{
  "best_metric": 0.8855701147554326,
  "best_model_checkpoint": "/scratch/mrahma45/pixel/finetuned_models/bert/bert-base-finetuned-pos-ud-Japanese-GSD/checkpoint-2000",
  "epoch": 20.361990950226243,
  "global_step": 4500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.45,
      "learning_rate": 4.9500000000000004e-05,
      "loss": 1.3362,
      "step": 100
    },
    {
      "epoch": 0.9,
      "learning_rate": 4.966778523489933e-05,
      "loss": 0.5269,
      "step": 200
    },
    {
      "epoch": 1.36,
      "learning_rate": 4.933221476510068e-05,
      "loss": 0.432,
      "step": 300
    },
    {
      "epoch": 1.81,
      "learning_rate": 4.8996644295302016e-05,
      "loss": 0.4029,
      "step": 400
    },
    {
      "epoch": 2.26,
      "learning_rate": 4.8661073825503355e-05,
      "loss": 0.3774,
      "step": 500
    },
    {
      "epoch": 2.26,
      "eval_accuracy": 0.8758850817937658,
      "eval_loss": 0.37171030044555664,
      "eval_runtime": 2.6026,
      "eval_samples_per_second": 194.803,
      "eval_steps_per_second": 24.591,
      "step": 500
    },
    {
      "epoch": 2.71,
      "learning_rate": 4.83255033557047e-05,
      "loss": 0.3581,
      "step": 600
    },
    {
      "epoch": 3.17,
      "learning_rate": 4.798993288590604e-05,
      "loss": 0.3378,
      "step": 700
    },
    {
      "epoch": 3.62,
      "learning_rate": 4.765436241610739e-05,
      "loss": 0.3239,
      "step": 800
    },
    {
      "epoch": 4.07,
      "learning_rate": 4.7318791946308726e-05,
      "loss": 0.3132,
      "step": 900
    },
    {
      "epoch": 4.52,
      "learning_rate": 4.698322147651007e-05,
      "loss": 0.2882,
      "step": 1000
    },
    {
      "epoch": 4.52,
      "eval_accuracy": 0.8841865386180516,
      "eval_loss": 0.3594276010990143,
      "eval_runtime": 2.6266,
      "eval_samples_per_second": 193.024,
      "eval_steps_per_second": 24.366,
      "step": 1000
    },
    {
      "epoch": 4.98,
      "learning_rate": 4.664765100671141e-05,
      "loss": 0.2882,
      "step": 1100
    },
    {
      "epoch": 5.43,
      "learning_rate": 4.631208053691276e-05,
      "loss": 0.2499,
      "step": 1200
    },
    {
      "epoch": 5.88,
      "learning_rate": 4.5976510067114097e-05,
      "loss": 0.2633,
      "step": 1300
    },
    {
      "epoch": 6.33,
      "learning_rate": 4.564093959731544e-05,
      "loss": 0.2364,
      "step": 1400
    },
    {
      "epoch": 6.79,
      "learning_rate": 4.5305369127516775e-05,
      "loss": 0.2304,
      "step": 1500
    },
    {
      "epoch": 6.79,
      "eval_accuracy": 0.8854887279238219,
      "eval_loss": 0.3736213743686676,
      "eval_runtime": 2.6247,
      "eval_samples_per_second": 193.165,
      "eval_steps_per_second": 24.384,
      "step": 1500
    },
    {
      "epoch": 7.24,
      "learning_rate": 4.496979865771812e-05,
      "loss": 0.2144,
      "step": 1600
    },
    {
      "epoch": 7.69,
      "learning_rate": 4.463422818791946e-05,
      "loss": 0.2067,
      "step": 1700
    },
    {
      "epoch": 8.14,
      "learning_rate": 4.4298657718120806e-05,
      "loss": 0.1925,
      "step": 1800
    },
    {
      "epoch": 8.6,
      "learning_rate": 4.3963087248322146e-05,
      "loss": 0.1764,
      "step": 1900
    },
    {
      "epoch": 9.05,
      "learning_rate": 4.362751677852349e-05,
      "loss": 0.1778,
      "step": 2000
    },
    {
      "epoch": 9.05,
      "eval_accuracy": 0.8855701147554326,
      "eval_loss": 0.43164435029029846,
      "eval_runtime": 2.6105,
      "eval_samples_per_second": 194.219,
      "eval_steps_per_second": 24.517,
      "step": 2000
    },
    {
      "epoch": 9.5,
      "learning_rate": 4.329194630872484e-05,
      "loss": 0.1529,
      "step": 2100
    },
    {
      "epoch": 9.95,
      "learning_rate": 4.295637583892618e-05,
      "loss": 0.1583,
      "step": 2200
    },
    {
      "epoch": 10.41,
      "learning_rate": 4.262080536912752e-05,
      "loss": 0.1356,
      "step": 2300
    },
    {
      "epoch": 10.86,
      "learning_rate": 4.228523489932886e-05,
      "loss": 0.1374,
      "step": 2400
    },
    {
      "epoch": 11.31,
