{
  "best_global_step": 2142,
  "best_metric": 0.8572526627868673,
  "best_model_checkpoint": "./data/ynat-model/checkpoint-2142",
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 2142,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0700280112044818,
      "grad_norm": 4.37371301651001,
      "learning_rate": 4.88562091503268e-05,
      "loss": 0.368,
      "step": 50
    },
    {
      "epoch": 0.1400560224089636,
      "grad_norm": 2.0869579315185547,
      "learning_rate": 4.7689075630252105e-05,
      "loss": 0.3318,
      "step": 100
    },
    {
      "epoch": 0.21008403361344538,
      "grad_norm": 5.720030784606934,
      "learning_rate": 4.6521942110177405e-05,
      "loss": 0.3172,
      "step": 150
    },
    {
      "epoch": 0.2801120448179272,
      "grad_norm": 11.471839904785156,
      "learning_rate": 4.535480859010271e-05,
      "loss": 0.2934,
      "step": 200
    },
    {
      "epoch": 0.35014005602240894,
      "grad_norm": 3.376629114151001,
      "learning_rate": 4.418767507002801e-05,
      "loss": 0.3005,
      "step": 250
    },
    {
      "epoch": 0.42016806722689076,
      "grad_norm": 4.57407808303833,
      "learning_rate": 4.302054154995332e-05,
      "loss": 0.2652,
      "step": 300
    },
    {
      "epoch": 0.49019607843137253,
      "grad_norm": 4.265451431274414,
      "learning_rate": 4.185340802987862e-05,
      "loss": 0.2649,
      "step": 350
    },
    {
      "epoch": 0.5602240896358543,
      "grad_norm": 4.1893086433410645,
      "learning_rate": 4.068627450980392e-05,
      "loss": 0.2446,
      "step": 400
    },
    {
      "epoch": 0.6302521008403361,
      "grad_norm": 2.6485085487365723,
      "learning_rate": 3.951914098972922e-05,
      "loss": 0.2269,
      "step": 450
    },
    {
      "epoch": 0.7002801120448179,
      "grad_norm": 3.4019579887390137,
      "learning_rate": 3.8352007469654535e-05,
      "loss": 0.2324,
      "step": 500
    },
    {
      "epoch": 0.7703081232492998,
      "grad_norm": 1.2434275150299072,
      "learning_rate": 3.7184873949579834e-05,
      "loss": 0.2329,
      "step": 550
    },
    {
      "epoch": 0.8403361344537815,
      "grad_norm": 5.956329345703125,
      "learning_rate": 3.6017740429505134e-05,
      "loss": 0.207,
      "step": 600
    },
    {
      "epoch": 0.9103641456582633,
      "grad_norm": 2.210266351699829,
      "learning_rate": 3.485060690943044e-05,
      "loss": 0.2278,
      "step": 650
    },
    {
      "epoch": 0.9803921568627451,
      "grad_norm": 3.8561065196990967,
      "learning_rate": 3.3683473389355746e-05,
      "loss": 0.2255,
      "step": 700
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8529702426704733,
      "eval_f1": 0.8516564005044691,
      "eval_loss": 0.4749363362789154,
      "eval_precision": 0.8525157577020369,
      "eval_recall": 0.8535400561743023,
      "eval_runtime": 12.9676,
      "eval_samples_per_second": 702.29,
      "eval_steps_per_second": 11.028,
      "step": 714
    },
    {
      "epoch": 1.050420168067227,
      "grad_norm": 2.6144492626190186,
      "learning_rate": 3.2516339869281046e-05,
      "loss": 0.2636,
      "step": 750
    },
    {
      "epoch": 1.1204481792717087,
      "grad_norm": 4.134031772613525,
      "learning_rate": 3.134920634920635e-05,
      "loss": 0.2475,
      "step": 800
    },
    {
      "epoch": 1.1904761904761905,
      "grad_norm": 4.822914123535156,
      "learning_rate": 3.018207282913165e-05,
      "loss": 0.2398,
      "step": 850
    },
    {
      "epoch": 1.2605042016806722,
      "grad_norm": 4.078502655029297,
      "learning_rate": 2.901493930905696e-05,
      "loss": 0.2369,
      "step": 900
    },
    {
      "epoch": 1.330532212885154,
      "grad_norm": 5.511151313781738,
      "learning_rate": 2.784780578898226e-05,
      "loss": 0.2519,
      "step": 950
    },
    {
      "epoch": 1.4005602240896358,
      "grad_norm": 14.258254051208496,
      "learning_rate": 2.6680672268907564e-05,
      "loss": 0.2603,
      "step": 1000
    },
    {
      "epoch": 1.4705882352941178,
      "grad_norm": 37.48504638671875,
      "learning_rate": 2.5513538748832866e-05,
      "loss": 0.2329,
      "step": 1050
    },
    {
      "epoch": 1.5406162464985993,
      "grad_norm": 9.023612976074219,
      "learning_rate": 2.434640522875817e-05,
      "loss": 0.2524,
      "step": 1100
