{
  "best_metric": 0.046924859285354614,
  "best_model_checkpoint": "man_woman_face_image_detection/checkpoint-28722",
  "epoch": 6.0,
  "eval_steps": 500,
  "global_step": 28722,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.10444955086693128,
      "grad_norm": 0.17686675488948822,
      "learning_rate": 6.89013671875e-07,
      "loss": 0.1192,
      "step": 500
    },
    {
      "epoch": 0.20889910173386256,
      "grad_norm": 7.6029133796691895,
      "learning_rate": 6.76806640625e-07,
      "loss": 0.1074,
      "step": 1000
    },
    {
      "epoch": 0.3133486526007938,
      "grad_norm": 0.25385862588882446,
      "learning_rate": 6.645996093749999e-07,
      "loss": 0.0978,
      "step": 1500
    },
    {
      "epoch": 0.4177982034677251,
      "grad_norm": 1.1110018491744995,
      "learning_rate": 6.52392578125e-07,
      "loss": 0.0842,
      "step": 2000
    },
    {
      "epoch": 0.5222477543346563,
      "grad_norm": 0.5810139179229736,
      "learning_rate": 6.40185546875e-07,
      "loss": 0.0862,
      "step": 2500
    },
    {
      "epoch": 0.6266973052015876,
      "grad_norm": 5.968545436859131,
      "learning_rate": 6.279785156249999e-07,
      "loss": 0.0834,
      "step": 3000
    },
    {
      "epoch": 0.7311468560685189,
      "grad_norm": 2.3612782955169678,
      "learning_rate": 6.15771484375e-07,
      "loss": 0.0788,
      "step": 3500
    },
    {
      "epoch": 0.8355964069354502,
      "grad_norm": 2.7137603759765625,
      "learning_rate": 6.03564453125e-07,
      "loss": 0.0815,
      "step": 4000
    },
    {
      "epoch": 0.9400459578023814,
      "grad_norm": 2.546800374984741,
      "learning_rate": 5.91357421875e-07,
      "loss": 0.0714,
      "step": 4500
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.984920292977165,
      "eval_loss": 0.051657598465681076,
      "eval_model_preparation_time": 0.0052,
      "eval_runtime": 1021.407,
      "eval_samples_per_second": 99.984,
      "eval_steps_per_second": 12.498,
      "step": 4787
    },
    {
      "epoch": 1.0444955086693126,
      "grad_norm": 2.2125747203826904,
      "learning_rate": 5.791503906249999e-07,
      "loss": 0.0807,
      "step": 5000
    },
    {
      "epoch": 1.148945059536244,
      "grad_norm": 0.20245374739170074,
      "learning_rate": 5.66943359375e-07,
      "loss": 0.0743,
      "step": 5500
    },
    {
      "epoch": 1.2533946104031752,
      "grad_norm": 0.28860384225845337,
      "learning_rate": 5.54736328125e-07,
      "loss": 0.0719,
      "step": 6000
    },
    {
      "epoch": 1.3578441612701067,
      "grad_norm": 0.21754620969295502,
      "learning_rate": 5.425292968749999e-07,
      "loss": 0.0691,
      "step": 6500
    },
    {
      "epoch": 1.4622937121370378,
      "grad_norm": 3.9786083698272705,
      "learning_rate": 5.30322265625e-07,
      "loss": 0.0721,
      "step": 7000
    },
    {
      "epoch": 1.566743263003969,
      "grad_norm": 0.5568149089813232,
      "learning_rate": 5.18115234375e-07,
      "loss": 0.0728,
      "step": 7500
    },
    {
      "epoch": 1.6711928138709005,
      "grad_norm": 7.5547027587890625,
      "learning_rate": 5.059082031249999e-07,
      "loss": 0.0709,
      "step": 8000
    },
    {
      "epoch": 1.7756423647378317,
      "grad_norm": 3.4100685119628906,
      "learning_rate": 4.937011718749999e-07,
      "loss": 0.0689,
      "step": 8500
    },
    {
      "epoch": 1.8800919156047629,
      "grad_norm": 0.25586971640586853,
      "learning_rate": 4.81494140625e-07,
      "loss": 0.0686,
      "step": 9000
    },
    {
      "epoch": 1.9845414664716943,
      "grad_norm": 0.20959605276584625,
      "learning_rate": 4.69287109375e-07,
      "loss": 0.0632,
      "step": 9500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9853021816615095,
      "eval_loss": 0.04927506670355797,
      "eval_model_preparation_time": 0.0052,
      "eval_runtime": 1036.5941,
      "eval_samples_per_second": 98.519,
