{
  "best_metric": 1.0297563076019287,
  "best_model_checkpoint": "pokemon_types_image_detection/checkpoint-8350",
  "epoch": 50.0,
  "eval_steps": 500,
  "global_step": 8350,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.06616541353383458,
      "eval_loss": 4.913679122924805,
      "eval_runtime": 17.1752,
      "eval_samples_per_second": 77.437,
      "eval_steps_per_second": 9.723,
      "step": 167
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.3090225563909774,
      "eval_loss": 4.762322902679443,
      "eval_runtime": 18.0222,
      "eval_samples_per_second": 73.798,
      "eval_steps_per_second": 9.266,
      "step": 334
    },
    {
      "epoch": 2.99,
      "learning_rate": 9.457831325301205e-06,
      "loss": 4.8044,
      "step": 500
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.562406015037594,
      "eval_loss": 4.578239440917969,
      "eval_runtime": 17.9833,
      "eval_samples_per_second": 73.957,
      "eval_steps_per_second": 9.286,
      "step": 501
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7180451127819549,
      "eval_loss": 4.36922025680542,
      "eval_runtime": 17.5366,
      "eval_samples_per_second": 75.842,
      "eval_steps_per_second": 9.523,
      "step": 668
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8285714285714286,
      "eval_loss": 4.163111209869385,
      "eval_runtime": 17.4507,
      "eval_samples_per_second": 76.215,
      "eval_steps_per_second": 9.57,
      "step": 835
    },
    {
      "epoch": 5.99,
      "learning_rate": 8.855421686746989e-06,
      "loss": 4.1907,
      "step": 1000
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.856390977443609,
      "eval_loss": 3.971830129623413,
      "eval_runtime": 17.1602,
      "eval_samples_per_second": 77.505,
      "eval_steps_per_second": 9.732,
      "step": 1002
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8827067669172932,
      "eval_loss": 3.7871174812316895,
      "eval_runtime": 17.4448,
      "eval_samples_per_second": 76.241,
      "eval_steps_per_second": 9.573,
      "step": 1169
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9030075187969925,
      "eval_loss": 3.613436222076416,
      "eval_runtime": 17.2789,
      "eval_samples_per_second": 76.973,
      "eval_steps_per_second": 9.665,
      "step": 1336
    },
    {
      "epoch": 8.98,
      "learning_rate": 8.253012048192773e-06,
      "loss": 3.5887,
      "step": 1500
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.9142857142857143,
      "eval_loss": 3.453962802886963,
      "eval_runtime": 16.8987,
      "eval_samples_per_second": 78.704,
      "eval_steps_per_second": 9.882,
      "step": 1503
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.924812030075188,
      "eval_loss": 3.299895763397217,
      "eval_runtime": 16.7986,
      "eval_samples_per_second": 79.173,
      "eval_steps_per_second": 9.941,
      "step": 1670
    },
    {
      "epoch": 11.0,
      "eval_accuracy": 0.9345864661654135,
      "eval_loss": 3.156660795211792,
      "eval_runtime": 16.7869,
      "eval_samples_per_second": 79.228,
      "eval_steps_per_second": 9.948,
      "step": 1837
    },
    {
      "epoch": 11.98,
      "learning_rate": 7.650602409638555e-06,
      "loss": 3.0841,
      "step": 2000
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.937593984962406,
      "eval_loss": 3.020425319671631,
      "eval_runtime": 16.7815,
      "eval_samples_per_second": 79.254,
      "eval_steps_per_second": 9.951,
      "step": 2004
    },
    {
      "epoch": 13.0,
      "eval_accuracy": 0.9421052631578948,
