{
  "best_metric": 0.042002853006124496,
  "best_model_checkpoint": "man_woman_face_image_detection/checkpoint-33509",
  "epoch": 7.0,
  "eval_steps": 500,
  "global_step": 33509,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.10444955086693128,
      "grad_norm": 2.75286865234375,
      "learning_rate": 9.865507038464987e-07,
      "loss": 0.0606,
      "step": 500
    },
    {
      "epoch": 0.20889910173386256,
      "grad_norm": 0.4504856765270233,
      "learning_rate": 9.716070414537195e-07,
      "loss": 0.0604,
      "step": 1000
    },
    {
      "epoch": 0.3133486526007938,
      "grad_norm": 4.492558479309082,
      "learning_rate": 9.566633790609401e-07,
      "loss": 0.0553,
      "step": 1500
    },
    {
      "epoch": 0.4177982034677251,
      "grad_norm": 0.11193133890628815,
      "learning_rate": 9.417197166681609e-07,
      "loss": 0.0592,
      "step": 2000
    },
    {
      "epoch": 0.5222477543346563,
      "grad_norm": 3.955646514892578,
      "learning_rate": 9.267760542753818e-07,
      "loss": 0.0582,
      "step": 2500
    },
    {
      "epoch": 0.6266973052015876,
      "grad_norm": 0.09266256541013718,
      "learning_rate": 9.118323918826025e-07,
      "loss": 0.0599,
      "step": 3000
    },
    {
      "epoch": 0.7311468560685189,
      "grad_norm": 0.46259480714797974,
      "learning_rate": 8.968887294898233e-07,
      "loss": 0.0649,
      "step": 3500
    },
    {
      "epoch": 0.8355964069354502,
      "grad_norm": 0.03443637862801552,
      "learning_rate": 8.819450670970441e-07,
      "loss": 0.0566,
      "step": 4000
    },
    {
      "epoch": 0.9400459578023814,
      "grad_norm": 2.622413158416748,
      "learning_rate": 8.670014047042648e-07,
      "loss": 0.0582,
      "step": 4500
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.986144295170577,
      "eval_loss": 0.04410167410969734,
      "eval_model_preparation_time": 0.0053,
      "eval_runtime": 1053.7087,
      "eval_samples_per_second": 96.919,
      "eval_steps_per_second": 12.115,
      "step": 4787
    },
    {
      "epoch": 1.0444955086693126,
      "grad_norm": 1.0063446760177612,
      "learning_rate": 8.520577423114857e-07,
      "loss": 0.0626,
      "step": 5000
    },
    {
      "epoch": 1.148945059536244,
      "grad_norm": 3.5057384967803955,
      "learning_rate": 8.371140799187064e-07,
      "loss": 0.0584,
      "step": 5500
    },
    {
      "epoch": 1.2533946104031752,
      "grad_norm": 0.10573850572109222,
      "learning_rate": 8.221704175259272e-07,
      "loss": 0.0576,
      "step": 6000
    },
    {
      "epoch": 1.3578441612701067,
      "grad_norm": 7.020801067352295,
      "learning_rate": 8.07226755133148e-07,
      "loss": 0.0543,
      "step": 6500
    },
    {
      "epoch": 1.4622937121370378,
      "grad_norm": 9.971087455749512,
      "learning_rate": 7.922830927403687e-07,
      "loss": 0.0517,
      "step": 7000
    },
    {
      "epoch": 1.566743263003969,
      "grad_norm": 3.649143695831299,
      "learning_rate": 7.773394303475895e-07,
      "loss": 0.0542,
      "step": 7500
    },
    {
      "epoch": 1.6711928138709005,
      "grad_norm": 0.7077023983001709,
      "learning_rate": 7.623957679548103e-07,
      "loss": 0.0576,
      "step": 8000
    },
    {
      "epoch": 1.7756423647378317,
      "grad_norm": 9.549636840820312,
      "learning_rate": 7.474521055620311e-07,
      "loss": 0.0558,
      "step": 8500
    },
    {
      "epoch": 1.8800919156047629,
      "grad_norm": 0.12657281756401062,
