{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 49.53191489361702,
  "eval_steps": 500,
  "global_step": 337,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.09,
      "learning_rate": 9.999918433243253e-06,
      "loss": 2.0691,
      "step": 1
    },
    {
      "epoch": 0.43,
      "learning_rate": 9.997960964140946e-06,
      "loss": 2.078,
      "step": 5
    },
    {
      "epoch": 0.51,
      "eval_loss": 2.030493974685669,
      "eval_runtime": 3.9443,
      "eval_samples_per_second": 21.043,
      "eval_steps_per_second": 1.521,
      "step": 6
    },
    {
      "epoch": 1.28,
      "learning_rate": 9.991845519630679e-06,
      "loss": 2.0322,
      "step": 10
    },
    {
      "epoch": 1.53,
      "eval_loss": 1.952285647392273,
      "eval_runtime": 3.904,
      "eval_samples_per_second": 21.26,
      "eval_steps_per_second": 1.537,
      "step": 13
    },
    {
      "epoch": 2.13,
      "learning_rate": 9.981658654313458e-06,
      "loss": 1.9849,
      "step": 15
    },
    {
      "epoch": 2.55,
      "learning_rate": 9.96740867674275e-06,
      "loss": 1.9301,
      "step": 20
    },
    {
      "epoch": 2.55,
      "eval_loss": 1.8813837766647339,
      "eval_runtime": 3.9106,
      "eval_samples_per_second": 21.224,
      "eval_steps_per_second": 1.534,
      "step": 20
    },
    {
      "epoch": 3.4,
      "learning_rate": 9.949107209404664e-06,
      "loss": 1.8757,
      "step": 25
    },
    {
      "epoch": 3.57,
      "eval_loss": 1.8208731412887573,
      "eval_runtime": 3.8907,
      "eval_samples_per_second": 21.333,
      "eval_steps_per_second": 1.542,
      "step": 27
    },
    {
      "epoch": 4.26,
      "learning_rate": 9.926769179238467e-06,
      "loss": 1.841,
      "step": 30
    },
    {
      "epoch": 4.51,
      "eval_loss": 1.7722175121307373,
      "eval_runtime": 3.8932,
      "eval_samples_per_second": 21.319,
      "eval_steps_per_second": 1.541,
      "step": 33
    },
    {
      "epoch": 5.11,
      "learning_rate": 9.900412805461968e-06,
      "loss": 1.7965,
      "step": 35
    },
    {
      "epoch": 5.53,
      "learning_rate": 9.870059584711668e-06,
      "loss": 1.7661,
      "step": 40
    },
    {
      "epoch": 5.53,
      "eval_loss": 1.7291498184204102,
      "eval_runtime": 3.8915,
      "eval_samples_per_second": 21.329,
      "eval_steps_per_second": 1.542,
      "step": 40
    },
    {
      "epoch": 6.38,
      "learning_rate": 9.835734273509787e-06,
      "loss": 1.731,
      "step": 45
    },
    {
      "epoch": 6.55,
      "eval_loss": 1.6903846263885498,
      "eval_runtime": 3.9149,
      "eval_samples_per_second": 21.201,
      "eval_steps_per_second": 1.533,
      "step": 47
    },
    {
      "epoch": 7.23,
      "learning_rate": 9.797464868072489e-06,
      "loss": 1.713,
      "step": 50
    },
    {
      "epoch": 7.57,
      "eval_loss": 1.6531494855880737,
      "eval_runtime": 3.8992,
      "eval_samples_per_second": 21.286,
      "eval_steps_per_second": 1.539,
      "step": 54
    },
    {
      "epoch": 8.09,
      "learning_rate": 9.755282581475769e-06,
      "loss": 1.6854,
      "step": 55
    },
    {
      "epoch": 8.51,
      "learning_rate": 9.709221818197626e-06,
      "loss": 1.6557,
      "step": 60
    },
    {
      "epoch": 8.51,
      "eval_loss": 1.6243276596069336,
      "eval_runtime": 3.917,
      "eval_samples_per_second": 21.19,
      "eval_steps_per_second": 1.532,
      "step": 60
    },
    {
      "epoch": 9.36,
      "learning_rate": 9.659320146057263e-06,
      "loss": 1.6319,
      "step": 65
    },
    {
      "epoch": 9.53,
      "eval_loss": 1.5888899564743042,
      "eval_runtime": 3.9117,
      "eval_samples_per_second": 21.218,
      "eval_steps_per_second": 1.534,