      "eval_loss": 2.8917837142944336,
      "eval_runtime": 16.9633,
      "eval_samples_per_second": 78.405,
      "eval_steps_per_second": 9.845,
      "step": 2171
    },
    {
      "epoch": 14.0,
      "eval_accuracy": 0.943609022556391,
      "eval_loss": 2.7655045986175537,
      "eval_runtime": 17.3813,
      "eval_samples_per_second": 76.519,
      "eval_steps_per_second": 9.608,
      "step": 2338
    },
    {
      "epoch": 14.97,
      "learning_rate": 7.048192771084338e-06,
      "loss": 2.6523,
      "step": 2500
    },
    {
      "epoch": 15.0,
      "eval_accuracy": 0.9488721804511279,
      "eval_loss": 2.6490113735198975,
      "eval_runtime": 17.1632,
      "eval_samples_per_second": 77.491,
      "eval_steps_per_second": 9.73,
      "step": 2505
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.9503759398496241,
      "eval_loss": 2.5374104976654053,
      "eval_runtime": 17.1217,
      "eval_samples_per_second": 77.679,
      "eval_steps_per_second": 9.754,
      "step": 2672
    },
    {
      "epoch": 17.0,
      "eval_accuracy": 0.9533834586466166,
      "eval_loss": 2.431523084640503,
      "eval_runtime": 17.2715,
      "eval_samples_per_second": 77.005,
      "eval_steps_per_second": 9.669,
      "step": 2839
    },
    {
      "epoch": 17.96,
      "learning_rate": 6.445783132530121e-06,
      "loss": 2.2791,
      "step": 3000
    },
    {
      "epoch": 18.0,
      "eval_accuracy": 0.9533834586466166,
      "eval_loss": 2.3305835723876953,
      "eval_runtime": 17.1667,
      "eval_samples_per_second": 77.476,
      "eval_steps_per_second": 9.728,
      "step": 3006
    },
    {
      "epoch": 19.0,
      "eval_accuracy": 0.9548872180451128,
      "eval_loss": 2.232471466064453,
      "eval_runtime": 17.2703,
      "eval_samples_per_second": 77.011,
      "eval_steps_per_second": 9.67,
      "step": 3173
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.9541353383458646,
      "eval_loss": 2.1426422595977783,
      "eval_runtime": 17.61,
      "eval_samples_per_second": 75.525,
      "eval_steps_per_second": 9.483,
      "step": 3340
    },
    {
      "epoch": 20.96,
      "learning_rate": 5.843373493975905e-06,
      "loss": 1.9581,
      "step": 3500
    },
    {
      "epoch": 21.0,
      "eval_accuracy": 0.9571428571428572,
      "eval_loss": 2.056194543838501,
      "eval_runtime": 16.9422,
      "eval_samples_per_second": 78.502,
      "eval_steps_per_second": 9.857,
      "step": 3507
    },
    {
      "epoch": 22.0,
      "eval_accuracy": 0.956390977443609,
      "eval_loss": 1.9756702184677124,
      "eval_runtime": 16.9704,
      "eval_samples_per_second": 78.372,
      "eval_steps_per_second": 9.841,
      "step": 3674
    },
    {
      "epoch": 23.0,
      "eval_accuracy": 0.9571428571428572,
      "eval_loss": 1.8963149785995483,
      "eval_runtime": 16.7265,
      "eval_samples_per_second": 79.515,
      "eval_steps_per_second": 9.984,
      "step": 3841
    },
    {
      "epoch": 23.95,
      "learning_rate": 5.240963855421687e-06,
      "loss": 1.6869,
      "step": 4000
    },
    {
      "epoch": 24.0,
      "eval_accuracy": 0.9593984962406015,
      "eval_loss": 1.8221393823623657,
      "eval_runtime": 16.7647,
      "eval_samples_per_second": 79.333,
      "eval_steps_per_second": 9.961,
      "step": 4008
    },
    {
      "epoch": 25.0,
      "eval_accuracy": 0.9586466165413534,
      "eval_loss": 1.7550489902496338,
      "eval_runtime": 16.9054,
      "eval_samples_per_second": 78.673,
      "eval_steps_per_second": 9.879,
      "step": 4175
    },
    {
      "epoch": 26.0,
