{
  "best_metric": 1.143545389175415,
  "best_model_checkpoint": "facial_age_image_detection/checkpoint-7200",
  "epoch": 30.0,
  "eval_steps": 500,
  "global_step": 7200,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6146026781350796,
      "eval_loss": 1.1694062948226929,
      "eval_runtime": 105.8046,
      "eval_samples_per_second": 96.697,
      "eval_steps_per_second": 3.024,
      "step": 240
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6155800996969993,
      "eval_loss": 1.168785572052002,
      "eval_runtime": 110.1471,
      "eval_samples_per_second": 92.885,
      "eval_steps_per_second": 2.905,
      "step": 480
    },
    {
      "epoch": 2.08,
      "grad_norm": 8.204376220703125,
      "learning_rate": 4.685314685314685e-07,
      "loss": 0.7138,
      "step": 500
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6099110546378653,
      "eval_loss": 1.1679891347885132,
      "eval_runtime": 102.9363,
      "eval_samples_per_second": 99.392,
      "eval_steps_per_second": 3.109,
      "step": 720
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6177304271332226,
      "eval_loss": 1.1608326435089111,
      "eval_runtime": 101.8768,
      "eval_samples_per_second": 100.425,
      "eval_steps_per_second": 3.141,
      "step": 960
    },
    {
      "epoch": 4.17,
      "grad_norm": 14.190340042114258,
      "learning_rate": 4.335664335664335e-07,
      "loss": 0.7034,
      "step": 1000
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.6098133124816734,
      "eval_loss": 1.1647840738296509,
      "eval_runtime": 103.2275,
      "eval_samples_per_second": 99.111,
      "eval_steps_per_second": 3.1,
      "step": 1200
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6123546085426644,
      "eval_loss": 1.1614784002304077,
      "eval_runtime": 106.5014,
      "eval_samples_per_second": 96.064,
      "eval_steps_per_second": 3.005,
      "step": 1440
    },
    {
      "epoch": 6.25,
      "grad_norm": 18.558853149414062,
      "learning_rate": 3.9860139860139855e-07,
      "loss": 0.6951,
      "step": 1500
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.6186101065389502,
      "eval_loss": 1.160543441772461,
      "eval_runtime": 105.023,
      "eval_samples_per_second": 97.417,
      "eval_steps_per_second": 3.047,
      "step": 1680
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.6172417163522628,
      "eval_loss": 1.1572130918502808,
      "eval_runtime": 105.5204,
      "eval_samples_per_second": 96.958,
      "eval_steps_per_second": 3.033,
      "step": 1920
    },
    {
      "epoch": 8.33,
      "grad_norm": 9.678888320922852,
      "learning_rate": 3.636363636363636e-07,
      "loss": 0.6885,
      "step": 2000
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.6171439741960708,
      "eval_loss": 1.1544899940490723,
      "eval_runtime": 102.6064,
      "eval_samples_per_second": 99.711,
      "eval_steps_per_second": 3.119,
      "step": 2160
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.6179259114456065,
      "eval_loss": 1.155420184135437,
      "eval_runtime": 104.82,
      "eval_samples_per_second": 97.605,
      "eval_steps_per_second": 3.053,
      "step": 2400
    },
    {
      "epoch": 10.42,
      "grad_norm": 13.007719039916992,
      "learning_rate": 3.286713286713286e-07,
      "loss": 0.6806,
      "step": 2500
    },
    {
      "epoch": 11.0,
      "eval_accuracy": 0.6176326849770306,
      "eval_loss": 1.1572375297546387,
      "eval_runtime": 103.6156,
      "eval_samples_per_second": 98.74,
      "eval_steps_per_second": 3.088,
      "step": 2640
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.6146026781350796,
      "eval_loss": 1.1544034481048584,
      "eval_runtime": 102.3651,
      "eval_samples_per_second": 99.946,
      "eval_steps_per_second": 3.126,
      "step": 2880
    },
    {
      "epoch": 12.5,
      "grad_norm": 12.4281005859375,
      "learning_rate": 2.9370629370629366e-07,
      "loss": 0.6767,
      "step": 3000
    },
    {
      "epoch": 13.0,
      "eval_accuracy": 0.6182191379141824,
      "eval_loss": 1.1541900634765625,
      "eval_runtime": 105.249,
      "eval_samples_per_second": 97.208,
      "eval_steps_per_second": 3.04,
      "step": 3120
    },
    {
      "epoch": 14.0,
      "eval_accuracy": 0.6143094516665037,
      "eval_loss": 1.1535159349441528,
      "eval_runtime": 106.0372,
      "eval_samples_per_second": 96.485,
      "eval_steps_per_second": 3.018,
      "step": 3360
    },
    {
      "epoch": 14.58,
      "grad_norm": 7.5727410316467285,
      "learning_rate": 2.5874125874125877e-07,
      "loss": 0.667,
      "step": 3500
    },
    {
      "epoch": 15.0,
      "eval_accuracy": 0.6190988173199101,
      "eval_loss": 1.1515381336212158,
      "eval_runtime": 102.4541,
      "eval_samples_per_second": 99.859,
      "eval_steps_per_second": 3.123,
      "step": 3600
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.6183168800703743,
