{
  "best_metric": 0.8810105949470253,
  "best_model_checkpoint": "/kaggle/working/danbooru-effnet/checkpoint-776",
  "epoch": 9.971014492753623,
  "eval_steps": 500,
  "global_step": 860,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.12,
      "learning_rate": 5.8139534883720935e-06,
      "loss": 7.541,
      "step": 10
    },
    {
      "epoch": 0.23,
      "learning_rate": 1.1627906976744187e-05,
      "loss": 7.3979,
      "step": 20
    },
    {
      "epoch": 0.35,
      "learning_rate": 1.744186046511628e-05,
      "loss": 7.1218,
      "step": 30
    },
    {
      "epoch": 0.46,
      "learning_rate": 2.3255813953488374e-05,
      "loss": 6.7652,
      "step": 40
    },
    {
      "epoch": 0.58,
      "learning_rate": 2.9069767441860467e-05,
      "loss": 6.1784,
      "step": 50
    },
    {
      "epoch": 0.7,
      "learning_rate": 3.488372093023256e-05,
      "loss": 5.3812,
      "step": 60
    },
    {
      "epoch": 0.81,
      "learning_rate": 4.0697674418604655e-05,
      "loss": 4.271,
      "step": 70
    },
    {
      "epoch": 0.93,
      "learning_rate": 4.651162790697675e-05,
      "loss": 2.9823,
      "step": 80
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7734311328443357,
      "eval_loss": 1.3710215091705322,
      "eval_runtime": 24.4306,
      "eval_samples_per_second": 50.224,
      "eval_steps_per_second": 1.596,
      "step": 86
    },
    {
      "epoch": 1.04,
      "learning_rate": 4.9741602067183466e-05,
      "loss": 1.7243,
      "step": 90
    },
    {
      "epoch": 1.16,
      "learning_rate": 4.9095607235142123e-05,
      "loss": 0.9988,
      "step": 100
    },
    {
      "epoch": 1.28,
      "learning_rate": 4.8449612403100775e-05,
      "loss": 0.6847,
      "step": 110
    },
    {
      "epoch": 1.39,
      "learning_rate": 4.780361757105943e-05,
      "loss": 0.6244,
      "step": 120
    },
    {
      "epoch": 1.51,
      "learning_rate": 4.715762273901809e-05,
      "loss": 0.5514,
      "step": 130
    },
    {
      "epoch": 1.62,
      "learning_rate": 4.651162790697675e-05,
      "loss": 0.5363,
      "step": 140
    },
    {
      "epoch": 1.74,
      "learning_rate": 4.5865633074935406e-05,
      "loss": 0.5036,
      "step": 150
    },
    {
      "epoch": 1.86,
      "learning_rate": 4.521963824289406e-05,
      "loss": 0.5194,
      "step": 160
    },
    {
      "epoch": 1.97,
      "learning_rate": 4.4573643410852715e-05,
      "loss": 0.4826,
      "step": 170
    },
    {
      "epoch": 1.99,
      "eval_accuracy": 0.8378158109209454,
      "eval_loss": 0.4534945785999298,
      "eval_runtime": 13.0595,
      "eval_samples_per_second": 93.955,
      "eval_steps_per_second": 2.986,
      "step": 172
    },
    {
      "epoch": 2.09,
      "learning_rate": 4.392764857881137e-05,
      "loss": 0.4388,
      "step": 180
    },
    {
      "epoch": 2.2,
      "learning_rate": 4.328165374677003e-05,
      "loss": 0.4391,
      "step": 190
    },
    {
      "epoch": 2.32,
      "learning_rate": 4.263565891472868e-05,
      "loss": 0.44,
      "step": 200
    },
    {
      "epoch": 2.43,
      "learning_rate": 4.198966408268734e-05,
      "loss": 0.4375,
      "step": 210
    },
    {
      "epoch": 2.55,
      "learning_rate": 4.1343669250646e-05,
      "loss": 0.4578,
      "step": 220
    },
    {
