch": 3.55,
      "learning_rate": 4.66383276225707e-06,
      "loss": 0.9073,
      "step": 2424
    },
    {
      "epoch": 3.55,
      "learning_rate": 4.628304925177318e-06,
      "loss": 0.8648,
      "step": 2425
    },
    {
      "epoch": 3.55,
      "learning_rate": 4.5929097208418425e-06,
      "loss": 0.8846,
      "step": 2426
    },
    {
      "epoch": 3.56,
      "learning_rate": 4.557647198474491e-06,
      "loss": 0.9423,
      "step": 2427
    },
    {
      "epoch": 3.56,
      "learning_rate": 4.5225174071146455e-06,
      "loss": 0.8671,
      "step": 2428
    },
    {
      "epoch": 3.56,
      "learning_rate": 4.487520395617029e-06,
      "loss": 0.9262,
      "step": 2429
    },
    {
      "epoch": 3.56,
      "learning_rate": 4.452656212651751e-06,
      "loss": 0.8335,
      "step": 2430
    },
    {
      "epoch": 3.56,
      "learning_rate": 4.417924906704185e-06,
      "loss": 0.8801,
      "step": 2431
    },
    {
      "epoch": 3.56,
      "learning_rate": 4.383326526074916e-06,
      "loss": 0.895,
      "step": 2432
    },
    {
      "epoch": 3.56,
      "learning_rate": 4.3488611188796545e-06,
      "loss": 0.9591,
      "step": 2433
    },
    {
      "epoch": 3.57,
      "learning_rate": 4.314528733049206e-06,
      "loss": 0.8998,
      "step": 2434
    },
    {
      "epoch": 3.57,
      "learning_rate": 4.280329416329365e-06,
      "loss": 0.9133,
      "step": 2435
    },
    {
      "epoch": 3.57,
      "learning_rate": 4.24626321628091e-06,
      "loss": 0.8861,
      "step": 2436
    },
    {
      "epoch": 3.57,
      "learning_rate": 4.212330180279456e-06,
      "loss": 0.8638,
      "step": 2437
    },
    {
      "epoch": 3.57,
      "learning_rate": 4.1785303555154085e-06,
      "loss": 0.8281,
      "step": 2438
    },
    {
      "epoch": 3.57,
      "learning_rate": 4.144863788993991e-06,
      "loss": 0.9246,
      "step": 2439
    },
    {
      "epoch": 3.58,
      "learning_rate": 4.111330527535029e-06,
      "loss": 0.9092,
      "step": 2440
    },
    {
      "epoch": 3.58,
      "learning_rate": 4.077930617773007e-06,
      "loss": 0.8681,
      "step": 2441
    },
    {
      "epoch": 3.58,
      "learning_rate": 4.044664106156915e-06,
      "loss": 0.8603,
      "step": 2442
    },
    {
      "epoch": 3.58,
      "learning_rate": 4.0115310389503004e-06,
      "loss": 0.8667,
      "step": 2443
    },
    {
      "epoch": 3.58,
      "learning_rate": 3.9785314622310495e-06,
      "loss": 0.921,
      "step": 2444
    },
    {
      "epoch": 3.58,
      "learning_rate": 3.945665421891465e-06,
      "loss": 0.8891,
      "step": 2445
    },
    {
      "epoch": 3.58,
      "learning_rate": 3.912932963638116e-06,
      "loss": 0.8857,
      "step": 2446
    },
    {
      "epoch": 3.59,
      "learning_rate": 3.880334132991792e-06,
      "loss": 0.8933,
      "step": 2447
    },
    {
      "epoch": 3.59,
      "learning_rate": 3.84786897528745e-06,
      "loss": 0.961,
      "step": 2448
    },
    {
      "epoch": 3.59,
      "learning_rate": 3.815537535674174e-06,
      "loss": 0.9356,
      "step": 2449
    },
    {
      "epoch": 3.59,
      "learning_rate": 3.783339859115065e-06,
      "loss": 0.907,
      "step": 2450
    },
    {
      "epoch": 3.59,
      "learning_rate": 3.751275990387193e-06,
      "loss": 0.7907,
      "step": 2451
    },
    {
      "epoch": 3.59,
      "learning_rate": 3.7193459740815674e-06,
      "loss": 0.832,
      "step": 2452
    },
    {
      "epoch": 3.59,
      "learning_rate": 3.687549854603023e-06,
      "loss": 0.8745,
      "step": 2453
    },
    {
      "epoch": 3.6,
      "learning_rate": 3.655887676170222e-06,
      "loss": 0.8743,
      "step": 2454
    },
    {
      "epoch": 3.6,
      "learning_rate": 3.624359482815509e-06,
      "loss": 0.8262,
      "step": 2455
    },
    {
      "epoch": 3.6,
      "learning_rate": 3.592965318384944e-06,
      "loss": 0.8567,
      "step": 2456
    },
    {
      "epoch": 3.6,