      "step": 67
    },
    {
      "epoch": 10.21,
      "learning_rate": 9.60561826557425e-06,
      "loss": 1.5989,
      "step": 70
    },
    {
      "epoch": 10.55,
      "eval_loss": 1.5500489473342896,
      "eval_runtime": 3.8961,
      "eval_samples_per_second": 21.303,
      "eval_steps_per_second": 1.54,
      "step": 74
    },
    {
      "epoch": 11.06,
      "learning_rate": 9.548159976772593e-06,
      "loss": 1.581,
      "step": 75
    },
    {
      "epoch": 11.49,
      "learning_rate": 9.486992143456792e-06,
      "loss": 1.5556,
      "step": 80
    },
    {
      "epoch": 11.57,
      "eval_loss": 1.5097848176956177,
      "eval_runtime": 3.8656,
      "eval_samples_per_second": 21.471,
      "eval_steps_per_second": 1.552,
      "step": 81
    },
    {
      "epoch": 12.34,
      "learning_rate": 9.422164654989073e-06,
      "loss": 1.5165,
      "step": 85
    },
    {
      "epoch": 12.51,
      "eval_loss": 1.4754232168197632,
      "eval_runtime": 3.8958,
      "eval_samples_per_second": 21.305,
      "eval_steps_per_second": 1.54,
      "step": 87
    },
    {
      "epoch": 13.19,
      "learning_rate": 9.353730385598887e-06,
      "loss": 1.4945,
      "step": 90
    },
    {
      "epoch": 13.53,
      "eval_loss": 1.4281667470932007,
      "eval_runtime": 3.8886,
      "eval_samples_per_second": 21.345,
      "eval_steps_per_second": 1.543,
      "step": 94
    },
    {
      "epoch": 14.04,
      "learning_rate": 9.281745151257946e-06,
      "loss": 1.4578,
      "step": 95
    },
    {
      "epoch": 14.47,
      "learning_rate": 9.206267664155906e-06,
      "loss": 1.4198,
      "step": 100
    },
    {
      "epoch": 14.55,
      "eval_loss": 1.377801775932312,
      "eval_runtime": 3.8937,
      "eval_samples_per_second": 21.317,
      "eval_steps_per_second": 1.541,
      "step": 101
    },
    {
      "epoch": 15.32,
      "learning_rate": 9.12735948481387e-06,
      "loss": 1.3823,
      "step": 105
    },
    {
      "epoch": 15.57,
      "eval_loss": 1.3291016817092896,
      "eval_runtime": 3.8883,
      "eval_samples_per_second": 21.346,
      "eval_steps_per_second": 1.543,
      "step": 108
    },
    {
      "epoch": 16.17,
      "learning_rate": 9.045084971874738e-06,
      "loss": 1.3576,
      "step": 110
    },
    {
      "epoch": 16.51,
      "eval_loss": 1.2924872636795044,
      "eval_runtime": 3.9257,
      "eval_samples_per_second": 21.143,
      "eval_steps_per_second": 1.528,
      "step": 114
    },
    {
      "epoch": 17.02,
      "learning_rate": 8.959511229611377e-06,
      "loss": 1.3142,
      "step": 115
    },
    {
      "epoch": 17.45,
      "learning_rate": 8.870708053195414e-06,
      "loss": 1.2917,
      "step": 120
    },
    {
      "epoch": 17.53,
      "eval_loss": 1.25697660446167,
      "eval_runtime": 3.9218,
      "eval_samples_per_second": 21.163,
      "eval_steps_per_second": 1.53,
      "step": 121
    },
    {
      "epoch": 18.3,
      "learning_rate": 8.778747871771293e-06,
      "loss": 1.2599,
      "step": 125
    },
    {
      "epoch": 18.55,
      "eval_loss": 1.2282600402832031,
      "eval_runtime": 3.9116,
      "eval_samples_per_second": 21.219,
      "eval_steps_per_second": 1.534,
      "step": 128
    },
    {
      "epoch": 19.15,
      "learning_rate": 8.683705689382025e-06,
      "loss": 1.2529,
      "step": 130
    },
    {
      "epoch": 19.57,
      "learning_rate": 8.585659023794818e-06,
      "loss": 1.2257,
      "step": 135
    },
    {
      "epoch": 19.57,
      "eval_loss": 1.203251838684082,
      "eval_runtime": 3.8892,
      "eval_samples_per_second": 21.341,
      "eval_steps_per_second": 1.543,
      "step": 135
    },
    {
      "epoch": 20.43,
