{
  "best_metric": 1.1146559715270996,
  "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": 0.6217378555370932,
      "eval_loss": 1.1227059364318848,
      "eval_runtime": 116.8455,
      "eval_samples_per_second": 87.56,
      "eval_steps_per_second": 2.739,
      "step": 240
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6248656045352361,
      "eval_loss": 1.1205345392227173,
      "eval_runtime": 101.698,
      "eval_samples_per_second": 100.602,
      "eval_steps_per_second": 3.147,
      "step": 480
    },
    {
      "epoch": 2.08,
      "grad_norm": 14.197029113769531,
      "learning_rate": 2.7157894736842105e-07,
      "loss": 0.7372,
      "step": 500
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6255497996285798,
      "eval_loss": 1.1209293603897095,
      "eval_runtime": 102.6145,
      "eval_samples_per_second": 99.703,
      "eval_steps_per_second": 3.118,
      "step": 720
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6272114162838432,
      "eval_loss": 1.119669795036316,
      "eval_runtime": 101.8458,
      "eval_samples_per_second": 100.456,
      "eval_steps_per_second": 3.142,
      "step": 960
    },
    {
      "epoch": 4.17,
      "grad_norm": 13.897222518920898,
      "learning_rate": 2.4e-07,
      "loss": 0.7296,
      "step": 1000
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6242791515980842,
      "eval_loss": 1.1204010248184204,
      "eval_runtime": 103.9165,
      "eval_samples_per_second": 98.454,
      "eval_steps_per_second": 3.079,
      "step": 1200
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6246701202228521,
      "eval_loss": 1.1215792894363403,
      "eval_runtime": 106.6746,
      "eval_samples_per_second": 95.909,
      "eval_steps_per_second": 3.0,
      "step": 1440
    },
    {
      "epoch": 6.25,
      "grad_norm": 11.151677131652832,
      "learning_rate": 2.0842105263157896e-07,
      "loss": 0.72,
      "step": 1500
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.6236926986609325,
      "eval_loss": 1.1201025247573853,
      "eval_runtime": 102.583,
      "eval_samples_per_second": 99.734,
      "eval_steps_per_second": 3.119,
      "step": 1680
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.6247678623790441,
      "eval_loss": 1.1184368133544922,
      "eval_runtime": 102.2621,
      "eval_samples_per_second": 100.047,
      "eval_steps_per_second": 3.129,
      "step": 1920
    },
    {
      "epoch": 8.33,
      "grad_norm": 13.550016403198242,
      "learning_rate": 1.7684210526315786e-07,
      "loss": 0.7137,
      "step": 2000
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.625158831003812,
      "eval_loss": 1.1169142723083496,
      "eval_runtime": 101.4179,
      "eval_samples_per_second": 100.88,
      "eval_steps_per_second": 3.155,
      "step": 2160
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.6285798064705307,
      "eval_loss": 1.1156556606292725,
      "eval_runtime": 109.0217,
      "eval_samples_per_second": 93.844,
      "eval_steps_per_second": 2.935,
      "step": 2400
    },
    {
      "epoch": 10.42,
      "grad_norm": 17.68604850769043,
      "learning_rate": 1.4526315789473685e-07,
      "loss": 0.715,
      "step": 2500
    },
    {
      "epoch": 11.0,
      "eval_accuracy": 0.6254520574723879,
      "eval_loss": 1.1168755292892456,
      "eval_runtime": 112.108,
      "eval_samples_per_second": 91.26,
      "eval_steps_per_second": 2.854,
      "step": 2640
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.6256475417847718,
      "eval_loss": 1.1173280477523804,
      "eval_runtime": 111.2518,
      "eval_samples_per_second": 91.963,
      "eval_steps_per_second": 2.876,
      "step": 2880
    },
    {
      "epoch": 12.5,
      "grad_norm": 13.063650131225586,
      "learning_rate": 1.1368421052631579e-07,
      "loss": 0.7087,
      "step": 3000
    },
    {
      "epoch": 13.0,
      "eval_accuracy": 0.6273091584400352,
      "eval_loss": 1.116219162940979,
      "eval_runtime": 102.5044,
      "eval_samples_per_second": 99.81,
      "eval_steps_per_second": 3.122,
      "step": 3120
    },
    {
      "epoch": 14.0,
      "eval_accuracy": 0.6236926986609325,
      "eval_loss": 1.1165720224380493,
      "eval_runtime": 117.8614,
      "eval_samples_per_second": 86.805,
      "eval_steps_per_second": 2.715,
      "step": 3360
    },
    {
      "epoch": 14.58,
      "grad_norm": 22.076231002807617,
      "learning_rate": 8.210526315789474e-08,
      "loss": 0.705,
      "step": 3500
    },
    {
      "epoch": 15.0,
      "eval_accuracy": 0.6240836672857003,
      "eval_loss": 1.1172147989273071,
      "eval_runtime": 107.0013,
      "eval_samples_per_second": 95.616,
      "eval_steps_per_second": 2.991,
      "step": 3600
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.6259407682533477,
      "eval_loss": 1.1154069900512695,
      "eval_runtime": 107.8292,
      "eval_samples_per_second": 94.882,
      "eval_steps_per_second": 2.968,
      "step": 3840
    },
    {
      "epoch": 16.67,
      "grad_norm": 10.493719100952148,
      "learning_rate": 5.052631578947368e-08,
      "loss": 0.6994,
      "step": 4000
    },
    {
      "epoch": 17.0,
      "eval_accuracy": 0.6261362525657316,
      "eval_loss": 1.1150434017181396,
      "eval_runtime": 103.249,
      "eval_samples_per_second": 99.091,
      "eval_steps_per_second": 3.099,
      "step": 4080
    },
    {
      "epoch": 18.0,
      "eval_accuracy": 0.6241814094418923,
      "eval_loss": 1.1151177883148193,
      "eval_runtime": 102.8465,
      "eval_samples_per_second": 99.478,
      "eval_steps_per_second": 3.111,
      "step": 4320
    },
    {
      "epoch": 18.75,
      "grad_norm": 15.010687828063965,
      "learning_rate": 1.8947368421052632e-08,
      "loss": 0.7018,
      "step": 4500
    },
    {
      "epoch": 19.0,
      "eval_accuracy": 0.6252565731600039,
      "eval_loss": 1.1146559715270996,
      "eval_runtime": 104.246,
      "eval_samples_per_second": 98.143,
      "eval_steps_per_second": 3.07,
      "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
}