      "learning_rate": 3.561705226538148e-06,
      "loss": 0.8733,
      "step": 2457
    },
    {
      "epoch": 3.6,
      "learning_rate": 3.5305792507483117e-06,
      "loss": 0.9398,
      "step": 2458
    },
    {
      "epoch": 3.6,
      "learning_rate": 3.4995874343021094e-06,
      "loss": 0.8461,
      "step": 2459
    },
    {
      "epoch": 3.61,
      "learning_rate": 3.4687298202996655e-06,
      "loss": 0.8773,
      "step": 2460
    },
    {
      "epoch": 3.61,
      "learning_rate": 3.4380064516544297e-06,
      "loss": 0.9523,
      "step": 2461
    },
    {
      "epoch": 3.61,
      "learning_rate": 3.40741737109318e-06,
      "loss": 0.8672,
      "step": 2462
    },
    {
      "epoch": 3.61,
      "learning_rate": 3.3769626211559102e-06,
      "loss": 0.8343,
      "step": 2463
    },
    {
      "epoch": 3.61,
      "learning_rate": 3.3466422441958634e-06,
      "loss": 0.8847,
      "step": 2464
    },
    {
      "epoch": 3.61,
      "learning_rate": 3.3164562823793653e-06,
      "loss": 0.8854,
      "step": 2465
    },
    {
      "epoch": 3.61,
      "learning_rate": 3.286404777685792e-06,
      "loss": 0.8703,
      "step": 2466
    },
    {
      "epoch": 3.62,
      "learning_rate": 3.25648777190759e-06,
      "loss": 0.8572,
      "step": 2467
    },
    {
      "epoch": 3.62,
      "learning_rate": 3.226705306650113e-06,
      "loss": 0.8701,
      "step": 2468
    },
    {
      "epoch": 3.62,
      "learning_rate": 3.1970574233316397e-06,
      "loss": 0.8247,
      "step": 2469
    },
    {
      "epoch": 3.62,
      "learning_rate": 3.167544163183256e-06,
      "loss": 0.9345,
      "step": 2470
    },
    {
      "epoch": 3.62,
      "learning_rate": 3.138165567248863e-06,
      "loss": 0.8971,
      "step": 2471
    },
    {
      "epoch": 3.62,
      "learning_rate": 3.1089216763850572e-06,
      "loss": 0.9541,
      "step": 2472
    },
    {
      "epoch": 3.62,
      "learning_rate": 3.079812531261095e-06,
      "loss": 0.8111,
      "step": 2473
    },
    {
      "epoch": 3.63,
      "learning_rate": 3.050838172358883e-06,
      "loss": 0.8347,
      "step": 2474
    },
    {
      "epoch": 3.63,
      "learning_rate": 3.021998639972845e-06,
      "loss": 0.8189,
      "step": 2475
    },
    {
      "epoch": 3.63,
      "learning_rate": 2.9932939742099208e-06,
      "loss": 0.8757,
      "step": 2476
    },
    {
      "epoch": 3.63,
      "learning_rate": 2.9647242149895006e-06,
      "loss": 0.9051,
      "step": 2477
    },
    {
      "epoch": 3.63,
      "learning_rate": 2.936289402043313e-06,
      "loss": 0.9259,
      "step": 2478
    },
    {
      "epoch": 3.63,
      "learning_rate": 2.9079895749154927e-06,
      "loss": 0.8777,
      "step": 2479
    },
    {
      "epoch": 3.63,
      "learning_rate": 2.8798247729623806e-06,
      "loss": 0.8496,
      "step": 2480
    },
    {
      "epoch": 3.64,
      "learning_rate": 2.851795035352578e-06,
      "loss": 0.8094,
      "step": 2481
    },
    {
      "epoch": 3.64,
      "learning_rate": 2.8239004010668367e-06,
      "loss": 0.8504,
      "step": 2482
    },
    {
      "epoch": 3.64,
      "learning_rate": 2.796140908898026e-06,
      "loss": 0.8756,
      "step": 2483
    },
    {
      "epoch": 3.64,
      "learning_rate": 2.7685165974510986e-06,
      "loss": 0.9191,
      "step": 2484
    },
    {
      "epoch": 3.64,
      "learning_rate": 2.741027505142979e-06,
      "loss": 0.8877,
      "step": 2485
    },
    {
      "epoch": 3.64,
      "learning_rate": 2.7136736702025433e-06,
      "loss": 0.8375,
      "step": 2486
    },
    {
      "epoch": 3.65,
      "learning_rate": 2.6864551306705842e-06,
      "loss": 0.9617,
      "step": 2487
    },
    {
      "epoch": 3.65,
      "learning_rate": 2.659371924399734e-06,
      "loss": 0.8413,
      "step": 2488
    },
    {
      "epoch": 3.65,
      "learning_rate": 2.6324240890544193e-06,
      "loss": 0.9178,
      "step": 2489
    },
    {
      "epoch": 3.65,
