{
  "best_metric": 1.0019118785858154,
  "best_model_checkpoint": "fairface_age_image_detection/checkpoint-9376",
  "epoch": 4.0,
  "eval_steps": 500,
  "global_step": 9376,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.21331058020477817,
      "grad_norm": 2.756800413131714,
      "learning_rate": 5.710486811065838e-06,
      "loss": 1.7206,
      "step": 500
    },
    {
      "epoch": 0.42662116040955633,
      "grad_norm": 3.085395097732544,
      "learning_rate": 5.3888054900278794e-06,
      "loss": 1.3151,
      "step": 1000
    },
    {
      "epoch": 0.6399317406143344,
      "grad_norm": 3.533961057662964,
      "learning_rate": 5.067124168989921e-06,
      "loss": 1.212,
      "step": 1500
    },
    {
      "epoch": 0.8532423208191127,
      "grad_norm": 5.493643283843994,
      "learning_rate": 4.745442847951962e-06,
      "loss": 1.145,
      "step": 2000
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5652,
      "eval_loss": 1.115075707435608,
      "eval_model_preparation_time": 0.0037,
      "eval_runtime": 903.327,
      "eval_samples_per_second": 11.07,
      "eval_steps_per_second": 1.384,
      "step": 2344
    },
    {
      "epoch": 1.0665529010238908,
      "grad_norm": 5.28899621963501,
      "learning_rate": 4.423761526914004e-06,
      "loss": 1.1006,
      "step": 2500
    },
    {
      "epoch": 1.2798634812286689,
      "grad_norm": 6.52961540222168,
      "learning_rate": 4.102080205876046e-06,
      "loss": 1.0519,
      "step": 3000
    },
    {
      "epoch": 1.493174061433447,
      "grad_norm": 10.652497291564941,
      "learning_rate": 3.7803988848380875e-06,
      "loss": 1.0292,
      "step": 3500
    },
    {
      "epoch": 1.7064846416382253,
      "grad_norm": 5.245483875274658,
      "learning_rate": 3.4587175638001284e-06,
      "loss": 1.016,
      "step": 4000
    },
    {
      "epoch": 1.9197952218430034,
      "grad_norm": 4.070522308349609,
      "learning_rate": 3.1370362427621706e-06,
      "loss": 1.0056,
      "step": 4500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5831,
      "eval_loss": 1.030907392501831,
      "eval_model_preparation_time": 0.0037,
      "eval_runtime": 917.4025,
      "eval_samples_per_second": 10.9,
      "eval_steps_per_second": 1.363,
      "step": 4688
    },
    {
      "epoch": 2.1331058020477816,
      "grad_norm": 6.568565845489502,
      "learning_rate": 2.815354921724212e-06,
      "loss": 0.9796,
      "step": 5000
    },
    {
      "epoch": 2.34641638225256,
      "grad_norm": 10.742039680480957,
      "learning_rate": 2.4936736006862537e-06,
      "loss": 0.9587,
      "step": 5500
    },
    {
      "epoch": 2.5597269624573378,
      "grad_norm": 4.3592424392700195,
      "learning_rate": 2.171992279648295e-06,
      "loss": 0.9455,
      "step": 6000
    },
    {
      "epoch": 2.773037542662116,
      "grad_norm": 5.696556091308594,
      "learning_rate": 1.8503109586103366e-06,
      "loss": 0.9363,
      "step": 6500
    },
    {
      "epoch": 2.986348122866894,
      "grad_norm": 5.746116638183594,
      "learning_rate": 1.5286296375723786e-06,
      "loss": 0.9347,
      "step": 7000
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.5883,
      "eval_loss": 1.0057746171951294,
      "eval_model_preparation_time": 0.0037,
      "eval_runtime": 926.1536,
      "eval_samples_per_second": 10.797,
      "eval_steps_per_second": 1.35,
      "step": 7032
    },
    {
      "epoch": 3.1996587030716723,
      "grad_norm": 4.218583106994629,
      "learning_rate": 1.20694831653442e-06,
      "loss": 0.9035,
      "step": 7500
    },
    {
      "epoch": 3.4129692832764507,
      "grad_norm": 6.724698543548584,
      "learning_rate": 8.852669954964614e-07,
      "loss": 0.9081,
      "step": 8000
    },
    {
      "epoch": 3.6262798634812285,
      "grad_norm": 6.658480167388916,
      "learning_rate": 5.635856744585031e-07,
      "loss": 0.905,
      "step": 8500
    },
    {
      "epoch": 3.839590443686007,
      "grad_norm": 4.382744789123535,
      "learning_rate": 2.4190435342054474e-07,
      "loss": 0.8969,
      "step": 9000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.5892,
      "eval_loss": 1.0019118785858154,
      "eval_model_preparation_time": 0.0037,
      "eval_runtime": 929.5968,
      "eval_samples_per_second": 10.757,
      "eval_steps_per_second": 1.345,
      "step": 9376
    }
  ],
  "logging_steps": 500,
  "max_steps": 9376,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "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": 2.32490554103808e+19,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}