      "eval_steps_per_second": 12.315,
      "step": 9574
    },
    {
      "epoch": 2.0889910173386252,
      "grad_norm": 10.270435333251953,
      "learning_rate": 4.5708007812499997e-07,
      "loss": 0.0663,
      "step": 10000
    },
    {
      "epoch": 2.1934405682055567,
      "grad_norm": 4.9907331466674805,
      "learning_rate": 4.44873046875e-07,
      "loss": 0.0674,
      "step": 10500
    },
    {
      "epoch": 2.297890119072488,
      "grad_norm": 0.05801746994256973,
      "learning_rate": 4.32666015625e-07,
      "loss": 0.0619,
      "step": 11000
    },
    {
      "epoch": 2.402339669939419,
      "grad_norm": 0.07323583215475082,
      "learning_rate": 4.20458984375e-07,
      "loss": 0.0631,
      "step": 11500
    },
    {
      "epoch": 2.5067892208063505,
      "grad_norm": 1.5988435745239258,
      "learning_rate": 4.0825195312499994e-07,
      "loss": 0.0643,
      "step": 12000
    },
    {
      "epoch": 2.611238771673282,
      "grad_norm": 0.5066471695899963,
      "learning_rate": 3.96044921875e-07,
      "loss": 0.0713,
      "step": 12500
    },
    {
      "epoch": 2.7156883225402133,
      "grad_norm": 1.4767159223556519,
      "learning_rate": 3.83837890625e-07,
      "loss": 0.0727,
      "step": 13000
    },
    {
      "epoch": 2.8201378734071443,
      "grad_norm": 6.386782169342041,
      "learning_rate": 3.7163085937499995e-07,
      "loss": 0.0575,
      "step": 13500
    },
    {
      "epoch": 2.9245874242740757,
      "grad_norm": 1.1449916362762451,
      "learning_rate": 3.5942382812500003e-07,
      "loss": 0.0626,
      "step": 14000
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9854490619247189,
      "eval_loss": 0.048059992492198944,
      "eval_model_preparation_time": 0.0052,
      "eval_runtime": 1093.4649,
      "eval_samples_per_second": 93.395,
      "eval_steps_per_second": 11.675,
      "step": 14361
    },
    {
      "epoch": 3.0290369751410067,
      "grad_norm": 8.965994834899902,
      "learning_rate": 3.47216796875e-07,
      "loss": 0.061,
      "step": 14500
    },
    {
      "epoch": 3.133486526007938,
      "grad_norm": 0.24369557201862335,
      "learning_rate": 3.3500976562499996e-07,
      "loss": 0.0585,
      "step": 15000
    },
    {
      "epoch": 3.2379360768748695,
      "grad_norm": 2.182089328765869,
      "learning_rate": 3.22802734375e-07,
      "loss": 0.0603,
      "step": 15500
    },
    {
      "epoch": 3.342385627741801,
      "grad_norm": 8.162790298461914,
      "learning_rate": 3.1059570312499995e-07,
      "loss": 0.0608,
      "step": 16000
    },
    {
      "epoch": 3.446835178608732,
      "grad_norm": 3.445383071899414,
      "learning_rate": 2.98388671875e-07,
      "loss": 0.0662,
      "step": 16500
    },
    {
      "epoch": 3.5512847294756633,
      "grad_norm": 0.21630489826202393,
      "learning_rate": 2.86181640625e-07,
      "loss": 0.0619,
      "step": 17000
    },
    {
      "epoch": 3.6557342803425943,
      "grad_norm": 8.268571853637695,
      "learning_rate": 2.7397460937499996e-07,
      "loss": 0.0601,
      "step": 17500
    },
    {
      "epoch": 3.7601838312095257,
      "grad_norm": 1.3148396015167236,
      "learning_rate": 2.61767578125e-07,
      "loss": 0.0655,
      "step": 18000
    },
    {
      "epoch": 3.864633382076457,
      "grad_norm": 1.5834839344024658,
      "learning_rate": 2.4956054687499995e-07,
      "loss": 0.0589,
      "step": 18500
    },
    {
      "epoch": 3.9690829329433885,
      "grad_norm": 6.2100510597229,
      "learning_rate": 2.37353515625e-07,
      "loss": 0.0578,
      "step": 19000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9855959421879285,
      "eval_loss": 0.04732793942093849,
      "eval_model_preparation_time": 0.0052,
      "eval_runtime": 1129.7256,
      "eval_samples_per_second": 90.397,
      "eval_steps_per_second": 11.3,
      "step": 19148
    },
    {
      "epoch": 4.07353248381032,
      "grad_norm": 0.3075694441795349,