      "eval_accuracy": 0.9609022556390977,
      "eval_loss": 1.689659595489502,
      "eval_runtime": 16.9823,
      "eval_samples_per_second": 78.317,
      "eval_steps_per_second": 9.834,
      "step": 4342
    },
    {
      "epoch": 26.95,
      "learning_rate": 4.63855421686747e-06,
      "loss": 1.4554,
      "step": 4500
    },
    {
      "epoch": 27.0,
      "eval_accuracy": 0.9609022556390977,
      "eval_loss": 1.6294366121292114,
      "eval_runtime": 17.4283,
      "eval_samples_per_second": 76.313,
      "eval_steps_per_second": 9.582,
      "step": 4509
    },
    {
      "epoch": 28.0,
      "eval_accuracy": 0.9609022556390977,
      "eval_loss": 1.5733779668807983,
      "eval_runtime": 17.3167,
      "eval_samples_per_second": 76.805,
      "eval_steps_per_second": 9.644,
      "step": 4676
    },
    {
      "epoch": 29.0,
      "eval_accuracy": 0.9593984962406015,
      "eval_loss": 1.5163401365280151,
      "eval_runtime": 17.2582,
      "eval_samples_per_second": 77.065,
      "eval_steps_per_second": 9.677,
      "step": 4843
    },
    {
      "epoch": 29.94,
      "learning_rate": 4.036144578313254e-06,
      "loss": 1.2697,
      "step": 5000
    },
    {
      "epoch": 30.0,
      "eval_accuracy": 0.9624060150375939,
      "eval_loss": 1.4674514532089233,
      "eval_runtime": 17.3678,
      "eval_samples_per_second": 76.579,
      "eval_steps_per_second": 9.616,
      "step": 5010
    },
    {
      "epoch": 31.0,
      "eval_accuracy": 0.9624060150375939,
      "eval_loss": 1.419551968574524,
      "eval_runtime": 17.2763,
      "eval_samples_per_second": 76.984,
      "eval_steps_per_second": 9.666,
      "step": 5177
    },
    {
      "epoch": 32.0,
      "eval_accuracy": 0.9624060150375939,
      "eval_loss": 1.3748358488082886,
      "eval_runtime": 17.3245,
      "eval_samples_per_second": 76.77,
      "eval_steps_per_second": 9.64,
      "step": 5344
    },
    {
      "epoch": 32.93,
      "learning_rate": 3.4337349397590364e-06,
      "loss": 1.1151,
      "step": 5500
    },
    {
      "epoch": 33.0,
      "eval_accuracy": 0.9631578947368421,
      "eval_loss": 1.3351458311080933,
      "eval_runtime": 17.2177,
      "eval_samples_per_second": 77.246,
      "eval_steps_per_second": 9.699,
      "step": 5511
    },
    {
      "epoch": 34.0,
      "eval_accuracy": 0.9639097744360903,
      "eval_loss": 1.2986805438995361,
      "eval_runtime": 16.7721,
      "eval_samples_per_second": 79.299,
      "eval_steps_per_second": 9.957,
      "step": 5678
    },
    {
      "epoch": 35.0,
      "eval_accuracy": 0.9609022556390977,
      "eval_loss": 1.2629300355911255,
      "eval_runtime": 16.9981,
      "eval_samples_per_second": 78.244,
      "eval_steps_per_second": 9.825,
      "step": 5845
    },
    {
      "epoch": 35.93,
      "learning_rate": 2.83132530120482e-06,
      "loss": 0.9927,
      "step": 6000
    },
    {
      "epoch": 36.0,
      "eval_accuracy": 0.9631578947368421,
      "eval_loss": 1.2311846017837524,
      "eval_runtime": 17.1178,
      "eval_samples_per_second": 77.697,
      "eval_steps_per_second": 9.756,
      "step": 6012
    },
    {
      "epoch": 37.0,
      "eval_accuracy": 0.9631578947368421,
      "eval_loss": 1.2015360593795776,
      "eval_runtime": 16.8722,
      "eval_samples_per_second": 78.828,
      "eval_steps_per_second": 9.898,
      "step": 6179
    },
    {
      "epoch": 38.0,
      "eval_accuracy": 0.9631578947368421,
      "eval_loss": 1.174642562866211,
      "eval_runtime": 16.9885,
      "eval_samples_per_second": 78.288,
      "eval_steps_per_second": 9.83,
      "step": 6346