      "learning_rate": 2.605611662110785e-06,
      "loss": 0.8993,
      "step": 2490
    },
    {
      "epoch": 3.65,
      "learning_rate": 2.578934680856715e-06,
      "loss": 0.8596,
      "step": 2491
    },
    {
      "epoch": 3.65,
      "learning_rate": 2.552393182391677e-06,
      "loss": 0.908,
      "step": 2492
    },
    {
      "epoch": 3.65,
      "learning_rate": 2.5259872036267564e-06,
      "loss": 0.8652,
      "step": 2493
    },
    {
      "epoch": 3.66,
      "learning_rate": 2.499716781284556e-06,
      "loss": 0.8642,
      "step": 2494
    },
    {
      "epoch": 3.66,
      "learning_rate": 2.473581951899184e-06,
      "loss": 0.8744,
      "step": 2495
    },
    {
      "epoch": 3.66,
      "learning_rate": 2.4475827518161356e-06,
      "loss": 0.8064,
      "step": 2496
    },
    {
      "epoch": 3.66,
      "learning_rate": 2.421719217192331e-06,
      "loss": 0.7801,
      "step": 2497
    },
    {
      "epoch": 3.66,
      "learning_rate": 2.395991383995999e-06,
      "loss": 0.9131,
      "step": 2498
    },
    {
      "epoch": 3.66,
      "learning_rate": 2.3703992880066638e-06,
      "loss": 0.8389,
      "step": 2499
    },
    {
      "epoch": 3.66,
      "learning_rate": 2.3449429648150665e-06,
      "loss": 0.8062,
      "step": 2500
    },
    {
      "epoch": 3.67,
      "learning_rate": 2.3196224498231443e-06,
      "loss": 0.9015,
      "step": 2501
    },
    {
      "epoch": 3.67,
      "learning_rate": 2.294437778243963e-06,
      "loss": 0.9179,
      "step": 2502
    },
    {
      "epoch": 3.67,
      "learning_rate": 2.2693889851016393e-06,
      "loss": 0.9013,
      "step": 2503
    },
    {
      "epoch": 3.67,
      "learning_rate": 2.2444761052313856e-06,
      "loss": 0.8867,
      "step": 2504
    },
    {
      "epoch": 3.67,
      "learning_rate": 2.219699173279355e-06,
      "loss": 0.8618,
      "step": 2505
    },
    {
      "epoch": 3.67,
      "learning_rate": 2.1950582237026394e-06,
      "loss": 0.8839,
      "step": 2506
    },
    {
      "epoch": 3.68,
      "learning_rate": 2.1705532907692615e-06,
      "loss": 0.9381,
      "step": 2507
    },
    {
      "epoch": 3.68,
      "learning_rate": 2.1461844085580385e-06,
      "loss": 0.793,
      "step": 2508
    },
    {
      "epoch": 3.68,
      "learning_rate": 2.1219516109586056e-06,
      "loss": 0.8423,
      "step": 2509
    },
    {
      "epoch": 3.68,
      "learning_rate": 2.0978549316713612e-06,
      "loss": 0.8737,
      "step": 2510
    },
    {
      "epoch": 3.68,
      "learning_rate": 2.073894404207366e-06,
      "loss": 0.8675,
      "step": 2511
    },
    {
      "epoch": 3.68,
      "learning_rate": 2.050070061888365e-06,
      "loss": 0.8494,
      "step": 2512
    },
    {
      "epoch": 3.68,
      "learning_rate": 2.0263819378466884e-06,
      "loss": 0.9556,
      "step": 2513
    },
    {
      "epoch": 3.69,
      "learning_rate": 2.002830065025263e-06,
      "loss": 0.8991,
      "step": 2514
    },
    {
      "epoch": 3.69,
      "learning_rate": 1.9794144761775212e-06,
      "loss": 0.8551,
      "step": 2515
    },
    {
      "epoch": 3.69,
      "learning_rate": 1.9561352038673263e-06,
      "loss": 0.8836,
      "step": 2516
    },
    {
      "epoch": 3.69,
      "learning_rate": 1.9329922804690258e-06,
      "loss": 0.9173,
      "step": 2517
    },
    {
      "epoch": 3.69,
      "learning_rate": 1.909985738167308e-06,
      "loss": 0.8696,
      "step": 2518
    },
    {
      "epoch": 3.69,
      "learning_rate": 1.8871156089572018e-06,
      "loss": 0.8177,
      "step": 2519
    },
    {
      "epoch": 3.69,
      "learning_rate": 1.8643819246440542e-06,
      "loss": 0.7828,
      "step": 2520
    },
    {
      "epoch": 3.69,
      "eval_loss": 1.0041353702545166,
      "eval_runtime": 2.6123,
      "eval_samples_per_second": 418.024,
      "eval_steps_per_second": 26.414,
      "step": 2520
    },
    {
      "epoch": 3.7,
      "learning_rate": 1.841784716843431e-06,
