{
  "best_metric": 0.03415210172533989,
  "best_model_checkpoint": "AlexWang99/byt5_3k_4d/checkpoint-376",
  "epoch": 94.0,
  "eval_steps": 500,
  "global_step": 376,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_loss": 0.47024813294410706,
      "eval_runtime": 10.75,
      "eval_samples_per_second": 930.235,
      "eval_steps_per_second": 1.209,
      "step": 4
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.4366174340248108,
      "eval_runtime": 10.8447,
      "eval_samples_per_second": 922.113,
      "eval_steps_per_second": 1.199,
      "step": 8
    },
    {
      "epoch": 2.5,
      "learning_rate": 4.875e-05,
      "loss": 0.8301,
      "step": 10
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.4202331006526947,
      "eval_runtime": 10.7807,
      "eval_samples_per_second": 927.582,
      "eval_steps_per_second": 1.206,
      "step": 12
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.38434356451034546,
      "eval_runtime": 10.8022,
      "eval_samples_per_second": 925.735,
      "eval_steps_per_second": 1.203,
      "step": 16
    },
    {
      "epoch": 5.0,
      "learning_rate": 4.75e-05,
      "loss": 0.7703,
      "step": 20
    },
    {
      "epoch": 5.0,
      "eval_loss": 0.38975098729133606,
      "eval_runtime": 10.7363,
      "eval_samples_per_second": 931.418,
      "eval_steps_per_second": 1.211,
      "step": 20
    },
    {
      "epoch": 6.0,
      "eval_loss": 0.35734379291534424,
      "eval_runtime": 10.725,
      "eval_samples_per_second": 932.399,
      "eval_steps_per_second": 1.212,
      "step": 24
    },
    {
      "epoch": 7.0,
      "eval_loss": 0.34219321608543396,
      "eval_runtime": 10.9668,
      "eval_samples_per_second": 911.84,
      "eval_steps_per_second": 1.185,
      "step": 28
    },
    {
      "epoch": 7.5,
      "learning_rate": 4.6250000000000006e-05,
      "loss": 0.7169,
      "step": 30
    },
    {
      "epoch": 8.0,
      "eval_loss": 0.330382376909256,
      "eval_runtime": 10.8312,
      "eval_samples_per_second": 923.257,
      "eval_steps_per_second": 1.2,
      "step": 32
    },
    {
      "epoch": 9.0,
      "eval_loss": 0.3048989772796631,
      "eval_runtime": 10.726,
      "eval_samples_per_second": 932.316,
      "eval_steps_per_second": 1.212,
      "step": 36
    },
    {
      "epoch": 10.0,
      "learning_rate": 4.5e-05,
      "loss": 0.6727,
      "step": 40
    },
    {
      "epoch": 10.0,
      "eval_loss": 0.30665355920791626,
      "eval_runtime": 10.9181,
      "eval_samples_per_second": 915.912,
      "eval_steps_per_second": 1.191,
      "step": 40
    },
    {
      "epoch": 11.0,
      "eval_loss": 0.29648929834365845,
      "eval_runtime": 10.7354,
      "eval_samples_per_second": 931.494,
      "eval_steps_per_second": 1.211,
      "step": 44
    },
    {
      "epoch": 12.0,
      "eval_loss": 0.26932698488235474,
      "eval_runtime": 10.8379,
      "eval_samples_per_second": 922.691,
      "eval_steps_per_second": 1.199,
      "step": 48
    },
    {
      "epoch": 12.5,
      "learning_rate": 4.375e-05,
      "loss": 0.6394,
      "step": 50
    },
    {
      "epoch": 13.0,
      "eval_loss": 0.271121084690094,
      "eval_runtime": 10.8381,
      "eval_samples_per_second": 922.67,
      "eval_steps_per_second": 1.199,
      "step": 52
    },
    {
      "epoch": 14.0,
