{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.9935205183585313,
  "eval_steps": 6,
  "global_step": 115,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 2e-05,
      "loss": 0.1557,
      "step": 1
    },
    {
      "epoch": 0.01,
      "eval_loss": 0.1552305519580841,
      "eval_runtime": 32.8038,
      "eval_samples_per_second": 8.536,
      "eval_steps_per_second": 2.134,
      "step": 1
    },
    {
      "epoch": 0.02,
      "learning_rate": 4e-05,
      "loss": 0.1617,
      "step": 2
    },
    {
      "epoch": 0.03,
      "learning_rate": 6e-05,
      "loss": 0.1577,
      "step": 3
    },
    {
      "epoch": 0.03,
      "learning_rate": 8e-05,
      "loss": 0.1411,
      "step": 4
    },
    {
      "epoch": 0.04,
      "learning_rate": 0.0001,
      "loss": 0.1191,
      "step": 5
    },
    {
      "epoch": 0.05,
      "learning_rate": 0.00012,
      "loss": 0.0936,
      "step": 6
    },
    {
      "epoch": 0.05,
      "eval_loss": 0.06945057213306427,
      "eval_runtime": 32.8918,
      "eval_samples_per_second": 8.513,
      "eval_steps_per_second": 2.128,
      "step": 6
    },
    {
      "epoch": 0.06,
      "learning_rate": 0.00014,
      "loss": 0.0662,
      "step": 7
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.00016,
      "loss": 0.0513,
      "step": 8
    },
    {
      "epoch": 0.08,
      "learning_rate": 0.00018,
      "loss": 0.0521,
      "step": 9
    },
    {
      "epoch": 0.09,
      "learning_rate": 0.0002,
      "loss": 0.0496,
      "step": 10
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.00019995524322835034,
      "loss": 0.0537,
      "step": 11
    },
    {
      "epoch": 0.1,
      "learning_rate": 0.0001998210129767735,
      "loss": 0.0447,
      "step": 12
    },
    {
      "epoch": 0.1,
      "eval_loss": 0.041263870894908905,
      "eval_runtime": 32.8403,
      "eval_samples_per_second": 8.526,
      "eval_steps_per_second": 2.132,
      "step": 12
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.00019959742939952392,
      "loss": 0.0405,
      "step": 13
    },
    {
      "epoch": 0.12,
      "learning_rate": 0.00019928469263418374,
      "loss": 0.0371,
      "step": 14
    },
    {
      "epoch": 0.13,
      "learning_rate": 0.00019888308262251285,
      "loss": 0.0365,
      "step": 15
    },
    {
      "epoch": 0.14,
      "learning_rate": 0.00019839295885986296,
      "loss": 0.0396,
      "step": 16
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00019781476007338058,
      "loss": 0.0415,
      "step": 17
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.00019714900382928675,
      "loss": 0.0347,
      "step": 18
    },
    {
      "epoch": 0.16,
      "eval_loss": 0.03573257848620415,
      "eval_runtime": 32.8461,
      "eval_samples_per_second": 8.525,
      "eval_steps_per_second": 2.131,
      "step": 18
    },
    {
      "epoch": 0.16,
      "learning_rate": 0.00019639628606958533,
      "loss": 0.0319,
      "step": 19
    },
    {
      "epoch": 0.17,
      "learning_rate": 0.0001955572805786141,
      "loss": 0.0339,
      "step": 20
    },
    {
      "epoch": 0.18,
      "learning_rate": 0.00019463273837991643,
      "loss": 0.0313,
      "step": 21
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00019362348706397373,
      "loss": 0.0356,
      "step": 22
    },
    {
      "epoch": 0.2,
      "learning_rate": 0.00019253043004739968,
      "loss": 0.0347,
      "step": 23
    },
    {
      "epoch": 0.21,
      "learning_rate": 0.0001913545457642601,