      "learning_rate": 2.2514648437499997e-07,
      "loss": 0.0588,
      "step": 19500
    },
    {
      "epoch": 4.1779820346772505,
      "grad_norm": 5.955320358276367,
      "learning_rate": 2.12939453125e-07,
      "loss": 0.0631,
      "step": 20000
    },
    {
      "epoch": 4.282431585544182,
      "grad_norm": 0.7760459780693054,
      "learning_rate": 2.00732421875e-07,
      "loss": 0.0601,
      "step": 20500
    },
    {
      "epoch": 4.386881136411113,
      "grad_norm": 6.535593032836914,
      "learning_rate": 1.8852539062499998e-07,
      "loss": 0.0598,
      "step": 21000
    },
    {
      "epoch": 4.491330687278045,
      "grad_norm": 1.7667866945266724,
      "learning_rate": 1.76318359375e-07,
      "loss": 0.0562,
      "step": 21500
    },
    {
      "epoch": 4.595780238144976,
      "grad_norm": 1.9524359703063965,
      "learning_rate": 1.64111328125e-07,
      "loss": 0.0615,
      "step": 22000
    },
    {
      "epoch": 4.700229789011908,
      "grad_norm": 0.6181924939155579,
      "learning_rate": 1.5190429687499999e-07,
      "loss": 0.0602,
      "step": 22500
    },
    {
      "epoch": 4.804679339878838,
      "grad_norm": 4.446329593658447,
      "learning_rate": 1.39697265625e-07,
      "loss": 0.0604,
      "step": 23000
    },
    {
      "epoch": 4.9091288907457695,
      "grad_norm": 12.356111526489258,
      "learning_rate": 1.27490234375e-07,
      "loss": 0.06,
      "step": 23500
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9857330304335905,
      "eval_loss": 0.0469873771071434,
      "eval_model_preparation_time": 0.0052,
      "eval_runtime": 1092.9703,
      "eval_samples_per_second": 93.437,
      "eval_steps_per_second": 11.68,
      "step": 23935
    },
    {
      "epoch": 5.013578441612701,
      "grad_norm": 0.1434878706932068,
      "learning_rate": 1.15283203125e-07,
      "loss": 0.0525,
      "step": 24000
    },
    {
      "epoch": 5.118027992479632,
      "grad_norm": 0.09600865095853806,
      "learning_rate": 1.0307617187499999e-07,
      "loss": 0.0555,
      "step": 24500
    },
    {
      "epoch": 5.222477543346564,
      "grad_norm": 1.8430639505386353,
      "learning_rate": 9.086914062499999e-08,
      "loss": 0.066,
      "step": 25000
    },
    {
      "epoch": 5.326927094213495,
      "grad_norm": 3.1898910999298096,
      "learning_rate": 7.866210937499999e-08,
      "loss": 0.0569,
      "step": 25500
    },
    {
      "epoch": 5.431376645080427,
      "grad_norm": 4.9123854637146,
      "learning_rate": 6.645507812499999e-08,
      "loss": 0.0618,
      "step": 26000
    },
    {
      "epoch": 5.535826195947357,
      "grad_norm": 2.0142762660980225,
      "learning_rate": 5.4248046875e-08,
      "loss": 0.0592,
      "step": 26500
    },
    {
      "epoch": 5.640275746814289,
      "grad_norm": 0.04716000333428383,
      "learning_rate": 4.2041015624999996e-08,
      "loss": 0.0558,
      "step": 27000
    },
    {
      "epoch": 5.74472529768122,
      "grad_norm": 0.09332660585641861,
      "learning_rate": 2.9833984375000004e-08,
      "loss": 0.0583,
      "step": 27500
    },
    {
      "epoch": 5.849174848548151,
      "grad_norm": 0.10846803337335587,
      "learning_rate": 1.7626953124999998e-08,
      "loss": 0.06,
      "step": 28000
    },
    {
      "epoch": 5.953624399415083,
      "grad_norm": 1.7731516361236572,
      "learning_rate": 5.419921875e-09,
      "loss": 0.055,
      "step": 28500
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.985811366573969,
      "eval_loss": 0.046924859285354614,
      "eval_model_preparation_time": 0.0052,
      "eval_runtime": 964.6086,
      "eval_samples_per_second": 105.871,
      "eval_steps_per_second": 13.234,
      "step": 28722
    }
  ],
  "logging_steps": 500,
  "max_steps": 28722,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 6,
  "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": 7.1223197622729966e+19,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}