      "loss": 0.8021,
      "step": 2521
    },
    {
      "epoch": 3.7,
      "learning_rate": 1.8193240169810943e-06,
      "loss": 0.878,
      "step": 2522
    },
    {
      "epoch": 3.7,
      "learning_rate": 1.7969998562929913e-06,
      "loss": 0.8315,
      "step": 2523
    },
    {
      "epoch": 3.7,
      "learning_rate": 1.7748122658251876e-06,
      "loss": 0.9345,
      "step": 2524
    },
    {
      "epoch": 3.7,
      "learning_rate": 1.7527612764338008e-06,
      "loss": 0.9141,
      "step": 2525
    },
    {
      "epoch": 3.7,
      "learning_rate": 1.7308469187849896e-06,
      "loss": 0.9472,
      "step": 2526
    },
    {
      "epoch": 3.7,
      "learning_rate": 1.7090692233548866e-06,
      "loss": 0.7864,
      "step": 2527
    },
    {
      "epoch": 3.71,
      "learning_rate": 1.6874282204295766e-06,
      "loss": 0.794,
      "step": 2528
    },
    {
      "epoch": 3.71,
      "learning_rate": 1.665923940105074e-06,
      "loss": 0.9277,
      "step": 2529
    },
    {
      "epoch": 3.71,
      "learning_rate": 1.6445564122872015e-06,
      "loss": 0.8783,
      "step": 2530
    },
    {
      "epoch": 3.71,
      "learning_rate": 1.623325666691644e-06,
      "loss": 0.9033,
      "step": 2531
    },
    {
      "epoch": 3.71,
      "learning_rate": 1.6022317328438506e-06,
      "loss": 0.9331,
      "step": 2532
    },
    {
      "epoch": 3.71,
      "learning_rate": 1.5812746400790001e-06,
      "loss": 0.814,
      "step": 2533
    },
    {
      "epoch": 3.72,
      "learning_rate": 1.56045441754199e-06,
      "loss": 0.936,
      "step": 2534
    },
    {
      "epoch": 3.72,
      "learning_rate": 1.5397710941873366e-06,
      "loss": 0.8678,
      "step": 2535
    },
    {
      "epoch": 3.72,
      "learning_rate": 1.5192246987791981e-06,
      "loss": 0.8692,
      "step": 2536
    },
    {
      "epoch": 3.72,
      "learning_rate": 1.4988152598913064e-06,
      "loss": 0.9103,
      "step": 2537
    },
    {
      "epoch": 3.72,
      "learning_rate": 1.478542805906913e-06,
      "loss": 0.8532,
      "step": 2538
    },
    {
      "epoch": 3.72,
      "learning_rate": 1.4584073650187878e-06,
      "loss": 0.8255,
      "step": 2539
    },
    {
      "epoch": 3.72,
      "learning_rate": 1.4384089652291543e-06,
      "loss": 0.8267,
      "step": 2540
    },
    {
      "epoch": 3.73,
      "learning_rate": 1.4185476343496317e-06,
      "loss": 0.8717,
      "step": 2541
    },
    {
      "epoch": 3.73,
      "learning_rate": 1.3988234000012367e-06,
      "loss": 0.9405,
      "step": 2542
    },
    {
      "epoch": 3.73,
      "learning_rate": 1.3792362896143162e-06,
      "loss": 0.8314,
      "step": 2543
    },
    {
      "epoch": 3.73,
      "learning_rate": 1.3597863304285475e-06,
      "loss": 0.8815,
      "step": 2544
    },
    {
      "epoch": 3.73,
      "learning_rate": 1.3404735494928266e-06,
      "loss": 0.8933,
      "step": 2545
    },
    {
      "epoch": 3.73,
      "learning_rate": 1.3212979736653142e-06,
      "loss": 0.8637,
      "step": 2546
    },
    {
      "epoch": 3.73,
      "learning_rate": 1.302259629613356e-06,
      "loss": 0.981,
      "step": 2547
    },
    {
      "epoch": 3.74,
      "learning_rate": 1.2833585438134287e-06,
      "loss": 0.9305,
      "step": 2548
    },
    {
      "epoch": 3.74,
      "learning_rate": 1.2645947425511395e-06,
      "loss": 0.9079,
      "step": 2549
    },
    {
      "epoch": 3.74,
      "learning_rate": 1.2459682519211923e-06,
      "loss": 0.8601,
      "step": 2550
    },
    {
      "epoch": 3.74,
      "learning_rate": 1.2274790978272998e-06,
      "loss": 0.8278,
      "step": 2551
    },
    {
      "epoch": 3.74,
      "learning_rate": 1.209127305982205e-06,
      "loss": 0.8512,
      "step": 2552
    },
    {
      "epoch": 3.74,
      "learning_rate": 1.1909129019076036e-06,
      "loss": 0.9256,
      "step": 2553
    },
    {
      "epoch": 3.75,