      "loss": 0.0314,
      "step": 24
    },
    {
      "epoch": 0.21,
      "eval_loss": 0.03238712623715401,
      "eval_runtime": 32.8447,
      "eval_samples_per_second": 8.525,
      "eval_steps_per_second": 2.131,
      "step": 24
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0001900968867902419,
      "loss": 0.0309,
      "step": 25
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.00018875857890045543,
      "loss": 0.033,
      "step": 26
    },
    {
      "epoch": 0.23,
      "learning_rate": 0.00018734082006171299,
      "loss": 0.0331,
      "step": 27
    },
    {
      "epoch": 0.24,
      "learning_rate": 0.00018584487936018661,
      "loss": 0.0304,
      "step": 28
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0001842720958654039,
      "loss": 0.0306,
      "step": 29
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.0001826238774315995,
      "loss": 0.0306,
      "step": 30
    },
    {
      "epoch": 0.26,
      "eval_loss": 0.030900534242391586,
      "eval_runtime": 32.8989,
      "eval_samples_per_second": 8.511,
      "eval_steps_per_second": 2.128,
      "step": 30
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.00018090169943749476,
      "loss": 0.0299,
      "step": 31
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.00017910710346563416,
      "loss": 0.0307,
      "step": 32
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00017724169592245995,
      "loss": 0.0294,
      "step": 33
    },
    {
      "epoch": 0.29,
      "learning_rate": 0.00017530714660036112,
      "loss": 0.0283,
      "step": 34
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00017330518718298264,
      "loss": 0.0328,
      "step": 35
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.0001712376096951345,
      "loss": 0.0276,
      "step": 36
    },
    {
      "epoch": 0.31,
      "eval_loss": 0.02941257879137993,
      "eval_runtime": 32.8187,
      "eval_samples_per_second": 8.532,
      "eval_steps_per_second": 2.133,
      "step": 36
    },
    {
      "epoch": 0.32,
      "learning_rate": 0.00016910626489868649,
      "loss": 0.0267,
      "step": 37
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00016691306063588583,
      "loss": 0.0293,
      "step": 38
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00016465996012157995,
      "loss": 0.0259,
      "step": 39
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00016234898018587337,
      "loss": 0.0286,
      "step": 40
    },
    {
      "epoch": 0.35,
      "learning_rate": 0.00015998218946879138,
      "loss": 0.0255,
      "step": 41
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00015756170656856737,
      "loss": 0.028,
      "step": 42
    },
    {
      "epoch": 0.36,
      "eval_loss": 0.028368661180138588,
      "eval_runtime": 32.9402,
      "eval_samples_per_second": 8.5,
      "eval_steps_per_second": 2.125,
      "step": 42
    },
    {
      "epoch": 0.37,
      "learning_rate": 0.00015508969814521025,
      "loss": 0.028,
      "step": 43
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.00015256837698105047,
      "loss": 0.0279,
      "step": 44
    },
    {
      "epoch": 0.39,
      "learning_rate": 0.00015000000000000001,
      "loss": 0.0255,
      "step": 45
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.00014738686624729986,
      "loss": 0.0283,
      "step": 46
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00014473131483156327,
      "loss": 0.0257,
      "step": 47
    },
    {
      "epoch": 0.41,