      "learning_rate": 7.325084431692519e-07,
      "loss": 0.0501,
      "step": 9000
    },
    {
      "epoch": 1.9845414664716943,
      "grad_norm": 4.699319362640381,
      "learning_rate": 7.175647807764726e-07,
      "loss": 0.0545,
      "step": 9500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9864968078022796,
      "eval_loss": 0.042747873812913895,
      "eval_model_preparation_time": 0.0053,
      "eval_runtime": 1036.1097,
      "eval_samples_per_second": 98.565,
      "eval_steps_per_second": 12.321,
      "step": 9574
    },
    {
      "epoch": 2.0889910173386252,
      "grad_norm": 1.3633027076721191,
      "learning_rate": 7.026211183836934e-07,
      "loss": 0.0485,
      "step": 10000
    },
    {
      "epoch": 2.1934405682055567,
      "grad_norm": 0.850628137588501,
      "learning_rate": 6.876774559909142e-07,
      "loss": 0.0607,
      "step": 10500
    },
    {
      "epoch": 2.297890119072488,
      "grad_norm": 4.338198661804199,
      "learning_rate": 6.72733793598135e-07,
      "loss": 0.0541,
      "step": 11000
    },
    {
      "epoch": 2.402339669939419,
      "grad_norm": 0.07052299380302429,
      "learning_rate": 6.577901312053558e-07,
      "loss": 0.0505,
      "step": 11500
    },
    {
      "epoch": 2.5067892208063505,
      "grad_norm": 4.150068283081055,
      "learning_rate": 6.428464688125765e-07,
      "loss": 0.0546,
      "step": 12000
    },
    {
      "epoch": 2.611238771673282,
      "grad_norm": 0.10252227634191513,
      "learning_rate": 6.279028064197973e-07,
      "loss": 0.0498,
      "step": 12500
    },
    {
      "epoch": 2.7156883225402133,
      "grad_norm": 0.12012413889169693,
      "learning_rate": 6.12959144027018e-07,
      "loss": 0.0459,
      "step": 13000
    },
    {
      "epoch": 2.8201378734071443,
      "grad_norm": 1.6249473094940186,
      "learning_rate": 5.980154816342389e-07,
      "loss": 0.0469,
      "step": 13500
    },
    {
      "epoch": 2.9245874242740757,
      "grad_norm": 2.401695489883423,
      "learning_rate": 5.830718192414597e-07,
      "loss": 0.051,
      "step": 14000
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9866926481532254,
      "eval_loss": 0.042268697172403336,
      "eval_model_preparation_time": 0.0053,
      "eval_runtime": 1031.115,
      "eval_samples_per_second": 99.042,
      "eval_steps_per_second": 12.381,
      "step": 14361
    },
    {
      "epoch": 3.0290369751410067,
      "grad_norm": 1.793472170829773,
      "learning_rate": 5.681281568486804e-07,
      "loss": 0.0598,
      "step": 14500
    },
    {
      "epoch": 3.133486526007938,
      "grad_norm": 7.33253812789917,
      "learning_rate": 5.531844944559012e-07,
      "loss": 0.0477,
      "step": 15000
    },
    {
      "epoch": 3.2379360768748695,
      "grad_norm": 0.11843205243349075,
      "learning_rate": 5.382408320631219e-07,
      "loss": 0.0527,
      "step": 15500
    },
    {
      "epoch": 3.342385627741801,
      "grad_norm": 1.8657333850860596,
      "learning_rate": 5.232971696703428e-07,
      "loss": 0.054,
      "step": 16000
    },
    {
      "epoch": 3.446835178608732,
      "grad_norm": 4.304357528686523,
      "learning_rate": 5.083535072775636e-07,
      "loss": 0.0521,
      "step": 16500
    },
    {
      "epoch": 3.5512847294756633,
      "grad_norm": 0.2691272795200348,
      "learning_rate": 4.934098448847843e-07,
      "loss": 0.0516,
      "step": 17000
    },
    {
      "epoch": 3.6557342803425943,