      "eval_loss": 0.25609534978866577,
      "eval_runtime": 10.7523,
      "eval_samples_per_second": 930.029,
      "eval_steps_per_second": 1.209,
      "step": 56
    },
    {
      "epoch": 15.0,
      "learning_rate": 4.25e-05,
      "loss": 0.6047,
      "step": 60
    },
    {
      "epoch": 15.0,
      "eval_loss": 0.2453787624835968,
      "eval_runtime": 10.9268,
      "eval_samples_per_second": 915.184,
      "eval_steps_per_second": 1.19,
      "step": 60
    },
    {
      "epoch": 16.0,
      "eval_loss": 0.23679418861865997,
      "eval_runtime": 10.7591,
      "eval_samples_per_second": 929.449,
      "eval_steps_per_second": 1.208,
      "step": 64
    },
    {
      "epoch": 17.0,
      "eval_loss": 0.22503486275672913,
      "eval_runtime": 10.8371,
      "eval_samples_per_second": 922.76,
      "eval_steps_per_second": 1.2,
      "step": 68
    },
    {
      "epoch": 17.5,
      "learning_rate": 4.125e-05,
      "loss": 0.565,
      "step": 70
    },
    {
      "epoch": 18.0,
      "eval_loss": 0.21100571751594543,
      "eval_runtime": 10.8584,
      "eval_samples_per_second": 920.947,
      "eval_steps_per_second": 1.197,
      "step": 72
    },
    {
      "epoch": 19.0,
      "eval_loss": 0.21086934208869934,
      "eval_runtime": 10.746,
      "eval_samples_per_second": 930.581,
      "eval_steps_per_second": 1.21,
      "step": 76
    },
    {
      "epoch": 20.0,
      "learning_rate": 4e-05,
      "loss": 0.5368,
      "step": 80
    },
    {
      "epoch": 20.0,
      "eval_loss": 0.1950305849313736,
      "eval_runtime": 10.7833,
      "eval_samples_per_second": 927.356,
      "eval_steps_per_second": 1.206,
      "step": 80
    },
    {
      "epoch": 21.0,
      "eval_loss": 0.19742541015148163,
      "eval_runtime": 10.7496,
      "eval_samples_per_second": 930.268,
      "eval_steps_per_second": 1.209,
      "step": 84
    },
    {
      "epoch": 22.0,
      "eval_loss": 0.18190620839595795,
      "eval_runtime": 11.0063,
      "eval_samples_per_second": 908.569,
      "eval_steps_per_second": 1.181,
      "step": 88
    },
    {
      "epoch": 22.5,
      "learning_rate": 3.875e-05,
      "loss": 0.518,
      "step": 90
    },
    {
      "epoch": 23.0,
      "eval_loss": 0.1795072704553604,
      "eval_runtime": 10.855,
      "eval_samples_per_second": 921.235,
      "eval_steps_per_second": 1.198,
      "step": 92
    },
    {
      "epoch": 24.0,
      "eval_loss": 0.16476453840732574,
      "eval_runtime": 10.7846,
      "eval_samples_per_second": 927.252,
      "eval_steps_per_second": 1.205,
      "step": 96
    },
    {
      "epoch": 25.0,
      "learning_rate": 3.7500000000000003e-05,
      "loss": 0.4862,
      "step": 100
    },
    {
      "epoch": 25.0,
      "eval_loss": 0.16748683154582977,
      "eval_runtime": 10.9227,
      "eval_samples_per_second": 915.525,
      "eval_steps_per_second": 1.19,
      "step": 100
    },
    {
      "epoch": 26.0,
      "eval_loss": 0.15504491329193115,
      "eval_runtime": 10.7793,
      "eval_samples_per_second": 927.702,
      "eval_steps_per_second": 1.206,
      "step": 104
    },
    {
      "epoch": 27.0,
      "eval_loss": 0.15300293266773224,
      "eval_runtime": 10.8547,
      "eval_samples_per_second": 921.263,
      "eval_steps_per_second": 1.198,
      "step": 108
    },