      "epoch": 2.67,
      "learning_rate": 4.0697674418604655e-05,
      "loss": 0.4513,
      "step": 230
    },
    {
      "epoch": 2.78,
      "learning_rate": 4.005167958656331e-05,
      "loss": 0.4169,
      "step": 240
    },
    {
      "epoch": 2.9,
      "learning_rate": 3.9405684754521964e-05,
      "loss": 0.4562,
      "step": 250
    },
    {
      "epoch": 2.99,
      "eval_accuracy": 0.8524857375713122,
      "eval_loss": 0.38659510016441345,
      "eval_runtime": 13.3081,
      "eval_samples_per_second": 92.199,
      "eval_steps_per_second": 2.931,
      "step": 258
    },
    {
      "epoch": 3.01,
      "learning_rate": 3.875968992248062e-05,
      "loss": 0.4149,
      "step": 260
    },
    {
      "epoch": 3.13,
      "learning_rate": 3.811369509043928e-05,
      "loss": 0.4,
      "step": 270
    },
    {
      "epoch": 3.25,
      "learning_rate": 3.746770025839794e-05,
      "loss": 0.4126,
      "step": 280
    },
    {
      "epoch": 3.36,
      "learning_rate": 3.682170542635659e-05,
      "loss": 0.415,
      "step": 290
    },
    {
      "epoch": 3.48,
      "learning_rate": 3.617571059431525e-05,
      "loss": 0.4347,
      "step": 300
    },
    {
      "epoch": 3.59,
      "learning_rate": 3.5529715762273905e-05,
      "loss": 0.4009,
      "step": 310
    },
    {
      "epoch": 3.71,
      "learning_rate": 3.488372093023256e-05,
      "loss": 0.4073,
      "step": 320
    },
    {
      "epoch": 3.83,
      "learning_rate": 3.4237726098191214e-05,
      "loss": 0.3998,
      "step": 330
    },
    {
      "epoch": 3.94,
      "learning_rate": 3.359173126614987e-05,
      "loss": 0.4149,
      "step": 340
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8533007334963325,
      "eval_loss": 0.3689855635166168,
      "eval_runtime": 13.1979,
      "eval_samples_per_second": 92.969,
      "eval_steps_per_second": 2.955,
      "step": 345
    },
    {
      "epoch": 4.06,
      "learning_rate": 3.294573643410852e-05,
      "loss": 0.3678,
      "step": 350
    },
    {
      "epoch": 4.17,
      "learning_rate": 3.229974160206719e-05,
      "loss": 0.3866,
      "step": 360
    },
    {
      "epoch": 4.29,
      "learning_rate": 3.1653746770025845e-05,
      "loss": 0.3978,
      "step": 370
    },
    {
      "epoch": 4.41,
      "learning_rate": 3.1007751937984497e-05,
      "loss": 0.3898,
      "step": 380
    },
    {
      "epoch": 4.52,
      "learning_rate": 3.0361757105943154e-05,
      "loss": 0.3866,
      "step": 390
    },
    {
      "epoch": 4.64,
      "learning_rate": 2.971576227390181e-05,
      "loss": 0.376,
      "step": 400
    },
    {
      "epoch": 4.75,
      "learning_rate": 2.9069767441860467e-05,
      "loss": 0.3798,
      "step": 410
    },
    {
      "epoch": 4.87,
      "learning_rate": 2.842377260981912e-05,
      "loss": 0.411,
      "step": 420
    },
    {
      "epoch": 4.99,
      "learning_rate": 2.777777777777778e-05,
      "loss": 0.3943,
      "step": 430
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8687856560717196,
      "eval_loss": 0.34427791833877563,
      "eval_runtime": 13.0107,
      "eval_samples_per_second": 94.307,
      "eval_steps_per_second": 2.998,
      "step": 431
    },
    {
      "epoch": 5.1,
      "learning_rate": 2.7131782945736434e-05,
      "loss": 0.3535,
