{
  "best_metric": 1.164973497390747,
  "best_model_checkpoint": "facial_age_image_detection/checkpoint-4800",
  "epoch": 20.0,
  "eval_steps": 500,
  "global_step": 4800,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6013097448929723,
      "eval_loss": 1.182183861732483,
      "eval_runtime": 106.5296,
      "eval_samples_per_second": 96.039,
      "eval_steps_per_second": 3.004,
      "step": 240
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6016029713615483,
      "eval_loss": 1.1812384128570557,
      "eval_runtime": 103.6605,
      "eval_samples_per_second": 98.697,
      "eval_steps_per_second": 3.087,
      "step": 480
    },
    {
      "epoch": 2.08,
      "grad_norm": 11.414525032043457,
      "learning_rate": 4.526315789473684e-07,
      "loss": 0.7517,
      "step": 500
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.5973023164891017,
      "eval_loss": 1.1846318244934082,
      "eval_runtime": 104.3279,
      "eval_samples_per_second": 98.066,
      "eval_steps_per_second": 3.067,
      "step": 720
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6031668458606196,
      "eval_loss": 1.178046464920044,
      "eval_runtime": 105.5564,
      "eval_samples_per_second": 96.924,
      "eval_steps_per_second": 3.032,
      "step": 960
    },
    {
      "epoch": 4.17,
      "grad_norm": 10.3371000289917,
      "learning_rate": 4e-07,
      "loss": 0.7404,
      "step": 1000
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6057081419216108,
      "eval_loss": 1.1763097047805786,
      "eval_runtime": 104.8017,
      "eval_samples_per_second": 97.622,
      "eval_steps_per_second": 3.053,
      "step": 1200
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6061968527025706,
      "eval_loss": 1.173648715019226,
      "eval_runtime": 104.8949,
      "eval_samples_per_second": 97.536,
      "eval_steps_per_second": 3.051,
      "step": 1440
    },
    {
      "epoch": 6.25,
      "grad_norm": 11.105679512023926,
      "learning_rate": 3.4736842105263157e-07,
      "loss": 0.7343,
      "step": 1500
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.6017984556739322,
      "eval_loss": 1.175057053565979,
      "eval_runtime": 104.1925,
      "eval_samples_per_second": 98.193,
      "eval_steps_per_second": 3.071,
      "step": 1680
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.6026781350796598,
      "eval_loss": 1.1735320091247559,
      "eval_runtime": 103.0462,
      "eval_samples_per_second": 99.286,
      "eval_steps_per_second": 3.105,
      "step": 1920
    },
    {
      "epoch": 8.33,
      "grad_norm": 15.928546905517578,
      "learning_rate": 2.947368421052631e-07,
      "loss": 0.7275,
      "step": 2000
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.6044374938911152,
      "eval_loss": 1.1707645654678345,
      "eval_runtime": 103.3329,
      "eval_samples_per_second": 99.01,
      "eval_steps_per_second": 3.097,
      "step": 2160
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.6085426644511778,
      "eval_loss": 1.169615387916565,
      "eval_runtime": 104.1315,
      "eval_samples_per_second": 98.251,
      "eval_steps_per_second": 3.073,
      "step": 2400
    },
    {
      "epoch": 10.42,
      "grad_norm": 15.35641098022461,
      "learning_rate": 2.4210526315789473e-07,
      "loss": 0.7201,
      "step": 2500
    },
    {
      "epoch": 11.0,
      "eval_accuracy": 0.6064900791711465,
      "eval_loss": 1.1700137853622437,
      "eval_runtime": 106.578,
      "eval_samples_per_second": 95.995,
      "eval_steps_per_second": 3.002,
      "step": 2640
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.6087381487635617,
      "eval_loss": 1.1673282384872437,
      "eval_runtime": 102.521,
      "eval_samples_per_second": 99.794,
      "eval_steps_per_second": 3.121,
      "step": 2880
    },
    {
      "epoch": 12.5,
      "grad_norm": 10.890913963317871,
      "learning_rate": 1.8947368421052632e-07,
      "loss": 0.7144,
      "step": 3000
    },
    {
      "epoch": 13.0,
      "eval_accuracy": 0.6035578144853876,
      "eval_loss": 1.1682850122451782,
      "eval_runtime": 103.3961,
      "eval_samples_per_second": 98.95,
      "eval_steps_per_second": 3.095,
      "step": 3120
    },
    {
      "epoch": 14.0,
      "eval_accuracy": 0.6070765321082983,
      "eval_loss": 1.1674185991287231,
      "eval_runtime": 103.2219,
      "eval_samples_per_second": 99.117,
      "eval_steps_per_second": 3.1,
      "step": 3360
    },
    {
      "epoch": 14.58,
      "grad_norm": 11.558945655822754,
      "learning_rate": 1.368421052631579e-07,
      "loss": 0.7137,
      "step": 3500
    },
    {
      "epoch": 15.0,
      "eval_accuracy": 0.6073697585768743,
      "eval_loss": 1.1667180061340332,
      "eval_runtime": 103.2403,
      "eval_samples_per_second": 99.099,
      "eval_steps_per_second": 3.1,
      "step": 3600
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.6068810477959143,
      "eval_loss": 1.166371464729309,
      "eval_runtime": 103.7779,
      "eval_samples_per_second": 98.586,
      "eval_steps_per_second": 3.084,
      "step": 3840
    },
    {
      "epoch": 16.67,
      "grad_norm": 15.126260757446289,
      "learning_rate": 8.421052631578947e-08,
      "loss": 0.7093,
      "step": 4000
    },
    {
      "epoch": 17.0,
      "eval_accuracy": 0.6044374938911152,
      "eval_loss": 1.1674280166625977,
      "eval_runtime": 102.8109,
      "eval_samples_per_second": 99.513,
      "eval_steps_per_second": 3.113,
      "step": 4080
    },
    {
      "epoch": 18.0,
      "eval_accuracy": 0.6042420095787313,
      "eval_loss": 1.1664873361587524,
      "eval_runtime": 102.8553,
      "eval_samples_per_second": 99.47,
      "eval_steps_per_second": 3.111,
      "step": 4320
    },
    {
      "epoch": 18.75,
      "grad_norm": 14.217340469360352,
      "learning_rate": 3.157894736842105e-08,
      "loss": 0.7035,
      "step": 4500
    },
    {
      "epoch": 19.0,
      "eval_accuracy": 0.6062945948587626,
      "eval_loss": 1.1650714874267578,
      "eval_runtime": 103.4609,
      "eval_samples_per_second": 98.888,
      "eval_steps_per_second": 3.093,
      "step": 4560
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.6068810477959143,
      "eval_loss": 1.164973497390747,
      "eval_runtime": 102.8512,
      "eval_samples_per_second": 99.474,
      "eval_steps_per_second": 3.111,
      "step": 4800
    }
  ],
  "logging_steps": 500,
  "max_steps": 4800,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 20,
  "save_steps": 500,
  "total_flos": 2.3786767829022114e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}