    {
      "epoch": 27.5,
      "learning_rate": 3.625e-05,
      "loss": 0.4628,
      "step": 110
    },
    {
      "epoch": 28.0,
      "eval_loss": 0.14466118812561035,
      "eval_runtime": 10.8549,
      "eval_samples_per_second": 921.246,
      "eval_steps_per_second": 1.198,
      "step": 112
    },
    {
      "epoch": 29.0,
      "eval_loss": 0.1442325860261917,
      "eval_runtime": 10.764,
      "eval_samples_per_second": 929.023,
      "eval_steps_per_second": 1.208,
      "step": 116
    },
    {
      "epoch": 30.0,
      "learning_rate": 3.5e-05,
      "loss": 0.4408,
      "step": 120
    },
    {
      "epoch": 30.0,
      "eval_loss": 0.13103845715522766,
      "eval_runtime": 10.9404,
      "eval_samples_per_second": 914.043,
      "eval_steps_per_second": 1.188,
      "step": 120
    },
    {
      "epoch": 31.0,
      "eval_loss": 0.13363589346408844,
      "eval_runtime": 10.7472,
      "eval_samples_per_second": 930.475,
      "eval_steps_per_second": 1.21,
      "step": 124
    },
    {
      "epoch": 32.0,
      "eval_loss": 0.12347704917192459,
      "eval_runtime": 10.8696,
      "eval_samples_per_second": 920.0,
      "eval_steps_per_second": 1.196,
      "step": 128
    },
    {
      "epoch": 32.5,
      "learning_rate": 3.375000000000001e-05,
      "loss": 0.4192,
      "step": 130
    },
    {
      "epoch": 33.0,
      "eval_loss": 0.1190723404288292,
      "eval_runtime": 10.8564,
      "eval_samples_per_second": 921.113,
      "eval_steps_per_second": 1.197,
      "step": 132
    },
    {
      "epoch": 34.0,
      "eval_loss": 0.11926523596048355,
      "eval_runtime": 10.7486,
      "eval_samples_per_second": 930.351,
      "eval_steps_per_second": 1.209,
      "step": 136
    },
    {
      "epoch": 35.0,
      "learning_rate": 3.2500000000000004e-05,
      "loss": 0.4133,
      "step": 140
    },
    {
      "epoch": 35.0,
      "eval_loss": 0.1123439222574234,
      "eval_runtime": 10.9392,
      "eval_samples_per_second": 914.14,
      "eval_steps_per_second": 1.188,
      "step": 140
    },
    {
      "epoch": 36.0,
      "eval_loss": 0.11556507647037506,
      "eval_runtime": 10.7494,
      "eval_samples_per_second": 930.287,
      "eval_steps_per_second": 1.209,
      "step": 144
    },
    {
      "epoch": 37.0,
      "eval_loss": 0.10512395203113556,
      "eval_runtime": 10.8606,
      "eval_samples_per_second": 920.755,
      "eval_steps_per_second": 1.197,
      "step": 148
    },
    {
      "epoch": 37.5,
      "learning_rate": 3.125e-05,
      "loss": 0.3922,
      "step": 150
    },
    {
      "epoch": 38.0,
      "eval_loss": 0.09992647171020508,
      "eval_runtime": 10.8473,
      "eval_samples_per_second": 921.889,
      "eval_steps_per_second": 1.198,
      "step": 152
    },
    {
      "epoch": 39.0,
      "eval_loss": 0.09912022948265076,
      "eval_runtime": 10.751,
      "eval_samples_per_second": 930.148,
      "eval_steps_per_second": 1.209,
      "step": 156
    },
    {
      "epoch": 40.0,
      "learning_rate": 3e-05,
      "loss": 0.3778,
      "step": 160
    },
    {
      "epoch": 40.0,
      "eval_loss": 0.09952548891305923,
      "eval_runtime": 10.9239,
      "eval_samples_per_second": 915.427,
      "eval_steps_per_second": 1.19,
      "step": 160
    },
    {
      "epoch": 41.0,
      "eval_loss": 0.09118303656578064,