      "learning_rate": 0.00014203572283095657,
      "loss": 0.0307,
      "step": 48
    },
    {
      "epoch": 0.41,
      "eval_loss": 0.028058379888534546,
      "eval_runtime": 32.8234,
      "eval_samples_per_second": 8.531,
      "eval_steps_per_second": 2.133,
      "step": 48
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.00013930250316539238,
      "loss": 0.0292,
      "step": 49
    },
    {
      "epoch": 0.43,
      "learning_rate": 0.00013653410243663952,
      "loss": 0.0267,
      "step": 50
    },
    {
      "epoch": 0.44,
      "learning_rate": 0.00013373299873828303,
      "loss": 0.0277,
      "step": 51
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.00013090169943749476,
      "loss": 0.0285,
      "step": 52
    },
    {
      "epoch": 0.46,
      "learning_rate": 0.00012804273893060028,
      "loss": 0.0267,
      "step": 53
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00012515867637445086,
      "loss": 0.0276,
      "step": 54
    },
    {
      "epoch": 0.47,
      "eval_loss": 0.027369534596800804,
      "eval_runtime": 32.8389,
      "eval_samples_per_second": 8.526,
      "eval_steps_per_second": 2.132,
      "step": 54
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00012225209339563145,
      "loss": 0.0288,
      "step": 55
    },
    {
      "epoch": 0.48,
      "learning_rate": 0.00011932559177955533,
      "loss": 0.028,
      "step": 56
    },
    {
      "epoch": 0.49,
      "learning_rate": 0.00011638179114151377,
      "loss": 0.0264,
      "step": 57
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.00011342332658176555,
      "loss": 0.0248,
      "step": 58
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.00011045284632676536,
      "loss": 0.0261,
      "step": 59
    },
    {
      "epoch": 0.52,
      "learning_rate": 0.00010747300935864243,
      "loss": 0.0251,
      "step": 60
    },
    {
      "epoch": 0.52,
      "eval_loss": 0.026654452085494995,
      "eval_runtime": 32.8756,
      "eval_samples_per_second": 8.517,
      "eval_steps_per_second": 2.129,
      "step": 60
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.00010448648303505151,
      "loss": 0.028,
      "step": 61
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.00010149594070152638,
      "loss": 0.0274,
      "step": 62
    },
    {
      "epoch": 0.54,
      "learning_rate": 9.850405929847366e-05,
      "loss": 0.0252,
      "step": 63
    },
    {
      "epoch": 0.55,
      "learning_rate": 9.551351696494854e-05,
      "loss": 0.0277,
      "step": 64
    },
    {
      "epoch": 0.56,
      "learning_rate": 9.252699064135758e-05,
      "loss": 0.0285,
      "step": 65
    },
    {
      "epoch": 0.57,
      "learning_rate": 8.954715367323468e-05,
      "loss": 0.0244,
      "step": 66
    },
    {
      "epoch": 0.57,
      "eval_loss": 0.02685328759253025,
      "eval_runtime": 32.8937,
      "eval_samples_per_second": 8.512,
      "eval_steps_per_second": 2.128,
      "step": 66
    },
    {
      "epoch": 0.58,
      "learning_rate": 8.657667341823448e-05,
      "loss": 0.0241,
      "step": 67
    },
    {
      "epoch": 0.59,
      "learning_rate": 8.361820885848624e-05,
      "loss": 0.0242,
      "step": 68
    },
    {
      "epoch": 0.6,
      "learning_rate": 8.067440822044469e-05,
      "loss": 0.0253,
      "step": 69
    },
    {
      "epoch": 0.6,
      "learning_rate": 7.774790660436858e-05,
      "loss": 0.0286,
      "step": 70
    },
    {
      "epoch": 0.61,
      "learning_rate": 7.484132362554915e-05,