      "learning_rate": 4.194966442953021e-05,
      "loss": 0.1217,
      "step": 2500
    },
    {
      "epoch": 11.31,
      "eval_accuracy": 0.8805241311955726,
      "eval_loss": 0.493513286113739,
      "eval_runtime": 2.6161,
      "eval_samples_per_second": 193.798,
      "eval_steps_per_second": 24.464,
      "step": 2500
    },
    {
      "epoch": 11.76,
      "learning_rate": 4.161409395973155e-05,
      "loss": 0.1193,
      "step": 2600
    },
    {
      "epoch": 12.22,
      "learning_rate": 4.127852348993289e-05,
      "loss": 0.1098,
      "step": 2700
    },
    {
      "epoch": 12.67,
      "learning_rate": 4.0942953020134226e-05,
      "loss": 0.103,
      "step": 2800
    },
    {
      "epoch": 13.12,
      "learning_rate": 4.060738255033557e-05,
      "loss": 0.1012,
      "step": 2900
    },
    {
      "epoch": 13.57,
      "learning_rate": 4.027181208053691e-05,
      "loss": 0.0896,
      "step": 3000
    },
    {
      "epoch": 13.57,
      "eval_accuracy": 0.8823146414910068,
      "eval_loss": 0.5315696001052856,
      "eval_runtime": 2.6096,
      "eval_samples_per_second": 194.28,
      "eval_steps_per_second": 24.525,
      "step": 3000
    },
    {
      "epoch": 14.03,
      "learning_rate": 3.993624161073826e-05,
      "loss": 0.0869,
      "step": 3100
    },
    {
      "epoch": 14.48,
      "learning_rate": 3.96006711409396e-05,
      "loss": 0.0784,
      "step": 3200
    },
    {
      "epoch": 14.93,
      "learning_rate": 3.926510067114094e-05,
      "loss": 0.0783,
      "step": 3300
    },
    {
      "epoch": 15.38,
      "learning_rate": 3.892953020134228e-05,
      "loss": 0.0691,
      "step": 3400
    },
    {
      "epoch": 15.84,
      "learning_rate": 3.859395973154363e-05,
      "loss": 0.0685,
      "step": 3500
    },
    {
      "epoch": 15.84,
      "eval_accuracy": 0.8823960283226174,
      "eval_loss": 0.5921140313148499,
      "eval_runtime": 2.6221,
      "eval_samples_per_second": 193.357,
      "eval_steps_per_second": 24.408,
      "step": 3500
    },
    {
      "epoch": 16.29,
      "learning_rate": 3.825838926174497e-05,
      "loss": 0.0646,
      "step": 3600
    },
    {
      "epoch": 16.74,
      "learning_rate": 3.7922818791946313e-05,
      "loss": 0.0592,
      "step": 3700
    },
    {
      "epoch": 17.19,
      "learning_rate": 3.758724832214765e-05,
      "loss": 0.0593,
      "step": 3800
    },
    {
      "epoch": 17.65,
      "learning_rate": 3.725167785234899e-05,
      "loss": 0.0546,
      "step": 3900
    },
    {
      "epoch": 18.1,
      "learning_rate": 3.691610738255034e-05,
      "loss": 0.0546,
      "step": 4000
    },
    {
      "epoch": 18.1,
      "eval_accuracy": 0.88272157564906,
      "eval_loss": 0.6464724540710449,
      "eval_runtime": 2.62,
      "eval_samples_per_second": 193.511,
      "eval_steps_per_second": 24.427,
      "step": 4000
    },
    {
      "epoch": 18.55,
      "learning_rate": 3.658053691275168e-05,
      "loss": 0.0494,
      "step": 4100
    },
    {
      "epoch": 19.0,
      "learning_rate": 3.624496644295302e-05,
      "loss": 0.0519,
      "step": 4200
    },
    {
      "epoch": 19.46,
      "learning_rate": 3.590939597315436e-05,
      "loss": 0.0422,
      "step": 4300
    },
    {
      "epoch": 19.91,
      "learning_rate": 3.557382550335571e-05,
      "loss": 0.0445,
      "step": 4400
    },
    {
      "epoch": 20.36,
      "learning_rate": 3.523825503355705e-05,
      "loss": 0.0395,
      "step": 4500
    },
    {
      "epoch": 20.36,
      "eval_accuracy": 0.8812566126800684,
      "eval_loss": 0.6766866445541382,
      "eval_runtime": 2.6147,
      "eval_samples_per_second": 193.903,
      "eval_steps_per_second": 24.477,
      "step": 4500
    },
    {
      "epoch": 20.36,
      "step": 4500,
      "total_flos": 1.875842476013568e+16,
      "train_loss": 0.20138724523120458,
      "train_runtime": 1165.6408,
      "train_samples_per_second": 411.791,
      "train_steps_per_second": 12.868
    }
  ],
  "max_steps": 15000,
  "num_train_epochs": 68,
  "total_flos": 1.875842476013568e+16,
  "trial_name": null,
  "trial_params": null
}