      "eval_runtime": 10.7525,
      "eval_samples_per_second": 930.014,
      "eval_steps_per_second": 1.209,
      "step": 164
    },
    {
      "epoch": 42.0,
      "eval_loss": 0.09032606333494186,
      "eval_runtime": 10.8423,
      "eval_samples_per_second": 922.31,
      "eval_steps_per_second": 1.199,
      "step": 168
    },
    {
      "epoch": 42.5,
      "learning_rate": 2.8749999999999997e-05,
      "loss": 0.3655,
      "step": 170
    },
    {
      "epoch": 43.0,
      "eval_loss": 0.08409227430820465,
      "eval_runtime": 10.8572,
      "eval_samples_per_second": 921.048,
      "eval_steps_per_second": 1.197,
      "step": 172
    },
    {
      "epoch": 44.0,
      "eval_loss": 0.0789598897099495,
      "eval_runtime": 10.7437,
      "eval_samples_per_second": 930.776,
      "eval_steps_per_second": 1.21,
      "step": 176
    },
    {
      "epoch": 45.0,
      "learning_rate": 2.7500000000000004e-05,
      "loss": 0.3526,
      "step": 180
    },
    {
      "epoch": 45.0,
      "eval_loss": 0.08267370611429214,
      "eval_runtime": 10.9337,
      "eval_samples_per_second": 914.6,
      "eval_steps_per_second": 1.189,
      "step": 180
    },
    {
      "epoch": 46.0,
      "eval_loss": 0.07559242099523544,
      "eval_runtime": 10.7359,
      "eval_samples_per_second": 931.456,
      "eval_steps_per_second": 1.211,
      "step": 184
    },
    {
      "epoch": 47.0,
      "eval_loss": 0.07468590885400772,
      "eval_runtime": 10.8561,
      "eval_samples_per_second": 921.142,
      "eval_steps_per_second": 1.197,
      "step": 188
    },
    {
      "epoch": 47.5,
      "learning_rate": 2.625e-05,
      "loss": 0.3378,
      "step": 190
    },
    {
      "epoch": 48.0,
      "eval_loss": 0.0737282931804657,
      "eval_runtime": 10.8372,
      "eval_samples_per_second": 922.747,
      "eval_steps_per_second": 1.2,
      "step": 192
    },
    {
      "epoch": 49.0,
      "eval_loss": 0.07465548813343048,
      "eval_runtime": 10.7608,
      "eval_samples_per_second": 929.3,
      "eval_steps_per_second": 1.208,
      "step": 196
    },
    {
      "epoch": 50.0,
      "learning_rate": 2.5e-05,
      "loss": 0.3308,
      "step": 200
    },
    {
      "epoch": 50.0,
      "eval_loss": 0.07294411957263947,
      "eval_runtime": 10.9347,
      "eval_samples_per_second": 914.518,
      "eval_steps_per_second": 1.189,
      "step": 200
    },
    {
      "epoch": 51.0,
      "eval_loss": 0.0665128082036972,
      "eval_runtime": 10.7481,
      "eval_samples_per_second": 930.401,
      "eval_steps_per_second": 1.21,
      "step": 204
    },
    {
      "epoch": 52.0,
      "eval_loss": 0.06627500057220459,
      "eval_runtime": 10.8409,
      "eval_samples_per_second": 922.432,
      "eval_steps_per_second": 1.199,
      "step": 208
    },
    {
      "epoch": 52.5,
      "learning_rate": 2.375e-05,
      "loss": 0.321,
      "step": 210
    },
    {
      "epoch": 53.0,
      "eval_loss": 0.06422976404428482,
      "eval_runtime": 10.8561,
      "eval_samples_per_second": 921.142,
      "eval_steps_per_second": 1.197,
      "step": 212
    },
    {
      "epoch": 54.0,
      "eval_loss": 0.06401015818119049,
      "eval_runtime": 10.7568,
      "eval_samples_per_second": 929.646,
      "eval_steps_per_second": 1.209,
      "step": 216
    },