      "loss": 0.0259,
      "step": 71
    },
    {
      "epoch": 0.62,
      "learning_rate": 7.195726106939974e-05,
      "loss": 0.0268,
      "step": 72
    },
    {
      "epoch": 0.62,
      "eval_loss": 0.026269590482115746,
      "eval_runtime": 32.8623,
      "eval_samples_per_second": 8.52,
      "eval_steps_per_second": 2.13,
      "step": 72
    },
    {
      "epoch": 0.63,
      "learning_rate": 6.909830056250527e-05,
      "loss": 0.0274,
      "step": 73
    },
    {
      "epoch": 0.64,
      "learning_rate": 6.626700126171702e-05,
      "loss": 0.0245,
      "step": 74
    },
    {
      "epoch": 0.65,
      "learning_rate": 6.34658975633605e-05,
      "loss": 0.0251,
      "step": 75
    },
    {
      "epoch": 0.66,
      "learning_rate": 6.069749683460765e-05,
      "loss": 0.0271,
      "step": 76
    },
    {
      "epoch": 0.67,
      "learning_rate": 5.796427716904347e-05,
      "loss": 0.0255,
      "step": 77
    },
    {
      "epoch": 0.67,
      "learning_rate": 5.526868516843673e-05,
      "loss": 0.0249,
      "step": 78
    },
    {
      "epoch": 0.67,
      "eval_loss": 0.026168059557676315,
      "eval_runtime": 32.8573,
      "eval_samples_per_second": 8.522,
      "eval_steps_per_second": 2.13,
      "step": 78
    },
    {
      "epoch": 0.68,
      "learning_rate": 5.261313375270014e-05,
      "loss": 0.027,
      "step": 79
    },
    {
      "epoch": 0.69,
      "learning_rate": 5.000000000000002e-05,
      "loss": 0.0241,
      "step": 80
    },
    {
      "epoch": 0.7,
      "learning_rate": 4.743162301894952e-05,
      "loss": 0.0245,
      "step": 81
    },
    {
      "epoch": 0.71,
      "learning_rate": 4.491030185478976e-05,
      "loss": 0.0233,
      "step": 82
    },
    {
      "epoch": 0.72,
      "learning_rate": 4.2438293431432665e-05,
      "loss": 0.0261,
      "step": 83
    },
    {
      "epoch": 0.73,
      "learning_rate": 4.001781053120863e-05,
      "loss": 0.0252,
      "step": 84
    },
    {
      "epoch": 0.73,
      "eval_loss": 0.02580336295068264,
      "eval_runtime": 32.827,
      "eval_samples_per_second": 8.53,
      "eval_steps_per_second": 2.132,
      "step": 84
    },
    {
      "epoch": 0.73,
      "learning_rate": 3.7651019814126654e-05,
      "loss": 0.0246,
      "step": 85
    },
    {
      "epoch": 0.74,
      "learning_rate": 3.534003987842005e-05,
      "loss": 0.0229,
      "step": 86
    },
    {
      "epoch": 0.75,
      "learning_rate": 3.308693936411421e-05,
      "loss": 0.025,
      "step": 87
    },
    {
      "epoch": 0.76,
      "learning_rate": 3.089373510131354e-05,
      "loss": 0.0247,
      "step": 88
    },
    {
      "epoch": 0.77,
      "learning_rate": 2.876239030486554e-05,
      "loss": 0.0251,
      "step": 89
    },
    {
      "epoch": 0.78,
      "learning_rate": 2.669481281701739e-05,
      "loss": 0.0259,
      "step": 90
    },
    {
      "epoch": 0.78,
      "eval_loss": 0.02568827196955681,
      "eval_runtime": 32.8573,
      "eval_samples_per_second": 8.522,
      "eval_steps_per_second": 2.13,
      "step": 90
    },
    {
      "epoch": 0.79,
      "learning_rate": 2.4692853399638917e-05,
      "loss": 0.0283,
      "step": 91
    },
    {
      "epoch": 0.79,
      "learning_rate": 2.275830407754006e-05,
      "loss": 0.0245,
      "step": 92
    },
    {
      "epoch": 0.8,
      "learning_rate": 2.0892896534365904e-05,
      "loss": 0.0308,
      "step": 93
    },
    {
      "epoch": 0.81,
      "learning_rate": 1.9098300562505266e-05,
