{
  "best_metric": 1.536595344543457,
  "best_model_checkpoint": "output/face/checkpoint-402",
  "epoch": 3.0,
  "global_step": 402,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.04,
      "learning_rate": 0.00013672921035335123,
      "loss": 2.5516,
      "step": 5
    },
    {
      "epoch": 0.07,
      "learning_rate": 0.00013532330330528217,
      "loss": 2.1537,
      "step": 10
    },
    {
      "epoch": 0.11,
      "learning_rate": 0.00013300157583778652,
      "loss": 2.0468,
      "step": 15
    },
    {
      "epoch": 0.15,
      "learning_rate": 0.00012979589515943672,
      "loss": 2.0981,
      "step": 20
    },
    {
      "epoch": 0.19,
      "learning_rate": 0.00012575026130737067,
      "loss": 1.9179,
      "step": 25
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0001209202032183722,
      "loss": 2.1116,
      "step": 30
    },
    {
      "epoch": 0.26,
      "learning_rate": 0.00011537201655836005,
      "loss": 1.9485,
      "step": 35
    },
    {
      "epoch": 0.3,
      "learning_rate": 0.00010918185377156764,
      "loss": 1.8987,
      "step": 40
    },
    {
      "epoch": 0.34,
      "learning_rate": 0.00010243467883907644,
      "loss": 1.9255,
      "step": 45
    },
    {
      "epoch": 0.37,
      "learning_rate": 9.522310109331633e-05,
      "loss": 1.9168,
      "step": 50
    },
    {
      "epoch": 0.41,
      "learning_rate": 8.764610409518253e-05,
      "loss": 1.8695,
      "step": 55
    },
    {
      "epoch": 0.45,
      "learning_rate": 7.980768702075116e-05,
      "loss": 1.8724,
      "step": 60
    },
    {
      "epoch": 0.49,
      "learning_rate": 7.181543720543789e-05,
      "loss": 1.854,
      "step": 65
    },
    {
      "epoch": 0.52,
      "learning_rate": 6.377905343835293e-05,
      "loss": 1.833,
      "step": 70
    },
    {
      "epoch": 0.56,
      "learning_rate": 5.580884027559068e-05,
      "loss": 1.8861,
      "step": 75
    },
    {
      "epoch": 0.6,
      "learning_rate": 4.8014194038976637e-05,
      "loss": 1.9461,
      "step": 80
    },
    {
      "epoch": 0.63,
      "learning_rate": 4.0502101280915654e-05,
      "loss": 1.8331,
      "step": 85
    },
    {
      "epoch": 0.67,
      "learning_rate": 3.337567032488183e-05,
      "loss": 1.8838,
      "step": 90
    },
    {
      "epoch": 0.71,
      "learning_rate": 2.673271603710427e-05,
      "loss": 1.8553,
      "step": 95
    },
    {
      "epoch": 0.75,
      "learning_rate": 2.0664417254371463e-05,
      "loss": 1.7774,
      "step": 100
    },
    {
      "epoch": 0.78,
      "learning_rate": 1.5254065295624607e-05,
      "loss": 1.7757,
      "step": 105
    },
    {
      "epoch": 0.82,
      "learning_rate": 1.0575920734825916e-05,
      "loss": 1.7125,
      "step": 110
    },
    {
      "epoch": 0.86,
      "learning_rate": 6.6941941266328415e-06,
      "loss": 1.848,
      "step": 115
    },
    {
      "epoch": 0.9,
      "learning_rate": 3.66216467507622e-06,
      "loss": 1.8553,
      "step": 120
    },
    {
      "epoch": 0.93,
      "learning_rate": 1.5214489420839643e-06,
      "loss": 1.9048,
      "step": 125
    },
    {
      "epoch": 0.97,
      "learning_rate": 3.014296332977907e-07,
      "loss": 1.587,
      "step": 130
    },
    {
      "epoch": 1.0,
      "eval_loss": 1.7584949731826782,
      "eval_runtime": 8.5759,
      "eval_samples_per_second": 22.738,
      "eval_steps_per_second": 2.915,
      "step": 134
    },
    {
      "epoch": 1.01,