      "learning_rate": 1.1728359109341446e-06,
      "loss": 0.8918,
      "step": 2554
    },
    {
      "epoch": 3.75,
      "learning_rate": 1.1548963582013961e-06,
      "loss": 0.8318,
      "step": 2555
    },
    {
      "epoch": 3.75,
      "learning_rate": 1.1370942686577347e-06,
      "loss": 0.8787,
      "step": 2556
    },
    {
      "epoch": 3.75,
      "learning_rate": 1.1194296670604233e-06,
      "loss": 0.9061,
      "step": 2557
    },
    {
      "epoch": 3.75,
      "learning_rate": 1.1019025779754666e-06,
      "loss": 0.8601,
      "step": 2558
    },
    {
      "epoch": 3.75,
      "learning_rate": 1.0845130257777114e-06,
      "loss": 0.8982,
      "step": 2559
    },
    {
      "epoch": 3.75,
      "learning_rate": 1.0672610346506463e-06,
      "loss": 0.8479,
      "step": 2560
    },
    {
      "epoch": 3.76,
      "learning_rate": 1.0501466285865126e-06,
      "loss": 0.8491,
      "step": 2561
    },
    {
      "epoch": 3.76,
      "learning_rate": 1.0331698313861937e-06,
      "loss": 0.9286,
      "step": 2562
    },
    {
      "epoch": 3.76,
      "learning_rate": 1.016330666659182e-06,
      "loss": 0.7757,
      "step": 2563
    },
    {
      "epoch": 3.76,
      "learning_rate": 9.996291578236228e-07,
      "loss": 0.8271,
      "step": 2564
    },
    {
      "epoch": 3.76,
      "learning_rate": 9.830653281061474e-07,
      "loss": 0.8242,
      "step": 2565
    },
    {
      "epoch": 3.76,
      "learning_rate": 9.666392005419856e-07,
      "loss": 0.8585,
      "step": 2566
    },
    {
      "epoch": 3.76,
      "learning_rate": 9.503507979748305e-07,
      "loss": 0.8939,
      "step": 2567
    },
    {
      "epoch": 3.77,
      "learning_rate": 9.342001430568292e-07,
      "loss": 0.9021,
      "step": 2568
    },
    {
      "epoch": 3.77,
      "learning_rate": 9.18187258248604e-07,
      "loss": 0.8612,
      "step": 2569
    },
    {
      "epoch": 3.77,
      "learning_rate": 9.023121658191636e-07,
      "loss": 0.9391,
      "step": 2570
    },
    {
      "epoch": 3.77,
      "learning_rate": 8.865748878458702e-07,
      "loss": 0.9055,
      "step": 2571
    },
    {
      "epoch": 3.77,
      "learning_rate": 8.709754462144615e-07,
      "loss": 0.974,
      "step": 2572
    },
    {
      "epoch": 3.77,
      "learning_rate": 8.555138626189618e-07,
      "loss": 0.9572,
      "step": 2573
    },
    {
      "epoch": 3.77,
      "learning_rate": 8.401901585616823e-07,
      "loss": 0.8094,
      "step": 2574
    },
    {
      "epoch": 3.78,
      "learning_rate": 8.250043553532094e-07,
      "loss": 0.8996,
      "step": 2575
    },
    {
      "epoch": 3.78,
      "learning_rate": 8.099564741123166e-07,
      "loss": 0.8258,
      "step": 2576
    },
    {
      "epoch": 3.78,
      "learning_rate": 7.950465357659864e-07,
      "loss": 0.868,
      "step": 2577
    },
    {
      "epoch": 3.78,
      "learning_rate": 7.802745610493656e-07,
      "loss": 0.8502,
      "step": 2578
    },
    {
      "epoch": 3.78,
      "learning_rate": 7.656405705057435e-07,
      "loss": 0.8995,
      "step": 2579
    },
    {
      "epoch": 3.78,
      "learning_rate": 7.511445844864962e-07,
      "loss": 0.9136,
      "step": 2580
    },
    {
      "epoch": 3.79,
      "learning_rate": 7.367866231511089e-07,
      "loss": 0.8804,
      "step": 2581
    },
    {
      "epoch": 3.79,
      "learning_rate": 7.22566706467076e-07,
      "loss": 0.8206,
      "step": 2582
    },
    {
      "epoch": 3.79,
      "learning_rate": 7.084848542099342e-07,
      "loss": 0.9475,
      "step": 2583
    },
    {
      "epoch": 3.79,
      "learning_rate": 6.945410859632295e-07,
      "loss": 0.8562,
      "step": 2584
    },
    {
      "epoch": 3.79,
      "learning_rate": 6.807354211184613e-07,
      "loss": 0.9127,
      "step": 2585
    },
    {
      "epoch": 3.79,
      "learning_rate": 6.670678788750717e-07,
      "loss": 0.9582,
      "step": 2586
    },
    {
      "epoch": 3.79,
      "learning_rate": 6.535384782404009e-07,