      "learning_rate": 8.48468784327647e-06,
      "loss": 1.2123,
      "step": 140
    },
    {
      "epoch": 20.51,
      "eval_loss": 1.1904693841934204,
      "eval_runtime": 3.8787,
      "eval_samples_per_second": 21.399,
      "eval_steps_per_second": 1.547,
      "step": 141
    },
    {
      "epoch": 21.28,
      "learning_rate": 8.380874501370098e-06,
      "loss": 1.1966,
      "step": 145
    },
    {
      "epoch": 21.53,
      "eval_loss": 1.172432541847229,
      "eval_runtime": 3.8728,
      "eval_samples_per_second": 21.432,
      "eval_steps_per_second": 1.549,
      "step": 148
    },
    {
      "epoch": 22.13,
      "learning_rate": 8.274303669726427e-06,
      "loss": 1.1938,
      "step": 150
    },
    {
      "epoch": 22.55,
      "learning_rate": 8.165062269044353e-06,
      "loss": 1.1694,
      "step": 155
    },
    {
      "epoch": 22.55,
      "eval_loss": 1.1592084169387817,
      "eval_runtime": 3.8798,
      "eval_samples_per_second": 21.393,
      "eval_steps_per_second": 1.546,
      "step": 155
    },
    {
      "epoch": 23.4,
      "learning_rate": 8.053239398177191e-06,
      "loss": 1.1665,
      "step": 160
    },
    {
      "epoch": 23.57,
      "eval_loss": 1.1471155881881714,
      "eval_runtime": 3.891,
      "eval_samples_per_second": 21.331,
      "eval_steps_per_second": 1.542,
      "step": 162
    },
    {
      "epoch": 24.26,
      "learning_rate": 7.938926261462366e-06,
      "loss": 1.1559,
      "step": 165
    },
    {
      "epoch": 24.51,
      "eval_loss": 1.1368687152862549,
      "eval_runtime": 3.9159,
      "eval_samples_per_second": 21.196,
      "eval_steps_per_second": 1.532,
      "step": 168
    },
    {
      "epoch": 25.11,
      "learning_rate": 7.822216094333847e-06,
      "loss": 1.1552,
      "step": 170
    },
    {
      "epoch": 25.53,
      "learning_rate": 7.703204087277989e-06,
      "loss": 1.1383,
      "step": 175
    },
    {
      "epoch": 25.53,
      "eval_loss": 1.128839373588562,
      "eval_runtime": 3.89,
      "eval_samples_per_second": 21.337,
      "eval_steps_per_second": 1.542,
      "step": 175
    },
    {
      "epoch": 26.38,
      "learning_rate": 7.5819873081948105e-06,
      "loss": 1.141,
      "step": 180
    },
    {
      "epoch": 26.55,
      "eval_loss": 1.1200487613677979,
      "eval_runtime": 3.8914,
      "eval_samples_per_second": 21.329,
      "eval_steps_per_second": 1.542,
      "step": 182
    },
    {
      "epoch": 27.23,
      "learning_rate": 7.45866462322802e-06,
      "loss": 1.1334,
      "step": 185
    },
    {
      "epoch": 27.57,
      "eval_loss": 1.1138092279434204,
      "eval_runtime": 3.8814,
      "eval_samples_per_second": 21.384,
      "eval_steps_per_second": 1.546,
      "step": 189
    },
    {
      "epoch": 28.09,
      "learning_rate": 7.333336616128369e-06,
      "loss": 1.1201,
      "step": 190
    },
    {
      "epoch": 28.51,
      "learning_rate": 7.206105506216107e-06,
      "loss": 1.1193,
      "step": 195
    },
    {
      "epoch": 28.51,
      "eval_loss": 1.1078517436981201,
      "eval_runtime": 3.9032,
      "eval_samples_per_second": 21.265,
      "eval_steps_per_second": 1.537,
      "step": 195
    },
    {
      "epoch": 29.36,
      "learning_rate": 7.0770750650094335e-06,
      "loss": 1.1079,
      "step": 200
    },
    {
      "epoch": 29.53,
      "eval_loss": 1.1015825271606445,
      "eval_runtime": 3.9112,
      "eval_samples_per_second": 21.221,
      "eval_steps_per_second": 1.534,
      "step": 202
    },
    {
      "epoch": 30.21,
      "learning_rate": 6.946350531586959e-06,
      "loss": 1.1188,
      "step": 205
    },
    {
      "epoch": 30.55,
      "eval_loss": 1.0960975885391235,