    {
      "epoch": 55.0,
      "learning_rate": 2.25e-05,
      "loss": 0.3084,
      "step": 220
    },
    {
      "epoch": 55.0,
      "eval_loss": 0.06319215148687363,
      "eval_runtime": 10.9357,
      "eval_samples_per_second": 914.439,
      "eval_steps_per_second": 1.189,
      "step": 220
    },
    {
      "epoch": 56.0,
      "eval_loss": 0.059920959174633026,
      "eval_runtime": 10.7511,
      "eval_samples_per_second": 930.141,
      "eval_steps_per_second": 1.209,
      "step": 224
    },
    {
      "epoch": 57.0,
      "eval_loss": 0.057977043092250824,
      "eval_runtime": 10.8555,
      "eval_samples_per_second": 921.188,
      "eval_steps_per_second": 1.198,
      "step": 228
    },
    {
      "epoch": 57.5,
      "learning_rate": 2.125e-05,
      "loss": 0.2967,
      "step": 230
    },
    {
      "epoch": 58.0,
      "eval_loss": 0.05669580027461052,
      "eval_runtime": 10.8434,
      "eval_samples_per_second": 922.219,
      "eval_steps_per_second": 1.199,
      "step": 232
    },
    {
      "epoch": 59.0,
      "eval_loss": 0.0525004044175148,
      "eval_runtime": 10.7506,
      "eval_samples_per_second": 930.18,
      "eval_steps_per_second": 1.209,
      "step": 236
    },
    {
      "epoch": 60.0,
      "learning_rate": 2e-05,
      "loss": 0.2928,
      "step": 240
    },
    {
      "epoch": 60.0,
      "eval_loss": 0.05224265158176422,
      "eval_runtime": 10.9382,
      "eval_samples_per_second": 914.231,
      "eval_steps_per_second": 1.188,
      "step": 240
    },
    {
      "epoch": 61.0,
      "eval_loss": 0.05358195677399635,
      "eval_runtime": 10.7598,
      "eval_samples_per_second": 929.387,
      "eval_steps_per_second": 1.208,
      "step": 244
    },
    {
      "epoch": 62.0,
      "eval_loss": 0.052435729652643204,
      "eval_runtime": 10.8537,
      "eval_samples_per_second": 921.341,
      "eval_steps_per_second": 1.198,
      "step": 248
    },
    {
      "epoch": 62.5,
      "learning_rate": 1.8750000000000002e-05,
      "loss": 0.2929,
      "step": 250
    },
    {
      "epoch": 63.0,
      "eval_loss": 0.056764379143714905,
      "eval_runtime": 10.8657,
      "eval_samples_per_second": 920.329,
      "eval_steps_per_second": 1.196,
      "step": 252
    },
    {
      "epoch": 64.0,
      "eval_loss": 0.0530422069132328,
      "eval_runtime": 10.7635,
      "eval_samples_per_second": 929.065,
      "eval_steps_per_second": 1.208,
      "step": 256
    },
    {
      "epoch": 65.0,
      "learning_rate": 1.75e-05,
      "loss": 0.283,
      "step": 260
    },
    {
      "epoch": 65.0,
      "eval_loss": 0.04763900488615036,
      "eval_runtime": 10.9405,
      "eval_samples_per_second": 914.038,
      "eval_steps_per_second": 1.188,
      "step": 260
    },
    {
      "epoch": 66.0,
      "eval_loss": 0.04787232354283333,
      "eval_runtime": 10.7497,
      "eval_samples_per_second": 930.255,
      "eval_steps_per_second": 1.209,
      "step": 264
    },
    {
      "epoch": 67.0,
      "eval_loss": 0.05069798231124878,
      "eval_runtime": 10.8615,
      "eval_samples_per_second": 920.685,
      "eval_steps_per_second": 1.197,
      "step": 268
    },
    {
      "epoch": 67.5,
      "learning_rate": 1.6250000000000002e-05,
      "loss": 0.2766,
      "step": 270
    },
    {
      "epoch": 68.0,