      "learning_rate": 1.885230146662962e-08,
      "loss": 1.6643,
      "step": 135
    },
    {
      "epoch": 1.04,
      "learning_rate": 6.775955029229377e-07,
      "loss": 1.839,
      "step": 140
    },
    {
      "epoch": 1.08,
      "learning_rate": 2.268617561892799e-06,
      "loss": 1.7973,
      "step": 145
    },
    {
      "epoch": 1.12,
      "learning_rate": 4.7700806733338495e-06,
      "loss": 1.6847,
      "step": 150
    },
    {
      "epoch": 1.16,
      "learning_rate": 8.147650640909638e-06,
      "loss": 1.9368,
      "step": 155
    },
    {
      "epoch": 1.19,
      "learning_rate": 1.235496813600306e-05,
      "loss": 1.7733,
      "step": 160
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.7334285009434683e-05,
      "loss": 1.7367,
      "step": 165
    },
    {
      "epoch": 1.27,
      "learning_rate": 2.3017256922096845e-05,
      "loss": 1.856,
      "step": 170
    },
    {
      "epoch": 1.31,
      "learning_rate": 2.932588141513589e-05,
      "loss": 1.8208,
      "step": 175
    },
    {
      "epoch": 1.34,
      "learning_rate": 3.6173568544064456e-05,
      "loss": 1.6543,
      "step": 180
    },
    {
      "epoch": 1.38,
      "learning_rate": 4.3466329381658294e-05,
      "loss": 1.7295,
      "step": 185
    },
    {
      "epoch": 1.42,
      "learning_rate": 5.110406607666898e-05,
      "loss": 1.7178,
      "step": 190
    },
    {
      "epoch": 1.46,
      "learning_rate": 5.898194576146263e-05,
      "loss": 1.7418,
      "step": 195
    },
    {
      "epoch": 1.49,
      "learning_rate": 6.699183945081632e-05,
      "loss": 1.6734,
      "step": 200
    },
    {
      "epoch": 1.53,
      "learning_rate": 7.502380618205647e-05,
      "loss": 1.7512,
      "step": 205
    },
    {
      "epoch": 1.57,
      "learning_rate": 8.296760202575751e-05,
      "loss": 1.7607,
      "step": 210
    },
    {
      "epoch": 1.6,
      "learning_rate": 9.071419325485517e-05,
      "loss": 1.6486,
      "step": 215
    },
    {
      "epoch": 1.64,
      "learning_rate": 9.815725290295059e-05,
      "loss": 1.7547,
      "step": 220
    },
    {
      "epoch": 1.68,
      "learning_rate": 0.00010519462017057694,
      "loss": 1.773,
      "step": 225
    },
    {
      "epoch": 1.72,
      "learning_rate": 0.00011172970264813468,
      "loss": 1.7159,
      "step": 230
    },
    {
      "epoch": 1.75,
      "learning_rate": 0.00011767280210908074,
      "loss": 1.6116,
      "step": 235
    },
    {
      "epoch": 1.79,
      "learning_rate": 0.00012294234567600346,
      "loss": 1.6329,
      "step": 240
    },
    {
      "epoch": 1.83,
      "learning_rate": 0.0001274660054610349,
      "loss": 1.6289,
      "step": 245
    },
    {
      "epoch": 1.87,
      "learning_rate": 0.00013118169131281192,
      "loss": 1.7835,
      "step": 250
    },
    {
      "epoch": 1.9,
      "learning_rate": 0.00013403840304389384,
      "loss": 1.7385,
      "step": 255
    },
    {
      "epoch": 1.94,
      "learning_rate": 0.00013599693044126453,
      "loss": 1.6406,
      "step": 260
    },
    {
      "epoch": 1.98,
      "learning_rate": 0.0001370303914518231,
      "loss": 1.7128,
      "step": 265
    },
    {
      "epoch": 2.0,
      "eval_loss": 1.7059965133666992,
      "eval_runtime": 8.588,
      "eval_samples_per_second": 22.706,
      "eval_steps_per_second": 2.911,
      "step": 268
    },
    {
      "epoch": 2.01,
      "learning_rate": 0.0001371246011559198,