      "step": 440
    },
    {
      "epoch": 5.22,
      "learning_rate": 2.648578811369509e-05,
      "loss": 0.3842,
      "step": 450
    },
    {
      "epoch": 5.33,
      "learning_rate": 2.5839793281653746e-05,
      "loss": 0.353,
      "step": 460
    },
    {
      "epoch": 5.45,
      "learning_rate": 2.5193798449612404e-05,
      "loss": 0.3797,
      "step": 470
    },
    {
      "epoch": 5.57,
      "learning_rate": 2.4547803617571062e-05,
      "loss": 0.378,
      "step": 480
    },
    {
      "epoch": 5.68,
      "learning_rate": 2.3901808785529716e-05,
      "loss": 0.371,
      "step": 490
    },
    {
      "epoch": 5.8,
      "learning_rate": 2.3255813953488374e-05,
      "loss": 0.3416,
      "step": 500
    },
    {
      "epoch": 5.91,
      "learning_rate": 2.260981912144703e-05,
      "loss": 0.394,
      "step": 510
    },
    {
      "epoch": 5.99,
      "eval_accuracy": 0.8793806030969845,
      "eval_loss": 0.33624228835105896,
      "eval_runtime": 13.2837,
      "eval_samples_per_second": 92.369,
      "eval_steps_per_second": 2.936,
      "step": 517
    },
    {
      "epoch": 6.03,
      "learning_rate": 2.1963824289405686e-05,
      "loss": 0.3905,
      "step": 520
    },
    {
      "epoch": 6.14,
      "learning_rate": 2.131782945736434e-05,
      "loss": 0.3574,
      "step": 530
    },
    {
      "epoch": 6.26,
      "learning_rate": 2.0671834625323e-05,
      "loss": 0.3718,
      "step": 540
    },
    {
      "epoch": 6.38,
      "learning_rate": 2.0025839793281657e-05,
      "loss": 0.3361,
      "step": 550
    },
    {
      "epoch": 6.49,
      "learning_rate": 1.937984496124031e-05,
      "loss": 0.3186,
      "step": 560
    },
    {
      "epoch": 6.61,
      "learning_rate": 1.873385012919897e-05,
      "loss": 0.3667,
      "step": 570
    },
    {
      "epoch": 6.72,
      "learning_rate": 1.8087855297157624e-05,
      "loss": 0.36,
      "step": 580
    },
    {
      "epoch": 6.84,
      "learning_rate": 1.744186046511628e-05,
      "loss": 0.3945,
      "step": 590
    },
    {
      "epoch": 6.96,
      "learning_rate": 1.6795865633074936e-05,
      "loss": 0.3416,
      "step": 600
    },
    {
      "epoch": 6.99,
      "eval_accuracy": 0.8712306438467807,
      "eval_loss": 0.33915451169013977,
      "eval_runtime": 13.2457,
      "eval_samples_per_second": 92.634,
      "eval_steps_per_second": 2.944,
      "step": 603
    },
    {
      "epoch": 7.07,
      "learning_rate": 1.6149870801033594e-05,
      "loss": 0.3618,
      "step": 610
    },
    {
      "epoch": 7.19,
      "learning_rate": 1.5503875968992248e-05,
      "loss": 0.3628,
      "step": 620
    },
    {
      "epoch": 7.3,
      "learning_rate": 1.4857881136950904e-05,
      "loss": 0.3514,
      "step": 630
    },
    {
      "epoch": 7.42,
      "learning_rate": 1.421188630490956e-05,
      "loss": 0.3777,
      "step": 640
    },
    {
      "epoch": 7.54,
      "learning_rate": 1.3565891472868217e-05,
      "loss": 0.3799,
      "step": 650
    },
    {
      "epoch": 7.65,
      "learning_rate": 1.2919896640826873e-05,
      "loss": 0.3875,
      "step": 660
    },
    {
      "epoch": 7.77,
      "learning_rate": 1.2273901808785531e-05,
      "loss": 0.343,
      "step": 670
    },
    {
      "epoch": 7.88,
      "learning_rate": 1.1627906976744187e-05,