    },
    {
      "epoch": 1.6106442577030813,
      "grad_norm": 4.550325870513916,
      "learning_rate": 2.3179271708683476e-05,
      "loss": 0.2249,
      "step": 1150
    },
    {
      "epoch": 1.680672268907563,
      "grad_norm": 5.294163227081299,
      "learning_rate": 2.201213818860878e-05,
      "loss": 0.229,
      "step": 1200
    },
    {
      "epoch": 1.7507002801120448,
      "grad_norm": 5.975066661834717,
      "learning_rate": 2.084500466853408e-05,
      "loss": 0.2346,
      "step": 1250
    },
    {
      "epoch": 1.8207282913165266,
      "grad_norm": 9.633162498474121,
      "learning_rate": 1.9677871148459384e-05,
      "loss": 0.2388,
      "step": 1300
    },
    {
      "epoch": 1.8907563025210083,
      "grad_norm": 2.6880970001220703,
      "learning_rate": 1.851073762838469e-05,
      "loss": 0.2372,
      "step": 1350
    },
    {
      "epoch": 1.9607843137254903,
      "grad_norm": 5.544015884399414,
      "learning_rate": 1.734360410830999e-05,
      "loss": 0.2296,
      "step": 1400
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.855825189414736,
      "eval_f1": 0.8561722862205893,
      "eval_loss": 0.44257503747940063,
      "eval_precision": 0.8489764689633209,
      "eval_recall": 0.865244799193242,
      "eval_runtime": 13.04,
      "eval_samples_per_second": 698.387,
      "eval_steps_per_second": 10.966,
      "step": 1428
    },
    {
      "epoch": 2.030812324929972,
      "grad_norm": 3.5343873500823975,
      "learning_rate": 1.6176470588235296e-05,
      "loss": 0.204,
      "step": 1450
    },
    {
      "epoch": 2.100840336134454,
      "grad_norm": 2.5762808322906494,
      "learning_rate": 1.5009337068160597e-05,
      "loss": 0.182,
      "step": 1500
    },
    {
      "epoch": 2.1708683473389354,
      "grad_norm": 2.4014556407928467,
      "learning_rate": 1.3842203548085902e-05,
      "loss": 0.1659,
      "step": 1550
    },
    {
      "epoch": 2.2408963585434174,
      "grad_norm": 5.196502685546875,
      "learning_rate": 1.2675070028011205e-05,
      "loss": 0.1721,
      "step": 1600
    },
    {
      "epoch": 2.310924369747899,
      "grad_norm": 2.636826992034912,
      "learning_rate": 1.1507936507936508e-05,
      "loss": 0.1781,
      "step": 1650
    },
    {
      "epoch": 2.380952380952381,
      "grad_norm": 3.6006743907928467,
      "learning_rate": 1.0340802987861812e-05,
      "loss": 0.1788,
      "step": 1700
    },
    {
      "epoch": 2.450980392156863,
      "grad_norm": 3.009672164916992,
      "learning_rate": 9.173669467787115e-06,
      "loss": 0.1574,
      "step": 1750
    },
    {
      "epoch": 2.5210084033613445,
      "grad_norm": 2.854926586151123,
      "learning_rate": 8.006535947712418e-06,
      "loss": 0.1511,
      "step": 1800
    },
    {
      "epoch": 2.5910364145658265,
      "grad_norm": 5.09183406829834,
      "learning_rate": 6.839402427637723e-06,
      "loss": 0.1993,
      "step": 1850
    },
    {
      "epoch": 2.661064425770308,
      "grad_norm": 1.9752312898635864,
      "learning_rate": 5.672268907563025e-06,
      "loss": 0.1701,
      "step": 1900
    },
    {
      "epoch": 2.73109243697479,
      "grad_norm": 3.669840097427368,
      "learning_rate": 4.505135387488329e-06,
      "loss": 0.1698,
      "step": 1950
    },
    {
      "epoch": 2.8011204481792715,
      "grad_norm": 2.8419978618621826,
      "learning_rate": 3.338001867413632e-06,
      "loss": 0.1637,
      "step": 2000
    },
    {
      "epoch": 2.8711484593837535,
      "grad_norm": 3.0742175579071045,
      "learning_rate": 2.1708683473389355e-06,
      "loss": 0.1591,
      "step": 2050
    },
    {
      "epoch": 2.9411764705882355,
      "grad_norm": 3.2953155040740967,
      "learning_rate": 1.0037348272642392e-06,
      "loss": 0.1686,
      "step": 2100
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8572526627868673,
      "eval_f1": 0.8569137587108026,
      "eval_loss": 0.4840508997440338,
      "eval_precision": 0.848038822955364,
      "eval_recall": 0.8670435119879699,
      "eval_runtime": 12.9882,
      "eval_samples_per_second": 701.177,
      "eval_steps_per_second": 11.01,
      "step": 2142
    }
  ],
  "logging_steps": 50,
  "max_steps": 2142,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1616745095802300.0,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": null
}