{
  "best_metric": 1.257996678352356,
  "best_model_checkpoint": "facial_age_image_detection/checkpoint-5320",
  "epoch": 19.0,
  "eval_steps": 500,
  "global_step": 5320,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5756548937833963,
      "eval_loss": 1.3048689365386963,
      "eval_runtime": 87.7602,
      "eval_samples_per_second": 87.431,
      "eval_steps_per_second": 2.735,
      "step": 280
    },
    {
      "epoch": 1.79,
      "grad_norm": 11.91701889038086,
      "learning_rate": 9.18918918918919e-07,
      "loss": 0.9042,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.5678352665189625,
      "eval_loss": 1.306204080581665,
      "eval_runtime": 81.8404,
      "eval_samples_per_second": 93.756,
      "eval_steps_per_second": 2.933,
      "step": 560
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.576176202267692,
      "eval_loss": 1.300585150718689,
      "eval_runtime": 81.0324,
      "eval_samples_per_second": 94.691,
      "eval_steps_per_second": 2.962,
      "step": 840
    },
    {
      "epoch": 3.57,
      "grad_norm": 12.9658784866333,
      "learning_rate": 8.288288288288288e-07,
      "loss": 0.8924,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.5730483513619185,
      "eval_loss": 1.2988545894622803,
      "eval_runtime": 82.0516,
      "eval_samples_per_second": 93.514,
      "eval_steps_per_second": 2.925,
      "step": 1120
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.5746122768148052,
      "eval_loss": 1.2918657064437866,
      "eval_runtime": 85.7442,
      "eval_samples_per_second": 89.487,
      "eval_steps_per_second": 2.799,
      "step": 1400
    },
    {
      "epoch": 5.36,
      "grad_norm": 11.397974014282227,
      "learning_rate": 7.387387387387387e-07,
      "loss": 0.8708,
      "step": 1500
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.5765671836309136,
      "eval_loss": 1.2833993434906006,
      "eval_runtime": 80.6066,
      "eval_samples_per_second": 95.191,
      "eval_steps_per_second": 2.977,
      "step": 1680
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.5842564837742734,
      "eval_loss": 1.276853322982788,
      "eval_runtime": 82.4272,
      "eval_samples_per_second": 93.088,
      "eval_steps_per_second": 2.912,
      "step": 1960
    },
    {
      "epoch": 7.14,
      "grad_norm": 7.821016311645508,
      "learning_rate": 6.486486486486486e-07,
      "loss": 0.8563,
      "step": 2000
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.5781311090838004,
      "eval_loss": 1.2758774757385254,
      "eval_runtime": 81.3168,
      "eval_samples_per_second": 94.359,
      "eval_steps_per_second": 2.951,
      "step": 2240
    },
    {
      "epoch": 8.93,
      "grad_norm": 9.624176025390625,
      "learning_rate": 5.585585585585585e-07,
      "loss": 0.8478,
      "step": 2500
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.5789130718102438,
      "eval_loss": 1.2749879360198975,
      "eval_runtime": 80.8906,
      "eval_samples_per_second": 94.856,
      "eval_steps_per_second": 2.967,
      "step": 2520
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.582301576958165,
      "eval_loss": 1.270504355430603,
      "eval_runtime": 83.0139,
      "eval_samples_per_second": 92.43,
      "eval_steps_per_second": 2.891,
      "step": 2800
    },
    {
      "epoch": 10.71,
      "grad_norm": 7.522056579589844,
      "learning_rate": 4.684684684684684e-07,
      "loss": 0.8341,
      "step": 3000
    },
    {
      "epoch": 11.0,
      "eval_accuracy": 0.5815196142317216,
      "eval_loss": 1.2686316967010498,
      "eval_runtime": 81.0469,
      "eval_samples_per_second": 94.674,
      "eval_steps_per_second": 2.961,
      "step": 3080
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.5804769972631305,
      "eval_loss": 1.2680425643920898,
      "eval_runtime": 80.5295,
      "eval_samples_per_second": 95.282,
      "eval_steps_per_second": 2.98,
      "step": 3360
    },
    {
      "epoch": 12.5,
      "grad_norm": 14.59212589263916,
      "learning_rate": 3.783783783783784e-07,
      "loss": 0.8298,
      "step": 3500
    },
    {
      "epoch": 13.0,
      "eval_accuracy": 0.5786524175680959,
      "eval_loss": 1.2673722505569458,
      "eval_runtime": 84.1895,
      "eval_samples_per_second": 91.14,
      "eval_steps_per_second": 2.851,
      "step": 3640
    },
    {
      "epoch": 14.0,
      "eval_accuracy": 0.5802163430209827,
      "eval_loss": 1.2637258768081665,
      "eval_runtime": 82.7707,
      "eval_samples_per_second": 92.702,
      "eval_steps_per_second": 2.9,
      "step": 3920
    },
    {
      "epoch": 14.29,
      "grad_norm": 11.842473030090332,
      "learning_rate": 2.882882882882883e-07,
      "loss": 0.8226,
      "step": 4000
    },
    {
      "epoch": 15.0,
      "eval_accuracy": 0.5813892871106477,
      "eval_loss": 1.2624099254608154,
      "eval_runtime": 82.4012,
      "eval_samples_per_second": 93.118,
      "eval_steps_per_second": 2.913,
      "step": 4200
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.5829532125635345,
      "eval_loss": 1.2608647346496582,
      "eval_runtime": 82.9157,
      "eval_samples_per_second": 92.54,
      "eval_steps_per_second": 2.895,
      "step": 4480
    },
    {
      "epoch": 16.07,
      "grad_norm": 7.111355304718018,
      "learning_rate": 1.9819819819819818e-07,
      "loss": 0.8104,
      "step": 4500
    },
    {
      "epoch": 17.0,
      "eval_accuracy": 0.5825622312003128,
      "eval_loss": 1.2595696449279785,
      "eval_runtime": 87.9608,
      "eval_samples_per_second": 87.232,
      "eval_steps_per_second": 2.728,
      "step": 4760
    },
    {
      "epoch": 17.86,
      "grad_norm": 10.99641227722168,
      "learning_rate": 1.0810810810810811e-07,
      "loss": 0.8098,
      "step": 5000
    },
    {
      "epoch": 18.0,
      "eval_accuracy": 0.5826925583213867,
      "eval_loss": 1.2582371234893799,
      "eval_runtime": 83.3175,
      "eval_samples_per_second": 92.093,
      "eval_steps_per_second": 2.881,
      "step": 5040
    },
    {
      "epoch": 19.0,
      "eval_accuracy": 0.5821712498370911,
      "eval_loss": 1.257996678352356,
      "eval_runtime": 83.5381,
      "eval_samples_per_second": 91.85,
      "eval_steps_per_second": 2.873,
      "step": 5320
    }
  ],
  "logging_steps": 500,
  "max_steps": 5600,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 20,
  "save_steps": 500,
  "total_flos": 2.636440398962747e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}