      "loss": 0.9698,
      "step": 2587
    },
    {
      "epoch": 3.8,
      "learning_rate": 6.401472380297091e-07,
      "loss": 0.8934,
      "step": 2588
    },
    {
      "epoch": 3.8,
      "learning_rate": 6.268941768660886e-07,
      "loss": 0.8583,
      "step": 2589
    },
    {
      "epoch": 3.8,
      "learning_rate": 6.137793131804737e-07,
      "loss": 0.7896,
      "step": 2590
    },
    {
      "epoch": 3.8,
      "learning_rate": 6.008026652116305e-07,
      "loss": 0.9075,
      "step": 2591
    },
    {
      "epoch": 3.8,
      "learning_rate": 5.87964251006079e-07,
      "loss": 0.8205,
      "step": 2592
    },
    {
      "epoch": 3.8,
      "learning_rate": 5.752640884181037e-07,
      "loss": 0.8261,
      "step": 2593
    },
    {
      "epoch": 3.8,
      "learning_rate": 5.627021951097545e-07,
      "loss": 0.8202,
      "step": 2594
    },
    {
      "epoch": 3.81,
      "learning_rate": 5.502785885507456e-07,
      "loss": 0.8793,
      "step": 2595
    },
    {
      "epoch": 3.81,
      "learning_rate": 5.379932860185122e-07,
      "loss": 0.8675,
      "step": 2596
    },
    {
      "epoch": 3.81,
      "learning_rate": 5.258463045981432e-07,
      "loss": 0.8722,
      "step": 2597
    },
    {
      "epoch": 3.81,
      "learning_rate": 5.13837661182337e-07,
      "loss": 0.8858,
      "step": 2598
    },
    {
      "epoch": 3.81,
      "learning_rate": 5.019673724714458e-07,
      "loss": 0.9106,
      "step": 2599
    },
    {
      "epoch": 3.81,
      "learning_rate": 4.902354549733978e-07,
      "loss": 0.8488,
      "step": 2600
    },
    {
      "epoch": 3.82,
      "learning_rate": 4.786419250036866e-07,
      "loss": 0.879,
      "step": 2601
    },
    {
      "epoch": 3.82,
      "learning_rate": 4.6718679868533734e-07,
      "loss": 0.9044,
      "step": 2602
    },
    {
      "epoch": 3.82,
      "learning_rate": 4.5587009194894004e-07,
      "loss": 0.8939,
      "step": 2603
    },
    {
      "epoch": 3.82,
      "learning_rate": 4.44691820532539e-07,
      "loss": 0.8272,
      "step": 2604
    },
    {
      "epoch": 3.82,
      "learning_rate": 4.336519999816879e-07,
      "loss": 0.8755,
      "step": 2605
    },
    {
      "epoch": 3.82,
      "learning_rate": 4.227506456493835e-07,
      "loss": 0.9187,
      "step": 2606
    },
    {
      "epoch": 3.82,
      "learning_rate": 4.1198777269605413e-07,
      "loss": 0.9211,
      "step": 2607
    },
    {
      "epoch": 3.83,
      "learning_rate": 4.0136339608957127e-07,
      "loss": 0.9451,
      "step": 2608
    },
    {
      "epoch": 3.83,
      "learning_rate": 3.908775306051604e-07,
      "loss": 0.8035,
      "step": 2609
    },
    {
      "epoch": 3.83,
      "learning_rate": 3.805301908254455e-07,
      "loss": 0.8989,
      "step": 2610
    },
    {
      "epoch": 3.83,
      "learning_rate": 3.7032139114041574e-07,
      "loss": 0.8378,
      "step": 2611
    },
    {
      "epoch": 3.83,
      "learning_rate": 3.6025114574734785e-07,
      "loss": 0.8598,
      "step": 2612
    },
    {
      "epoch": 3.83,
      "learning_rate": 3.5031946865088373e-07,
      "loss": 0.8903,
      "step": 2613
    },
    {
      "epoch": 3.83,
      "learning_rate": 3.405263736629416e-07,
      "loss": 0.8726,
      "step": 2614
    },
    {
      "epoch": 3.84,
      "learning_rate": 3.3087187440268287e-07,
      "loss": 0.9027,
      "step": 2615
    },
    {
      "epoch": 3.84,
      "learning_rate": 3.2135598429657853e-07,
      "loss": 0.8502,
      "step": 2616
    },
    {
      "epoch": 3.84,
      "learning_rate": 3.1197871657828725e-07,
      "loss": 0.9659,
      "step": 2617
    },
    {
      "epoch": 3.84,
      "learning_rate": 3.027400842887218e-07,
      "loss": 0.865,
      "step": 2618
    },
    {
      "epoch": 3.84,
      "learning_rate": 2.9364010027599364e-07,
      "loss": 0.8869,
      "step": 2619
    },
    {
      "epoch": 3.84,
      "learning_rate": 2.846787771953574e-07,