      "eval_loss": 0.04605843871831894,
      "eval_runtime": 10.8475,
      "eval_samples_per_second": 921.871,
      "eval_steps_per_second": 1.198,
      "step": 272
    },
    {
      "epoch": 69.0,
      "eval_loss": 0.04439115151762962,
      "eval_runtime": 10.7586,
      "eval_samples_per_second": 929.488,
      "eval_steps_per_second": 1.208,
      "step": 276
    },
    {
      "epoch": 70.0,
      "learning_rate": 1.5e-05,
      "loss": 0.2677,
      "step": 280
    },
    {
      "epoch": 70.0,
      "eval_loss": 0.0455540269613266,
      "eval_runtime": 10.9449,
      "eval_samples_per_second": 913.667,
      "eval_steps_per_second": 1.188,
      "step": 280
    },
    {
      "epoch": 71.0,
      "eval_loss": 0.04371872544288635,
      "eval_runtime": 10.758,
      "eval_samples_per_second": 929.543,
      "eval_steps_per_second": 1.208,
      "step": 284
    },
    {
      "epoch": 72.0,
      "eval_loss": 0.04281056672334671,
      "eval_runtime": 10.8458,
      "eval_samples_per_second": 922.019,
      "eval_steps_per_second": 1.199,
      "step": 288
    },
    {
      "epoch": 72.5,
      "learning_rate": 1.3750000000000002e-05,
      "loss": 0.2614,
      "step": 290
    },
    {
      "epoch": 73.0,
      "eval_loss": 0.04188177362084389,
      "eval_runtime": 10.8487,
      "eval_samples_per_second": 921.77,
      "eval_steps_per_second": 1.198,
      "step": 292
    },
    {
      "epoch": 74.0,
      "eval_loss": 0.04140578955411911,
      "eval_runtime": 10.7507,
      "eval_samples_per_second": 930.172,
      "eval_steps_per_second": 1.209,
      "step": 296
    },
    {
      "epoch": 75.0,
      "learning_rate": 1.25e-05,
      "loss": 0.2595,
      "step": 300
    },
    {
      "epoch": 75.0,
      "eval_loss": 0.04175864905118942,
      "eval_runtime": 10.9254,
      "eval_samples_per_second": 915.297,
      "eval_steps_per_second": 1.19,
      "step": 300
    },
    {
      "epoch": 76.0,
      "eval_loss": 0.04123968258500099,
      "eval_runtime": 10.7366,
      "eval_samples_per_second": 931.392,
      "eval_steps_per_second": 1.211,
      "step": 304
    },
    {
      "epoch": 77.0,
      "eval_loss": 0.03961934149265289,
      "eval_runtime": 10.8404,
      "eval_samples_per_second": 922.477,
      "eval_steps_per_second": 1.199,
      "step": 308
    },
    {
      "epoch": 77.5,
      "learning_rate": 1.125e-05,
      "loss": 0.2582,
      "step": 310
    },
    {
      "epoch": 78.0,
      "eval_loss": 0.03820143640041351,
      "eval_runtime": 10.8282,
      "eval_samples_per_second": 923.515,
      "eval_steps_per_second": 1.201,
      "step": 312
    },
    {
      "epoch": 79.0,
      "eval_loss": 0.038101743906736374,
      "eval_runtime": 10.7523,
      "eval_samples_per_second": 930.033,
      "eval_steps_per_second": 1.209,
      "step": 316
    },
    {
      "epoch": 80.0,
      "learning_rate": 1e-05,
      "loss": 0.2511,
      "step": 320
    },
    {
      "epoch": 80.0,
      "eval_loss": 0.03866468369960785,
      "eval_runtime": 10.9151,
      "eval_samples_per_second": 916.163,
      "eval_steps_per_second": 1.191,
      "step": 320
    },
    {
      "epoch": 81.0,
      "eval_loss": 0.0387905091047287,
      "eval_runtime": 10.7459,
      "eval_samples_per_second": 930.586,
      "eval_steps_per_second": 1.21,
      "step": 324