      "grad_norm": 6.170676231384277,
      "learning_rate": 4.784661824920051e-07,
      "loss": 0.0463,
      "step": 17500
    },
    {
      "epoch": 3.7601838312095257,
      "grad_norm": 3.405898094177246,
      "learning_rate": 4.635225200992259e-07,
      "loss": 0.0519,
      "step": 18000
    },
    {
      "epoch": 3.864633382076457,
      "grad_norm": 2.52126145362854,
      "learning_rate": 4.4857885770644665e-07,
      "loss": 0.0494,
      "step": 18500
    },
    {
      "epoch": 3.9690829329433885,
      "grad_norm": 11.166853904724121,
      "learning_rate": 4.336351953136674e-07,
      "loss": 0.0457,
      "step": 19000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9867220242058674,
      "eval_loss": 0.043168142437934875,
      "eval_model_preparation_time": 0.0053,
      "eval_runtime": 1034.639,
      "eval_samples_per_second": 98.705,
      "eval_steps_per_second": 12.339,
      "step": 19148
    },
    {
      "epoch": 4.07353248381032,
      "grad_norm": 0.24023988842964172,
      "learning_rate": 4.1869153292088824e-07,
      "loss": 0.0495,
      "step": 19500
    },
    {
      "epoch": 4.1779820346772505,
      "grad_norm": 7.678467273712158,
      "learning_rate": 4.03747870528109e-07,
      "loss": 0.0525,
      "step": 20000
    },
    {
      "epoch": 4.282431585544182,
      "grad_norm": 0.244981586933136,
      "learning_rate": 3.8880420813532977e-07,
      "loss": 0.0507,
      "step": 20500
    },
    {
      "epoch": 4.386881136411113,
      "grad_norm": 0.22300060093402863,
      "learning_rate": 3.7386054574255054e-07,
      "loss": 0.0509,
      "step": 21000
    },
    {
      "epoch": 4.491330687278045,
      "grad_norm": 5.288808345794678,
      "learning_rate": 3.589168833497713e-07,
      "loss": 0.0474,
      "step": 21500
    },
    {
      "epoch": 4.595780238144976,
      "grad_norm": 1.9191228151321411,
      "learning_rate": 3.4397322095699213e-07,
      "loss": 0.0488,
      "step": 22000
    },
    {
      "epoch": 4.700229789011908,
      "grad_norm": 5.287907123565674,
      "learning_rate": 3.290295585642129e-07,
      "loss": 0.0459,
      "step": 22500
    },
    {
      "epoch": 4.804679339878838,
      "grad_norm": 0.1149277314543724,
      "learning_rate": 3.1408589617143366e-07,
      "loss": 0.0447,
      "step": 23000
    },
    {
      "epoch": 4.9091288907457695,
      "grad_norm": 3.53507137298584,
      "learning_rate": 2.9914223377865443e-07,
      "loss": 0.0523,
      "step": 23500
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9870353687673808,
      "eval_loss": 0.0422465056180954,
      "eval_model_preparation_time": 0.0053,
      "eval_runtime": 991.7535,
      "eval_samples_per_second": 102.973,
      "eval_steps_per_second": 12.872,
      "step": 23935
    },
    {
      "epoch": 5.013578441612701,
      "grad_norm": 0.036819346249103546,
      "learning_rate": 2.841985713858752e-07,
      "loss": 0.0495,
      "step": 24000
    },
    {
      "epoch": 5.118027992479632,
      "grad_norm": 3.238111972808838,
      "learning_rate": 2.69254908993096e-07,
      "loss": 0.0507,
      "step": 24500
    },
    {
      "epoch": 5.222477543346564,
      "grad_norm": 0.4373936951160431,
      "learning_rate": 2.543112466003168e-07,
      "loss": 0.0499,
      "step": 25000
    },
    {
      "epoch": 5.326927094213495,
      "grad_norm": 0.06157074496150017,
      "learning_rate": 2.393675842075376e-07,
      "loss": 0.0481,
      "step": 25500
    },
    {
      "epoch": 5.431376645080427,