      "loss": 0.9345,
      "step": 2620
    },
    {
      "epoch": 3.85,
      "learning_rate": 2.758561275092886e-07,
      "loss": 0.9017,
      "step": 2621
    },
    {
      "epoch": 3.85,
      "learning_rate": 2.6717216348737253e-07,
      "loss": 0.8029,
      "step": 2622
    },
    {
      "epoch": 3.85,
      "learning_rate": 2.586268972063377e-07,
      "loss": 0.8922,
      "step": 2623
    },
    {
      "epoch": 3.85,
      "learning_rate": 2.5022034055003364e-07,
      "loss": 0.8617,
      "step": 2624
    },
    {
      "epoch": 3.85,
      "learning_rate": 2.419525052093863e-07,
      "loss": 0.8696,
      "step": 2625
    },
    {
      "epoch": 3.85,
      "learning_rate": 2.3382340268242042e-07,
      "loss": 0.9427,
      "step": 2626
    },
    {
      "epoch": 3.85,
      "learning_rate": 2.2583304427421515e-07,
      "loss": 0.9191,
      "step": 2627
    },
    {
      "epoch": 3.86,
      "learning_rate": 2.179814410969261e-07,
      "loss": 0.8317,
      "step": 2628
    },
    {
      "epoch": 3.86,
      "learning_rate": 2.1026860406970772e-07,
      "loss": 0.8796,
      "step": 2629
    },
    {
      "epoch": 3.86,
      "learning_rate": 2.0269454391874666e-07,
      "loss": 0.8337,
      "step": 2630
    },
    {
      "epoch": 3.86,
      "learning_rate": 1.952592711772505e-07,
      "loss": 0.793,
      "step": 2631
    },
    {
      "epoch": 3.86,
      "learning_rate": 1.8796279618537028e-07,
      "loss": 0.8609,
      "step": 2632
    },
    {
      "epoch": 3.86,
      "learning_rate": 1.8080512909028903e-07,
      "loss": 0.8442,
      "step": 2633
    },
    {
      "epoch": 3.86,
      "learning_rate": 1.7378627984612207e-07,
      "loss": 0.8088,
      "step": 2634
    },
    {
      "epoch": 3.87,
      "learning_rate": 1.669062582139169e-07,
      "loss": 0.8417,
      "step": 2635
    },
    {
      "epoch": 3.87,
      "learning_rate": 1.6016507376169777e-07,
      "loss": 0.8682,
      "step": 2636
    },
    {
      "epoch": 3.87,
      "learning_rate": 1.5356273586435432e-07,
      "loss": 0.8889,
      "step": 2637
    },
    {
      "epoch": 3.87,
      "learning_rate": 1.470992537037197e-07,
      "loss": 0.898,
      "step": 2638
    },
    {
      "epoch": 3.87,
      "learning_rate": 1.4077463626852582e-07,
      "loss": 0.9413,
      "step": 2639
    },
    {
      "epoch": 3.87,
      "learning_rate": 1.3458889235435922e-07,
      "loss": 0.8557,
      "step": 2640
    },
    {
      "epoch": 3.87,
      "learning_rate": 1.2854203056369417e-07,
      "loss": 0.8772,
      "step": 2641
    },
    {
      "epoch": 3.88,
      "learning_rate": 1.2263405930585948e-07,
      "loss": 0.9191,
      "step": 2642
    },
    {
      "epoch": 3.88,
      "learning_rate": 1.1686498679702728e-07,
      "loss": 0.8603,
      "step": 2643
    },
    {
      "epoch": 3.88,
      "learning_rate": 1.1123482106021322e-07,
      "loss": 0.813,
      "step": 2644
    },
    {
      "epoch": 3.88,
      "learning_rate": 1.0574356992525403e-07,
      "loss": 0.8411,
      "step": 2645
    },
    {
      "epoch": 3.88,
      "learning_rate": 1.003912410287744e-07,
      "loss": 0.8772,
      "step": 2646
    },
    {
      "epoch": 3.88,
      "learning_rate": 9.517784181422019e-08,
      "loss": 0.9134,
      "step": 2647
    },
    {
      "epoch": 3.89,
      "learning_rate": 9.010337953185843e-08,
      "loss": 0.8586,
      "step": 2648
    },
    {
      "epoch": 3.89,
      "learning_rate": 8.516786123867748e-08,
      "loss": 0.8817,
      "step": 2649
    },
    {
      "epoch": 3.89,
      "learning_rate": 8.037129379847574e-08,
      "loss": 0.8549,
      "step": 2650
    },
    {
      "epoch": 3.89,
      "learning_rate": 7.571368388181732e-08,
      "loss": 0.8673,
      "step": 2651
    },
    {
      "epoch": 3.89,
      "learning_rate": 7.119503796599868e-08,
      "loss": 0.8912,
      "step": 2652
    },
    {
      "epoch": 3.89,
      "learning_rate": 6.68153623350598e-08,