    },
    {
      "epoch": 82.0,
      "eval_loss": 0.03723302111029625,
      "eval_runtime": 10.8332,
      "eval_samples_per_second": 923.092,
      "eval_steps_per_second": 1.2,
      "step": 328
    },
    {
      "epoch": 82.5,
      "learning_rate": 8.75e-06,
      "loss": 0.2481,
      "step": 330
    },
    {
      "epoch": 83.0,
      "eval_loss": 0.03600754216313362,
      "eval_runtime": 10.8588,
      "eval_samples_per_second": 920.91,
      "eval_steps_per_second": 1.197,
      "step": 332
    },
    {
      "epoch": 84.0,
      "eval_loss": 0.036581408232450485,
      "eval_runtime": 10.7405,
      "eval_samples_per_second": 931.052,
      "eval_steps_per_second": 1.21,
      "step": 336
    },
    {
      "epoch": 85.0,
      "learning_rate": 7.5e-06,
      "loss": 0.2474,
      "step": 340
    },
    {
      "epoch": 85.0,
      "eval_loss": 0.03646053001284599,
      "eval_runtime": 10.9336,
      "eval_samples_per_second": 914.608,
      "eval_steps_per_second": 1.189,
      "step": 340
    },
    {
      "epoch": 86.0,
      "eval_loss": 0.03565111756324768,
      "eval_runtime": 10.7346,
      "eval_samples_per_second": 931.569,
      "eval_steps_per_second": 1.211,
      "step": 344
    },
    {
      "epoch": 87.0,
      "eval_loss": 0.035478148609399796,
      "eval_runtime": 10.8414,
      "eval_samples_per_second": 922.386,
      "eval_steps_per_second": 1.199,
      "step": 348
    },
    {
      "epoch": 87.5,
      "learning_rate": 6.25e-06,
      "loss": 0.2537,
      "step": 350
    },
    {
      "epoch": 88.0,
      "eval_loss": 0.03596709668636322,
      "eval_runtime": 10.8347,
      "eval_samples_per_second": 922.957,
      "eval_steps_per_second": 1.2,
      "step": 352
    },
    {
      "epoch": 89.0,
      "eval_loss": 0.03587077185511589,
      "eval_runtime": 10.7439,
      "eval_samples_per_second": 930.763,
      "eval_steps_per_second": 1.21,
      "step": 356
    },
    {
      "epoch": 90.0,
      "learning_rate": 5e-06,
      "loss": 0.2438,
      "step": 360
    },
    {
      "epoch": 90.0,
      "eval_loss": 0.03546379879117012,
      "eval_runtime": 10.9027,
      "eval_samples_per_second": 917.201,
      "eval_steps_per_second": 1.192,
      "step": 360
    },
    {
      "epoch": 91.0,
      "eval_loss": 0.03530960530042648,
      "eval_runtime": 10.7401,
      "eval_samples_per_second": 931.093,
      "eval_steps_per_second": 1.21,
      "step": 364
    },
    {
      "epoch": 92.0,
      "eval_loss": 0.034898195415735245,
      "eval_runtime": 10.838,
      "eval_samples_per_second": 922.682,
      "eval_steps_per_second": 1.199,
      "step": 368
    },
    {
      "epoch": 92.5,
      "learning_rate": 3.75e-06,
      "loss": 0.2461,
      "step": 370
    },
    {
      "epoch": 93.0,
      "eval_loss": 0.03429694473743439,
      "eval_runtime": 10.8462,
      "eval_samples_per_second": 921.982,
      "eval_steps_per_second": 1.199,
      "step": 372
    },
    {
      "epoch": 94.0,
      "eval_loss": 0.03415210172533989,
      "eval_runtime": 10.73,
      "eval_samples_per_second": 931.966,
      "eval_steps_per_second": 1.212,
      "step": 376
    }
  ],
  "logging_steps": 10,
  "max_steps": 400,
  "num_train_epochs": 100,
  "save_steps": 500,
  "total_flos": 8096480649216000.0,
  "trial_name": null,
  "trial_params": null
}