      "eval_loss": 1.1484887599945068,
      "eval_runtime": 111.0745,
      "eval_samples_per_second": 92.109,
      "eval_steps_per_second": 2.881,
      "step": 3840
    },
    {
      "epoch": 16.67,
      "grad_norm": 10.47463607788086,
      "learning_rate": 2.2377622377622377e-07,
      "loss": 0.6652,
      "step": 4000
    },
    {
      "epoch": 17.0,
      "eval_accuracy": 0.6168507477274948,
      "eval_loss": 1.1494479179382324,
      "eval_runtime": 104.2245,
      "eval_samples_per_second": 98.163,
      "eval_steps_per_second": 3.07,
      "step": 4080
    },
    {
      "epoch": 18.0,
      "eval_accuracy": 0.6195875281008699,
      "eval_loss": 1.1498117446899414,
      "eval_runtime": 103.8649,
      "eval_samples_per_second": 98.503,
      "eval_steps_per_second": 3.081,
      "step": 4320
    },
    {
      "epoch": 18.75,
      "grad_norm": 15.560189247131348,
      "learning_rate": 1.888111888111888e-07,
      "loss": 0.6619,
      "step": 4500
    },
    {
      "epoch": 19.0,
      "eval_accuracy": 0.619489785944678,
      "eval_loss": 1.1467111110687256,
      "eval_runtime": 104.6948,
      "eval_samples_per_second": 97.722,
      "eval_steps_per_second": 3.057,
      "step": 4560
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.6170462320398789,
      "eval_loss": 1.1486930847167969,
      "eval_runtime": 105.6513,
      "eval_samples_per_second": 96.837,
      "eval_steps_per_second": 3.029,
      "step": 4800
    },
    {
      "epoch": 20.83,
      "grad_norm": 10.84726333618164,
      "learning_rate": 1.5384615384615385e-07,
      "loss": 0.6586,
      "step": 5000
    },
    {
      "epoch": 21.0,
      "eval_accuracy": 0.6190988173199101,
      "eval_loss": 1.1456117630004883,
      "eval_runtime": 102.8016,
      "eval_samples_per_second": 99.522,
      "eval_steps_per_second": 3.113,
      "step": 5040
    },
    {
      "epoch": 22.0,
      "eval_accuracy": 0.6169484898836868,
      "eval_loss": 1.145732045173645,
      "eval_runtime": 103.7561,
      "eval_samples_per_second": 98.606,
      "eval_steps_per_second": 3.084,
      "step": 5280
    },
    {
      "epoch": 22.92,
      "grad_norm": 10.03998851776123,
      "learning_rate": 1.1888111888111887e-07,
      "loss": 0.652,
      "step": 5500
    },
    {
      "epoch": 23.0,
      "eval_accuracy": 0.6170462320398789,
      "eval_loss": 1.145518183708191,
      "eval_runtime": 102.871,
      "eval_samples_per_second": 99.455,
      "eval_steps_per_second": 3.111,
      "step": 5520
    },
    {
      "epoch": 24.0,
      "eval_accuracy": 0.6180236536017985,
      "eval_loss": 1.1446188688278198,
      "eval_runtime": 101.7602,
      "eval_samples_per_second": 100.54,
      "eval_steps_per_second": 3.145,
      "step": 5760
    },
    {
      "epoch": 25.0,
      "grad_norm": 13.074454307556152,
      "learning_rate": 8.391608391608391e-08,
      "loss": 0.6536,
      "step": 6000
    },
    {
      "epoch": 25.0,
      "eval_accuracy": 0.619489785944678,
      "eval_loss": 1.1448231935501099,
      "eval_runtime": 101.7751,
      "eval_samples_per_second": 100.526,
      "eval_steps_per_second": 3.144,
      "step": 6000
    },
    {
      "epoch": 26.0,
      "eval_accuracy": 0.6186101065389502,
      "eval_loss": 1.1442244052886963,
      "eval_runtime": 102.0024,
      "eval_samples_per_second": 100.302,
      "eval_steps_per_second": 3.137,
      "step": 6240
    },
    {
      "epoch": 27.0,
      "eval_accuracy": 0.6180236536017985,
      "eval_loss": 1.1441760063171387,
      "eval_runtime": 102.5718,
      "eval_samples_per_second": 99.745,
      "eval_steps_per_second": 3.12,
      "step": 6480
    },
    {
      "epoch": 27.08,
      "grad_norm": 12.248133659362793,
      "learning_rate": 4.895104895104895e-08,
      "loss": 0.6512,
      "step": 6500
    },
    {
      "epoch": 28.0,
      "eval_accuracy": 0.6179259114456065,
      "eval_loss": 1.1439582109451294,
      "eval_runtime": 103.381,
      "eval_samples_per_second": 98.964,
      "eval_steps_per_second": 3.095,
      "step": 6720
    },
    {
      "epoch": 29.0,
      "eval_accuracy": 0.6181213957579904,
      "eval_loss": 1.1439974308013916,
      "eval_runtime": 102.6407,
      "eval_samples_per_second": 99.678,
      "eval_steps_per_second": 3.118,
      "step": 6960
    },
    {
      "epoch": 29.17,
      "grad_norm": 7.591431617736816,
      "learning_rate": 1.3986013986013985e-08,
      "loss": 0.6473,
      "step": 7000
    },
    {
      "epoch": 30.0,
      "eval_accuracy": 0.6190010751637182,
      "eval_loss": 1.143545389175415,
      "eval_runtime": 105.5925,
      "eval_samples_per_second": 96.891,
      "eval_steps_per_second": 3.031,
      "step": 7200
    }
  ],
  "logging_steps": 500,
  "max_steps": 7200,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 30,
  "save_steps": 500,
  "total_flos": 3.5680151743533187e+19,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}