      "loss": 0.0239,
      "step": 94
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.7376122568400532e-05,
      "loss": 0.0238,
      "step": 95
    },
    {
      "epoch": 0.83,
      "learning_rate": 1.5727904134596083e-05,
      "loss": 0.0241,
      "step": 96
    },
    {
      "epoch": 0.83,
      "eval_loss": 0.025526922196149826,
      "eval_runtime": 32.8676,
      "eval_samples_per_second": 8.519,
      "eval_steps_per_second": 2.13,
      "step": 96
    },
    {
      "epoch": 0.84,
      "learning_rate": 1.415512063981339e-05,
      "loss": 0.0261,
      "step": 97
    },
    {
      "epoch": 0.85,
      "learning_rate": 1.2659179938287035e-05,
      "loss": 0.0246,
      "step": 98
    },
    {
      "epoch": 0.86,
      "learning_rate": 1.124142109954459e-05,
      "loss": 0.0295,
      "step": 99
    },
    {
      "epoch": 0.86,
      "learning_rate": 9.903113209758096e-06,
      "loss": 0.025,
      "step": 100
    },
    {
      "epoch": 0.87,
      "learning_rate": 8.645454235739903e-06,
      "loss": 0.0235,
      "step": 101
    },
    {
      "epoch": 0.88,
      "learning_rate": 7.46956995260033e-06,
      "loss": 0.0241,
      "step": 102
    },
    {
      "epoch": 0.88,
      "eval_loss": 0.025442615151405334,
      "eval_runtime": 32.9063,
      "eval_samples_per_second": 8.509,
      "eval_steps_per_second": 2.127,
      "step": 102
    },
    {
      "epoch": 0.89,
      "learning_rate": 6.37651293602628e-06,
      "loss": 0.0237,
      "step": 103
    },
    {
      "epoch": 0.9,
      "learning_rate": 5.367261620083575e-06,
      "loss": 0.0243,
      "step": 104
    },
    {
      "epoch": 0.91,
      "learning_rate": 4.442719421385922e-06,
      "loss": 0.0241,
      "step": 105
    },
    {
      "epoch": 0.92,
      "learning_rate": 3.6037139304146762e-06,
      "loss": 0.0237,
      "step": 106
    },
    {
      "epoch": 0.92,
      "learning_rate": 2.8509961707132494e-06,
      "loss": 0.0251,
      "step": 107
    },
    {
      "epoch": 0.93,
      "learning_rate": 2.1852399266194314e-06,
      "loss": 0.0253,
      "step": 108
    },
    {
      "epoch": 0.93,
      "eval_loss": 0.025354694575071335,
      "eval_runtime": 32.8944,
      "eval_samples_per_second": 8.512,
      "eval_steps_per_second": 2.128,
      "step": 108
    },
    {
      "epoch": 0.94,
      "learning_rate": 1.6070411401370334e-06,
      "loss": 0.024,
      "step": 109
    },
    {
      "epoch": 0.95,
      "learning_rate": 1.1169173774871478e-06,
      "loss": 0.0238,
      "step": 110
    },
    {
      "epoch": 0.96,
      "learning_rate": 7.153073658162646e-07,
      "loss": 0.0232,
      "step": 111
    },
    {
      "epoch": 0.97,
      "learning_rate": 4.025706004760932e-07,
      "loss": 0.0243,
      "step": 112
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.7898702322648453e-07,
      "loss": 0.0256,
      "step": 113
    },
    {
      "epoch": 0.98,
      "learning_rate": 4.475677164966774e-08,
      "loss": 0.0234,
      "step": 114
    },
    {
      "epoch": 0.98,
      "eval_loss": 0.02534579671919346,
      "eval_runtime": 32.8785,
      "eval_samples_per_second": 8.516,
      "eval_steps_per_second": 2.129,
      "step": 114
    },
    {
      "epoch": 0.99,
      "learning_rate": 0.0,
      "loss": 0.0243,
      "step": 115
    }
  ],
  "logging_steps": 1,
  "max_steps": 115,
  "num_train_epochs": 1,
  "save_steps": 500,
  "total_flos": 1.3013449078053274e+18,
  "trial_name": null,
  "trial_params": null
}