      "grad_norm": 0.5681213140487671,
      "learning_rate": 2.2442392181475835e-07,
      "loss": 0.048,
      "step": 26000
    },
    {
      "epoch": 5.535826195947357,
      "grad_norm": 0.2663944661617279,
      "learning_rate": 2.0948025942197914e-07,
      "loss": 0.0498,
      "step": 26500
    },
    {
      "epoch": 5.640275746814289,
      "grad_norm": 0.38863956928253174,
      "learning_rate": 1.945365970291999e-07,
      "loss": 0.0455,
      "step": 27000
    },
    {
      "epoch": 5.74472529768122,
      "grad_norm": 1.1037393808364868,
      "learning_rate": 1.795929346364207e-07,
      "loss": 0.0465,
      "step": 27500
    },
    {
      "epoch": 5.849174848548151,
      "grad_norm": 0.06277265399694443,
      "learning_rate": 1.6464927224364147e-07,
      "loss": 0.0443,
      "step": 28000
    },
    {
      "epoch": 5.953624399415083,
      "grad_norm": 0.3379976153373718,
      "learning_rate": 1.4970560985086224e-07,
      "loss": 0.0457,
      "step": 28500
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9869570326270025,
      "eval_loss": 0.04241248220205307,
      "eval_model_preparation_time": 0.0053,
      "eval_runtime": 1029.9456,
      "eval_samples_per_second": 99.155,
      "eval_steps_per_second": 12.395,
      "step": 28722
    },
    {
      "epoch": 6.058073950282013,
      "grad_norm": 4.613526821136475,
      "learning_rate": 1.3476194745808303e-07,
      "loss": 0.0428,
      "step": 29000
    },
    {
      "epoch": 6.162523501148945,
      "grad_norm": 0.2739298641681671,
      "learning_rate": 1.198182850653038e-07,
      "loss": 0.0474,
      "step": 29500
    },
    {
      "epoch": 6.266973052015876,
      "grad_norm": 0.4624876379966736,
      "learning_rate": 1.0487462267252457e-07,
      "loss": 0.0526,
      "step": 30000
    },
    {
      "epoch": 6.371422602882808,
      "grad_norm": 1.4040108919143677,
      "learning_rate": 8.993096027974535e-08,
      "loss": 0.0449,
      "step": 30500
    },
    {
      "epoch": 6.475872153749739,
      "grad_norm": 0.09033941477537155,
      "learning_rate": 7.498729788696613e-08,
      "loss": 0.0515,
      "step": 31000
    },
    {
      "epoch": 6.58032170461667,
      "grad_norm": 0.250262588262558,
      "learning_rate": 6.004363549418691e-08,
      "loss": 0.0467,
      "step": 31500
    },
    {
      "epoch": 6.684771255483602,
      "grad_norm": 1.7938841581344604,
      "learning_rate": 4.509997310140769e-08,
      "loss": 0.0545,
      "step": 32000
    },
    {
      "epoch": 6.789220806350532,
      "grad_norm": 0.6938722729682922,
      "learning_rate": 3.015631070862847e-08,
      "loss": 0.0485,
      "step": 32500
    },
    {
      "epoch": 6.893670357217464,
      "grad_norm": 0.07981687039136887,
      "learning_rate": 1.5212648315849248e-08,
      "loss": 0.0471,
      "step": 33000
    },
    {
      "epoch": 6.998119908084395,
      "grad_norm": 1.794087290763855,
      "learning_rate": 2.6898592307002603e-10,
      "loss": 0.0441,
      "step": 33500
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9871137049077592,
      "eval_loss": 0.042002853006124496,
      "eval_model_preparation_time": 0.0053,
      "eval_runtime": 1039.9182,
      "eval_samples_per_second": 98.204,
      "eval_steps_per_second": 12.276,
      "step": 33509
    }
  ],
  "logging_steps": 500,
  "max_steps": 33509,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 8.309373055985163e+19,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}