      "loss": 1.731,
      "step": 270
    },
    {
      "epoch": 2.05,
      "learning_rate": 0.0001362782664645371,
      "loss": 1.6293,
      "step": 275
    },
    {
      "epoch": 2.09,
      "learning_rate": 0.00013450300386777127,
      "loss": 1.5935,
      "step": 280
    },
    {
      "epoch": 2.13,
      "learning_rate": 0.00013182317999100535,
      "loss": 1.5613,
      "step": 285
    },
    {
      "epoch": 2.16,
      "learning_rate": 0.00012827557714724304,
      "loss": 1.8451,
      "step": 290
    },
    {
      "epoch": 2.2,
      "learning_rate": 0.000123908888476113,
      "loss": 1.7444,
      "step": 295
    },
    {
      "epoch": 2.24,
      "learning_rate": 0.00011878304959908774,
      "loss": 1.6749,
      "step": 300
    },
    {
      "epoch": 2.28,
      "learning_rate": 0.00011296841596437982,
      "loss": 1.5253,
      "step": 305
    },
    {
      "epoch": 2.31,
      "learning_rate": 0.00010654479717298873,
      "loss": 1.6357,
      "step": 310
    },
    {
      "epoch": 2.35,
      "learning_rate": 9.96003615403955e-05,
      "loss": 1.6636,
      "step": 315
    },
    {
      "epoch": 2.39,
      "learning_rate": 9.223042592950526e-05,
      "loss": 1.646,
      "step": 320
    },
    {
      "epoch": 2.43,
      "learning_rate": 8.453614746515988e-05,
      "loss": 1.7026,
      "step": 325
    },
    {
      "epoch": 2.46,
      "learning_rate": 7.662313508728495e-05,
      "loss": 1.6289,
      "step": 330
    },
    {
      "epoch": 2.5,
      "learning_rate": 6.860000000000001e-05,
      "loss": 1.4981,
      "step": 335
    },
    {
      "epoch": 2.54,
      "learning_rate": 6.0576864912715095e-05,
      "loss": 1.5564,
      "step": 340
    },
    {
      "epoch": 2.57,
      "learning_rate": 5.266385253484016e-05,
      "loss": 1.6555,
      "step": 345
    },
    {
      "epoch": 2.61,
      "learning_rate": 4.496957407049471e-05,
      "loss": 1.5796,
      "step": 350
    },
    {
      "epoch": 2.65,
      "learning_rate": 3.759963845960453e-05,
      "loss": 1.7131,
      "step": 355
    },
    {
      "epoch": 2.69,
      "learning_rate": 3.065520282701126e-05,
      "loss": 1.61,
      "step": 360
    },
    {
      "epoch": 2.72,
      "learning_rate": 2.423158403562016e-05,
      "loss": 1.6306,
      "step": 365
    },
    {
      "epoch": 2.76,
      "learning_rate": 1.8416950400912332e-05,
      "loss": 1.5807,
      "step": 370
    },
    {
      "epoch": 2.8,
      "learning_rate": 1.3291111523887019e-05,
      "loss": 1.5766,
      "step": 375
    },
    {
      "epoch": 2.84,
      "learning_rate": 8.924422852757e-06,
      "loss": 1.6478,
      "step": 380
    },
    {
      "epoch": 2.87,
      "learning_rate": 5.376820008994686e-06,
      "loss": 1.6091,
      "step": 385
    },
    {
      "epoch": 2.91,
      "learning_rate": 2.6969961322287634e-06,
      "loss": 1.512,
      "step": 390
    },
    {
      "epoch": 2.95,
      "learning_rate": 9.21733535462923e-07,
      "loss": 1.57,
      "step": 395
    },
    {
      "epoch": 2.99,
      "learning_rate": 7.53988440801922e-08,
      "loss": 1.5282,
      "step": 400
    },
    {
      "epoch": 3.0,
      "eval_loss": 1.536595344543457,
      "eval_runtime": 8.6245,
      "eval_samples_per_second": 22.726,
      "eval_steps_per_second": 2.899,
      "step": 402
    }
  ],
  "max_steps": 402,
  "num_train_epochs": 3,
  "total_flos": 418067251200000.0,
  "trial_name": null,
  "trial_params": null
}