{
  "best_metric": 1.1997283697128296,
  "best_model_checkpoint": "facial_age_image_detection/checkpoint-4560",
  "epoch": 19.0,
  "eval_steps": 500,
  "global_step": 4560,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5973023164891017,
      "eval_loss": 1.2148407697677612,
      "eval_runtime": 107.0191,
      "eval_samples_per_second": 95.6,
      "eval_steps_per_second": 2.99,
      "step": 240
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6003323233310527,
      "eval_loss": 1.2124874591827393,
      "eval_runtime": 108.7931,
      "eval_samples_per_second": 94.041,
      "eval_steps_per_second": 2.941,
      "step": 480
    },
    {
      "epoch": 2.08,
      "grad_norm": 10.869218826293945,
      "learning_rate": 4.526315789473684e-07,
      "loss": 0.7792,
      "step": 500
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6001368390186688,
      "eval_loss": 1.2103016376495361,
      "eval_runtime": 108.2046,
      "eval_samples_per_second": 94.552,
      "eval_steps_per_second": 2.957,
      "step": 720
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.5932948880852311,
      "eval_loss": 1.2177255153656006,
      "eval_runtime": 108.1913,
      "eval_samples_per_second": 94.564,
      "eval_steps_per_second": 2.958,
      "step": 960
    },
    {
      "epoch": 4.17,
      "grad_norm": 7.94642972946167,
      "learning_rate": 4e-07,
      "loss": 0.7693,
      "step": 1000
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.5987684488319812,
      "eval_loss": 1.2087459564208984,
      "eval_runtime": 108.1422,
      "eval_samples_per_second": 94.607,
      "eval_steps_per_second": 2.959,
      "step": 1200
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.5989639331443651,
      "eval_loss": 1.2106623649597168,
      "eval_runtime": 108.6434,
      "eval_samples_per_second": 94.17,
      "eval_steps_per_second": 2.945,
      "step": 1440
    },
    {
      "epoch": 6.25,
      "grad_norm": 12.206405639648438,
      "learning_rate": 3.4736842105263157e-07,
      "loss": 0.7579,
      "step": 1500
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.5941745674909589,
      "eval_loss": 1.2116234302520752,
      "eval_runtime": 110.2341,
      "eval_samples_per_second": 92.812,
      "eval_steps_per_second": 2.903,
      "step": 1680
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.596520379239566,
      "eval_loss": 1.209428071975708,
      "eval_runtime": 109.2117,
      "eval_samples_per_second": 93.68,
      "eval_steps_per_second": 2.93,
      "step": 1920
    },
    {
      "epoch": 8.33,
      "grad_norm": 13.978309631347656,
      "learning_rate": 2.947368421052631e-07,
      "loss": 0.7564,
      "step": 2000
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.5945655361157267,
      "eval_loss": 1.207972764968872,
      "eval_runtime": 109.0595,
      "eval_samples_per_second": 93.811,
      "eval_steps_per_second": 2.934,
      "step": 2160
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.5972045743329097,
      "eval_loss": 1.2053894996643066,
      "eval_runtime": 108.2001,
      "eval_samples_per_second": 94.556,
      "eval_steps_per_second": 2.957,
      "step": 2400
    },
    {
      "epoch": 10.42,
      "grad_norm": 13.184745788574219,
      "learning_rate": 2.4210526315789473e-07,
      "loss": 0.7462,
      "step": 2500
    },
    {
      "epoch": 11.0,
      "eval_accuracy": 0.596520379239566,
      "eval_loss": 1.2063862085342407,
      "eval_runtime": 109.0351,
      "eval_samples_per_second": 93.832,
      "eval_steps_per_second": 2.935,
      "step": 2640
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.5967158635519499,
      "eval_loss": 1.205006718635559,
      "eval_runtime": 107.7737,
      "eval_samples_per_second": 94.93,
      "eval_steps_per_second": 2.969,
      "step": 2880
    },
    {
      "epoch": 12.5,
      "grad_norm": 13.500107765197754,
      "learning_rate": 1.8947368421052632e-07,
      "loss": 0.7444,
      "step": 3000
    },
    {
      "epoch": 13.0,
      "eval_accuracy": 0.5974000586452937,
      "eval_loss": 1.2029664516448975,
      "eval_runtime": 108.3584,
      "eval_samples_per_second": 94.418,
      "eval_steps_per_second": 2.953,
      "step": 3120
    },
    {
      "epoch": 14.0,
      "eval_accuracy": 0.5982797380510214,
      "eval_loss": 1.2019948959350586,
      "eval_runtime": 108.2177,
      "eval_samples_per_second": 94.541,
      "eval_steps_per_second": 2.957,
      "step": 3360
    },
    {
      "epoch": 14.58,
      "grad_norm": 12.818633079528809,
      "learning_rate": 1.368421052631579e-07,
      "loss": 0.7402,
      "step": 3500
    },
    {
      "epoch": 15.0,
      "eval_accuracy": 0.5969113478643339,
      "eval_loss": 1.2034169435501099,
      "eval_runtime": 107.9426,
      "eval_samples_per_second": 94.782,
      "eval_steps_per_second": 2.965,
      "step": 3600
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.5980842537386375,
      "eval_loss": 1.2004482746124268,
      "eval_runtime": 108.7494,
      "eval_samples_per_second": 94.079,
      "eval_steps_per_second": 2.943,
      "step": 3840
    },
    {
      "epoch": 16.67,
      "grad_norm": 10.575230598449707,
      "learning_rate": 8.421052631578947e-08,
      "loss": 0.7357,
      "step": 4000
    },
    {
      "epoch": 17.0,
      "eval_accuracy": 0.5988661909881732,
      "eval_loss": 1.2010480165481567,
      "eval_runtime": 108.3609,
      "eval_samples_per_second": 94.416,
      "eval_steps_per_second": 2.953,
      "step": 4080
    },
    {
      "epoch": 18.0,
      "eval_accuracy": 0.5970090900205258,
      "eval_loss": 1.2006783485412598,
      "eval_runtime": 108.1842,
      "eval_samples_per_second": 94.57,
      "eval_steps_per_second": 2.958,
      "step": 4320
    },
    {
      "epoch": 18.75,
      "grad_norm": 10.900801658630371,
      "learning_rate": 3.157894736842105e-08,
      "loss": 0.7376,
      "step": 4500
    },
    {
      "epoch": 19.0,
      "eval_accuracy": 0.5974000586452937,
      "eval_loss": 1.1997283697128296,
      "eval_runtime": 108.4624,
      "eval_samples_per_second": 94.328,
      "eval_steps_per_second": 2.95,
      "step": 4560
    }
  ],
  "logging_steps": 500,
  "max_steps": 4800,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 20,
  "save_steps": 500,
  "total_flos": 2.2597429437571006e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}