      "loss": 0.9305,
      "step": 2653
    },
    {
      "epoch": 3.89,
      "learning_rate": 6.25746630798063e-08,
      "loss": 0.9182,
      "step": 2654
    },
    {
      "epoch": 3.9,
      "learning_rate": 5.847294609773179e-08,
      "loss": 0.8962,
      "step": 2655
    },
    {
      "epoch": 3.9,
      "learning_rate": 5.451021709307336e-08,
      "loss": 0.9156,
      "step": 2656
    },
    {
      "epoch": 3.9,
      "learning_rate": 5.068648157675604e-08,
      "loss": 0.8457,
      "step": 2657
    },
    {
      "epoch": 3.9,
      "learning_rate": 4.700174486644837e-08,
      "loss": 0.9179,
      "step": 2658
    },
    {
      "epoch": 3.9,
      "learning_rate": 4.3456012086462436e-08,
      "loss": 0.8552,
      "step": 2659
    },
    {
      "epoch": 3.9,
      "learning_rate": 4.0049288167842705e-08,
      "loss": 0.9144,
      "step": 2660
    },
    {
      "epoch": 3.9,
      "learning_rate": 3.6781577848277185e-08,
      "loss": 0.8938,
      "step": 2661
    },
    {
      "epoch": 3.91,
      "learning_rate": 3.365288567216407e-08,
      "loss": 0.8687,
      "step": 2662
    },
    {
      "epoch": 3.91,
      "learning_rate": 3.0663215990534014e-08,
      "loss": 0.9071,
      "step": 2663
    },
    {
      "epoch": 3.91,
      "learning_rate": 2.7812572961127824e-08,
      "loss": 0.7512,
      "step": 2664
    },
    {
      "epoch": 3.91,
      "learning_rate": 2.510096054829658e-08,
      "loss": 0.889,
      "step": 2665
    },
    {
      "epoch": 3.91,
      "learning_rate": 2.2528382523057113e-08,
      "loss": 0.8895,
      "step": 2666
    },
    {
      "epoch": 3.91,
      "learning_rate": 2.0094842463092012e-08,
      "loss": 0.92,
      "step": 2667
    },
    {
      "epoch": 3.92,
      "learning_rate": 1.7800343752683023e-08,
      "loss": 0.9548,
      "step": 2668
    },
    {
      "epoch": 3.92,
      "learning_rate": 1.564488958279986e-08,
      "loss": 0.8678,
      "step": 2669
    },
    {
      "epoch": 3.92,
      "learning_rate": 1.3628482951000277e-08,
      "loss": 0.8236,
      "step": 2670
    },
    {
      "epoch": 3.92,
      "learning_rate": 1.1751126661496692e-08,
      "loss": 0.8774,
      "step": 2671
    },
    {
      "epoch": 3.92,
      "learning_rate": 1.0012823325111776e-08,
      "loss": 0.9323,
      "step": 2672
    },
    {
      "epoch": 3.92,
      "learning_rate": 8.413575359289549e-09,
      "loss": 0.8945,
      "step": 2673
    },
    {
      "epoch": 3.92,
      "learning_rate": 6.953384988095391e-09,
      "loss": 0.8771,
      "step": 2674
    },
    {
      "epoch": 3.93,
      "learning_rate": 5.6322542422049266e-09,
      "loss": 0.8803,
      "step": 2675
    },
    {
      "epoch": 3.93,
      "learning_rate": 4.4501849589040355e-09,
      "loss": 0.8569,
      "step": 2676
    },
    {
      "epoch": 3.93,
      "learning_rate": 3.407178782088849e-09,
      "loss": 0.8887,
      "step": 2677
    },
    {
      "epoch": 3.93,
      "learning_rate": 2.5032371622546457e-09,
      "loss": 0.8661,
      "step": 2678
    },
    {
      "epoch": 3.93,
      "learning_rate": 1.7383613565291612e-09,
      "loss": 0.937,
      "step": 2679
    },
    {
      "epoch": 3.93,
      "learning_rate": 1.1125524285948707e-09,
      "loss": 0.8214,
      "step": 2680
    },
    {
      "epoch": 3.93,
      "learning_rate": 6.258112487667056e-10,
      "loss": 0.9334,
      "step": 2681
    },
    {
      "epoch": 3.94,
      "learning_rate": 2.781384939476439e-10,
      "loss": 0.8318,
      "step": 2682
    },
    {
      "epoch": 3.94,
      "learning_rate": 6.953464766201734e-11,
      "loss": 0.8687,
      "step": 2683
    },
    {
      "epoch": 3.94,
      "learning_rate": 0.0,
      "loss": 0.9612,
      "step": 2684
    }
  ],
  "logging_steps": 1,
  "max_steps": 2684,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 671,
  "total_flos": 8.33944645175083e+17,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}