{
  "best_metric": 0.03063076362013817,
  "best_model_checkpoint": "/workspace/training_output/age-gender-vit-large-patch16-384-v1/checkpoint-12285",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 12285,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.2035002035002035,
      "grad_norm": 0.2408568114042282,
      "learning_rate": 1.9185999185999185e-05,
      "loss": 0.1496,
      "step": 500
    },
    {
      "epoch": 0.407000407000407,
      "grad_norm": 0.0075986310839653015,
      "learning_rate": 1.8371998371998375e-05,
      "loss": 0.0806,
      "step": 1000
    },
    {
      "epoch": 0.6105006105006106,
      "grad_norm": 0.008936968632042408,
      "learning_rate": 1.755799755799756e-05,
      "loss": 0.0718,
      "step": 1500
    },
    {
      "epoch": 0.814000814000814,
      "grad_norm": 0.008683616295456886,
      "learning_rate": 1.6743996743996746e-05,
      "loss": 0.084,
      "step": 2000
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.990484429065744,
      "eval_loss": 0.04229868948459625,
      "eval_runtime": 319.7158,
      "eval_samples_per_second": 10.847,
      "eval_steps_per_second": 1.357,
      "step": 2457
    },
    {
      "epoch": 1.0175010175010175,
      "grad_norm": 0.016035890206694603,
      "learning_rate": 1.5929995929995933e-05,
      "loss": 0.0496,
      "step": 2500
    },
    {
      "epoch": 1.221001221001221,
      "grad_norm": 0.0016629098681733012,
      "learning_rate": 1.5115995115995116e-05,
      "loss": 0.0182,
      "step": 3000
    },
    {
      "epoch": 1.4245014245014245,
      "grad_norm": 0.004986602813005447,
      "learning_rate": 1.4301994301994305e-05,
      "loss": 0.0408,
      "step": 3500
    },
    {
      "epoch": 1.6280016280016278,
      "grad_norm": 12.01570987701416,
      "learning_rate": 1.348799348799349e-05,
      "loss": 0.0297,
      "step": 4000
    },
    {
      "epoch": 1.8315018315018317,
      "grad_norm": 0.0010504246456548572,
      "learning_rate": 1.2673992673992674e-05,
      "loss": 0.0233,
      "step": 4500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9927912341407151,
      "eval_loss": 0.035268500447273254,
      "eval_runtime": 325.0624,
      "eval_samples_per_second": 10.669,
      "eval_steps_per_second": 1.335,
      "step": 4914
    },
    {
      "epoch": 2.035002035002035,
      "grad_norm": 0.0030849070753902197,
      "learning_rate": 1.1859991859991862e-05,
      "loss": 0.0205,
      "step": 5000
    },
    {
      "epoch": 2.2385022385022384,
      "grad_norm": 0.0012894818792119622,
      "learning_rate": 1.1045991045991047e-05,
      "loss": 0.0097,
      "step": 5500
    },
    {
      "epoch": 2.442002442002442,
      "grad_norm": 0.0003367273311596364,
      "learning_rate": 1.0231990231990233e-05,
      "loss": 0.0051,
      "step": 6000
    },
    {
      "epoch": 2.6455026455026456,
      "grad_norm": 0.0010951660806313157,
      "learning_rate": 9.417989417989418e-06,
      "loss": 0.012,
      "step": 6500
    },
    {
      "epoch": 2.849002849002849,
      "grad_norm": 0.006617186591029167,
      "learning_rate": 8.603988603988605e-06,
      "loss": 0.009,
      "step": 7000
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.994232987312572,
      "eval_loss": 0.040446773171424866,
      "eval_runtime": 328.2567,
      "eval_samples_per_second": 10.565,
      "eval_steps_per_second": 1.322,
      "step": 7371
    },
    {
      "epoch": 3.0525030525030523,
      "grad_norm": 0.0005593279493041337,
      "learning_rate": 7.78998778998779e-06,
      "loss": 0.0122,
      "step": 7500
    },
    {
      "epoch": 3.256003256003256,
      "grad_norm": 5.2912073442712426e-05,
      "learning_rate": 6.975986975986977e-06,
      "loss": 0.0036,
      "step": 8000
    },
    {
      "epoch": 3.4595034595034595,
      "grad_norm": 0.0004164891433902085,
      "learning_rate": 6.161986161986162e-06,
      "loss": 0.0056,
      "step": 8500
    },
    {
      "epoch": 3.663003663003663,
      "grad_norm": 0.00016138925275299698,
      "learning_rate": 5.347985347985348e-06,
      "loss": 0.0095,
      "step": 9000
    },
    {
      "epoch": 3.8665038665038667,
      "grad_norm": 0.0001516578340670094,
      "learning_rate": 4.533984533984534e-06,
      "loss": 0.0056,
      "step": 9500
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9950980392156863,
      "eval_loss": 0.031163498759269714,
      "eval_runtime": 316.2806,
      "eval_samples_per_second": 10.965,
      "eval_steps_per_second": 1.372,
      "step": 9828
    },
    {
      "epoch": 4.07000407000407,
      "grad_norm": 5.3938976634526625e-05,
      "learning_rate": 3.7199837199837202e-06,
      "loss": 0.0001,
      "step": 10000
    },
    {
      "epoch": 4.273504273504273,
      "grad_norm": 0.00023969257017597556,
      "learning_rate": 2.9059829059829063e-06,
      "loss": 0.0009,
      "step": 10500
    },
    {
      "epoch": 4.477004477004477,
      "grad_norm": 0.0021630951669067144,
      "learning_rate": 2.091982091982092e-06,
      "loss": 0.0026,
      "step": 11000
    },
    {
      "epoch": 4.68050468050468,
      "grad_norm": 0.0001431748241884634,
      "learning_rate": 1.2779812779812782e-06,
      "loss": 0.0017,
      "step": 11500
    },
    {
      "epoch": 4.884004884004884,
      "grad_norm": 0.00023103418061509728,
      "learning_rate": 4.63980463980464e-07,
      "loss": 0.0,
      "step": 12000
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9950980392156863,
      "eval_loss": 0.03063076362013817,
      "eval_runtime": 325.9677,
      "eval_samples_per_second": 10.639,
      "eval_steps_per_second": 1.331,
      "step": 12285
    },
    {
      "epoch": 5.0,
      "step": 12285,
      "total_flos": 7.920463730427888e+19,
      "train_loss": 0.026289946709994166,
      "train_runtime": 16154.4085,
      "train_samples_per_second": 6.083,
      "train_steps_per_second": 0.76
    }
  ],
  "logging_steps": 500,
  "max_steps": 12285,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "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": 7.920463730427888e+19,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}