      "eval_runtime": 3.9135,
      "eval_samples_per_second": 21.209,
      "eval_steps_per_second": 1.533,
      "step": 209
    },
    {
      "epoch": 31.06,
      "learning_rate": 6.814038526753205e-06,
      "loss": 1.1013,
      "step": 210
    },
    {
      "epoch": 31.49,
      "learning_rate": 6.680246966077151e-06,
      "loss": 1.1006,
      "step": 215
    },
    {
      "epoch": 31.57,
      "eval_loss": 1.0916202068328857,
      "eval_runtime": 3.8812,
      "eval_samples_per_second": 21.385,
      "eval_steps_per_second": 1.546,
      "step": 216
    },
    {
      "epoch": 32.34,
      "learning_rate": 6.545084971874738e-06,
      "loss": 1.1016,
      "step": 220
    },
    {
      "epoch": 32.51,
      "eval_loss": 1.0850551128387451,
      "eval_runtime": 3.897,
      "eval_samples_per_second": 21.298,
      "eval_steps_per_second": 1.54,
      "step": 222
    },
    {
      "epoch": 33.19,
      "learning_rate": 6.408662784207149e-06,
      "loss": 1.0801,
      "step": 225
    },
    {
      "epoch": 33.53,
      "eval_loss": 1.078303337097168,
      "eval_runtime": 3.8819,
      "eval_samples_per_second": 21.381,
      "eval_steps_per_second": 1.546,
      "step": 229
    },
    {
      "epoch": 34.04,
      "learning_rate": 6.271091670967437e-06,
      "loss": 1.0942,
      "step": 230
    },
    {
      "epoch": 34.47,
      "learning_rate": 6.132483837128823e-06,
      "loss": 1.0846,
      "step": 235
    },
    {
      "epoch": 34.55,
      "eval_loss": 1.075777292251587,
      "eval_runtime": 3.9038,
      "eval_samples_per_second": 21.261,
      "eval_steps_per_second": 1.537,
      "step": 236
    },
    {
      "epoch": 35.32,
      "learning_rate": 5.9929523332287275e-06,
      "loss": 1.0828,
      "step": 240
    },
    {
      "epoch": 35.57,
      "eval_loss": 1.0725317001342773,
      "eval_runtime": 3.9021,
      "eval_samples_per_second": 21.271,
      "eval_steps_per_second": 1.538,
      "step": 243
    },
    {
      "epoch": 36.17,
      "learning_rate": 5.85261096316312e-06,
      "loss": 1.0758,
      "step": 245
    },
    {
      "epoch": 36.51,
      "eval_loss": 1.0693665742874146,
      "eval_runtime": 3.9189,
      "eval_samples_per_second": 21.179,
      "eval_steps_per_second": 1.531,
      "step": 249
    },
    {
      "epoch": 37.02,
      "learning_rate": 5.711574191366427e-06,
      "loss": 1.0707,
      "step": 250
    },
    {
      "epoch": 37.45,
      "learning_rate": 5.569957049452703e-06,
      "loss": 1.0749,
      "step": 255
    },
    {
      "epoch": 37.53,
      "eval_loss": 1.0646324157714844,
      "eval_runtime": 3.9232,
      "eval_samples_per_second": 21.156,
      "eval_steps_per_second": 1.529,
      "step": 256
    },
    {
      "epoch": 38.3,
      "learning_rate": 5.4278750423942e-06,
      "loss": 1.0626,
      "step": 260
    },
    {
      "epoch": 38.55,
      "eval_loss": 1.0626643896102905,
      "eval_runtime": 3.9102,
      "eval_samples_per_second": 21.227,
      "eval_steps_per_second": 1.534,
      "step": 263
    },
    {
      "epoch": 39.15,
      "learning_rate": 5.285444054313841e-06,
      "loss": 1.0698,
      "step": 265
    },
    {
      "epoch": 39.57,
      "learning_rate": 5.142780253968481e-06,
      "loss": 1.0575,
      "step": 270
    },
    {
      "epoch": 39.57,
      "eval_loss": 1.059247612953186,
      "eval_runtime": 3.9014,
      "eval_samples_per_second": 21.274,
      "eval_steps_per_second": 1.538,
      "step": 270
    },
    {
      "epoch": 40.43,
      "learning_rate": 5e-06,
      "loss": 1.0583,
      "step": 275
    },
    {
      "epoch": 40.51,
      "eval_loss": 1.0554794073104858,
      "eval_runtime": 3.924,
      "eval_samples_per_second": 21.152,