    },
    {
      "epoch": 38.92,
      "learning_rate": 2.2289156626506024e-06,
      "loss": 0.8987,
      "step": 6500
    },
    {
      "epoch": 39.0,
      "eval_accuracy": 0.9624060150375939,
      "eval_loss": 1.1511969566345215,
      "eval_runtime": 16.9611,
      "eval_samples_per_second": 78.415,
      "eval_steps_per_second": 9.846,
      "step": 6513
    },
    {
      "epoch": 40.0,
      "eval_accuracy": 0.9631578947368421,
      "eval_loss": 1.1287494897842407,
      "eval_runtime": 17.1976,
      "eval_samples_per_second": 77.336,
      "eval_steps_per_second": 9.711,
      "step": 6680
    },
    {
      "epoch": 41.0,
      "eval_accuracy": 0.9631578947368421,
      "eval_loss": 1.1099363565444946,
      "eval_runtime": 17.1353,
      "eval_samples_per_second": 77.618,
      "eval_steps_per_second": 9.746,
      "step": 6847
    },
    {
      "epoch": 41.92,
      "learning_rate": 1.6265060240963856e-06,
      "loss": 0.8287,
      "step": 7000
    },
    {
      "epoch": 42.0,
      "eval_accuracy": 0.9631578947368421,
      "eval_loss": 1.0933747291564941,
      "eval_runtime": 17.1628,
      "eval_samples_per_second": 77.493,
      "eval_steps_per_second": 9.73,
      "step": 7014
    },
    {
      "epoch": 43.0,
      "eval_accuracy": 0.9639097744360903,
      "eval_loss": 1.078044056892395,
      "eval_runtime": 17.3482,
      "eval_samples_per_second": 76.665,
      "eval_steps_per_second": 9.626,
      "step": 7181
    },
    {
      "epoch": 44.0,
      "eval_accuracy": 0.9646616541353383,
      "eval_loss": 1.06500244140625,
      "eval_runtime": 17.2738,
      "eval_samples_per_second": 76.995,
      "eval_steps_per_second": 9.668,
      "step": 7348
    },
    {
      "epoch": 44.91,
      "learning_rate": 1.0240963855421688e-06,
      "loss": 0.7796,
      "step": 7500
    },
    {
      "epoch": 45.0,
      "eval_accuracy": 0.9646616541353383,
      "eval_loss": 1.0539588928222656,
      "eval_runtime": 17.5661,
      "eval_samples_per_second": 75.714,
      "eval_steps_per_second": 9.507,
      "step": 7515
    },
    {
      "epoch": 46.0,
      "eval_accuracy": 0.9639097744360903,
      "eval_loss": 1.045282244682312,
      "eval_runtime": 17.4501,
      "eval_samples_per_second": 76.217,
      "eval_steps_per_second": 9.57,
      "step": 7682
    },
    {
      "epoch": 47.0,
      "eval_accuracy": 0.9646616541353383,
      "eval_loss": 1.0384619235992432,
      "eval_runtime": 17.119,
      "eval_samples_per_second": 77.691,
      "eval_steps_per_second": 9.755,
      "step": 7849
    },
    {
      "epoch": 47.9,
      "learning_rate": 4.216867469879518e-07,
      "loss": 0.7499,
      "step": 8000
    },
    {
      "epoch": 48.0,
      "eval_accuracy": 0.9639097744360903,
      "eval_loss": 1.0335502624511719,
      "eval_runtime": 16.7365,
      "eval_samples_per_second": 79.467,
      "eval_steps_per_second": 9.978,
      "step": 8016
    },
    {
      "epoch": 49.0,
      "eval_accuracy": 0.9646616541353383,
      "eval_loss": 1.0308656692504883,
      "eval_runtime": 16.8233,
      "eval_samples_per_second": 79.057,
      "eval_steps_per_second": 9.927,
      "step": 8183
    },
    {
      "epoch": 50.0,
      "eval_accuracy": 0.9646616541353383,
      "eval_loss": 1.0297563076019287,
      "eval_runtime": 17.0044,
      "eval_samples_per_second": 78.215,
      "eval_steps_per_second": 9.821,
      "step": 8350
    }
  ],
  "logging_steps": 500,
  "max_steps": 8350,
  "num_train_epochs": 50,
  "save_steps": 500,
  "total_flos": 2.0632452260191027e+19,
  "trial_name": null,
  "trial_params": null
}