      "loss": 0.3477,
      "step": 680
    },
    {
      "epoch": 8.0,
      "learning_rate": 1.0981912144702843e-05,
      "loss": 0.3644,
      "step": 690
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8753056234718827,
      "eval_loss": 0.33345112204551697,
      "eval_runtime": 13.3299,
      "eval_samples_per_second": 92.048,
      "eval_steps_per_second": 2.926,
      "step": 690
    },
    {
      "epoch": 8.12,
      "learning_rate": 1.03359173126615e-05,
      "loss": 0.3701,
      "step": 700
    },
    {
      "epoch": 8.23,
      "learning_rate": 9.689922480620156e-06,
      "loss": 0.3491,
      "step": 710
    },
    {
      "epoch": 8.35,
      "learning_rate": 9.043927648578812e-06,
      "loss": 0.3414,
      "step": 720
    },
    {
      "epoch": 8.46,
      "learning_rate": 8.397932816537468e-06,
      "loss": 0.3403,
      "step": 730
    },
    {
      "epoch": 8.58,
      "learning_rate": 7.751937984496124e-06,
      "loss": 0.3305,
      "step": 740
    },
    {
      "epoch": 8.7,
      "learning_rate": 7.10594315245478e-06,
      "loss": 0.3412,
      "step": 750
    },
    {
      "epoch": 8.81,
      "learning_rate": 6.4599483204134365e-06,
      "loss": 0.3633,
      "step": 760
    },
    {
      "epoch": 8.93,
      "learning_rate": 5.8139534883720935e-06,
      "loss": 0.3581,
      "step": 770
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.8810105949470253,
      "eval_loss": 0.32285985350608826,
      "eval_runtime": 13.3317,
      "eval_samples_per_second": 92.036,
      "eval_steps_per_second": 2.925,
      "step": 776
    },
    {
      "epoch": 9.04,
      "learning_rate": 5.16795865633075e-06,
      "loss": 0.2977,
      "step": 780
    },
    {
      "epoch": 9.16,
      "learning_rate": 4.521963824289406e-06,
      "loss": 0.3475,
      "step": 790
    },
    {
      "epoch": 9.28,
      "learning_rate": 3.875968992248062e-06,
      "loss": 0.3262,
      "step": 800
    },
    {
      "epoch": 9.39,
      "learning_rate": 3.2299741602067182e-06,
      "loss": 0.3532,
      "step": 810
    },
    {
      "epoch": 9.51,
      "learning_rate": 2.583979328165375e-06,
      "loss": 0.353,
      "step": 820
    },
    {
      "epoch": 9.62,
      "learning_rate": 1.937984496124031e-06,
      "loss": 0.3606,
      "step": 830
    },
    {
      "epoch": 9.74,
      "learning_rate": 1.2919896640826874e-06,
      "loss": 0.3519,
      "step": 840
    },
    {
      "epoch": 9.86,
      "learning_rate": 6.459948320413437e-07,
      "loss": 0.3531,
      "step": 850
    },
    {
      "epoch": 9.97,
      "learning_rate": 0.0,
      "loss": 0.3414,
      "step": 860
    },
    {
      "epoch": 9.97,
      "eval_accuracy": 0.8744906275468622,
      "eval_loss": 0.3368578255176544,
      "eval_runtime": 14.4016,
      "eval_samples_per_second": 85.199,
      "eval_steps_per_second": 2.708,
      "step": 860
    },
    {
      "epoch": 9.97,
      "step": 860,
      "total_flos": 2.1815473692672e+18,
      "train_loss": 0.9341721257498098,
      "train_runtime": 2440.787,
      "train_samples_per_second": 45.231,
      "train_steps_per_second": 0.352
    }
  ],
  "logging_steps": 10,
  "max_steps": 860,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 2.1815473692672e+18,
  "trial_name": null,
  "trial_params": null
}