      "eval_steps_per_second": 1.529,
      "step": 276
    },
    {
      "epoch": 41.28,
      "learning_rate": 4.85721974603152e-06,
      "loss": 1.0548,
      "step": 280
    },
    {
      "epoch": 41.53,
      "eval_loss": 1.051832914352417,
      "eval_runtime": 3.9145,
      "eval_samples_per_second": 21.203,
      "eval_steps_per_second": 1.533,
      "step": 283
    },
    {
      "epoch": 42.13,
      "learning_rate": 4.71455594568616e-06,
      "loss": 1.0506,
      "step": 285
    },
    {
      "epoch": 42.55,
      "learning_rate": 4.572124957605803e-06,
      "loss": 1.0495,
      "step": 290
    },
    {
      "epoch": 42.55,
      "eval_loss": 1.0468100309371948,
      "eval_runtime": 3.9072,
      "eval_samples_per_second": 21.243,
      "eval_steps_per_second": 1.536,
      "step": 290
    },
    {
      "epoch": 43.4,
      "learning_rate": 4.430042950547298e-06,
      "loss": 1.0449,
      "step": 295
    },
    {
      "epoch": 43.57,
      "eval_loss": 1.0469093322753906,
      "eval_runtime": 3.8777,
      "eval_samples_per_second": 21.404,
      "eval_steps_per_second": 1.547,
      "step": 297
    },
    {
      "epoch": 44.26,
      "learning_rate": 4.2884258086335755e-06,
      "loss": 1.0527,
      "step": 300
    },
    {
      "epoch": 44.51,
      "eval_loss": 1.041974663734436,
      "eval_runtime": 3.9068,
      "eval_samples_per_second": 21.245,
      "eval_steps_per_second": 1.536,
      "step": 303
    },
    {
      "epoch": 45.11,
      "learning_rate": 4.147389036836881e-06,
      "loss": 1.0413,
      "step": 305
    },
    {
      "epoch": 45.53,
      "learning_rate": 4.007047666771274e-06,
      "loss": 1.0411,
      "step": 310
    },
    {
      "epoch": 45.53,
      "eval_loss": 1.041487693786621,
      "eval_runtime": 3.9044,
      "eval_samples_per_second": 21.258,
      "eval_steps_per_second": 1.537,
      "step": 310
    },
    {
      "epoch": 46.38,
      "learning_rate": 3.867516162871177e-06,
      "loss": 1.0325,
      "step": 315
    },
    {
      "epoch": 46.55,
      "eval_loss": 1.0384427309036255,
      "eval_runtime": 3.8991,
      "eval_samples_per_second": 21.287,
      "eval_steps_per_second": 1.539,
      "step": 317
    },
    {
      "epoch": 47.23,
      "learning_rate": 3.7289083290325668e-06,
      "loss": 1.0404,
      "step": 320
    },
    {
      "epoch": 47.57,
      "eval_loss": 1.0352696180343628,
      "eval_runtime": 3.8837,
      "eval_samples_per_second": 21.371,
      "eval_steps_per_second": 1.545,
      "step": 324
    },
    {
      "epoch": 48.09,
      "learning_rate": 3.5913372157928515e-06,
      "loss": 1.034,
      "step": 325
    },
    {
      "epoch": 48.51,
      "learning_rate": 3.4549150281252635e-06,
      "loss": 1.0326,
      "step": 330
    },
    {
      "epoch": 48.51,
      "eval_loss": 1.033745527267456,
      "eval_runtime": 3.9032,
      "eval_samples_per_second": 21.265,
      "eval_steps_per_second": 1.537,
      "step": 330
    },
    {
      "epoch": 49.36,
      "learning_rate": 3.319753033922849e-06,
      "loss": 1.0262,
      "step": 335
    },
    {
      "epoch": 49.53,
      "eval_loss": 1.0316756963729858,
      "eval_runtime": 3.8951,
      "eval_samples_per_second": 21.309,
      "eval_steps_per_second": 1.54,
      "step": 337
    },
    {
      "epoch": 49.53,
      "step": 337,
      "total_flos": 7152558019706880.0,
      "train_loss": 1.3011242423637683,
      "train_runtime": 6393.7569,
      "train_samples_per_second": 5.834,
      "train_steps_per_second": 0.086
    }
  ],
  "logging_steps": 5,
  "max_steps": 550,
  "num_train_epochs": 50,
  "save_steps": 500,
  "total_flos": 7152558019706880.0,
  "trial_name": null,
  "trial_params": null
}