{
  "best_metric": 0.9564084340203743,
  "best_model_checkpoint": "/scratch/mrahma45/pixel/finetuned_models/mbert/mbert-base-finetuned-pos-ud-Chinese-GSD/checkpoint-1500",
  "epoch": 32.0,
  "global_step": 4000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.8,
      "learning_rate": 7.92e-05,
      "loss": 1.0164,
      "step": 100
    },
    {
      "epoch": 1.6,
      "learning_rate": 7.946845637583894e-05,
      "loss": 0.1696,
      "step": 200
    },
    {
      "epoch": 2.4,
      "learning_rate": 7.893154362416109e-05,
      "loss": 0.1167,
      "step": 300
    },
    {
      "epoch": 3.2,
      "learning_rate": 7.839463087248322e-05,
      "loss": 0.0845,
      "step": 400
    },
    {
      "epoch": 4.0,
      "learning_rate": 7.785771812080537e-05,
      "loss": 0.0603,
      "step": 500
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.9558556424228066,
      "eval_loss": 0.1726187914609909,
      "eval_runtime": 2.583,
      "eval_samples_per_second": 193.571,
      "eval_steps_per_second": 24.39,
      "step": 500
    },
    {
      "epoch": 4.8,
      "learning_rate": 7.732080536912752e-05,
      "loss": 0.0413,
      "step": 600
    },
    {
      "epoch": 5.6,
      "learning_rate": 7.678389261744967e-05,
      "loss": 0.0335,
      "step": 700
    },
    {
      "epoch": 6.4,
      "learning_rate": 7.624697986577182e-05,
      "loss": 0.0287,
      "step": 800
    },
    {
      "epoch": 7.2,
      "learning_rate": 7.571006711409396e-05,
      "loss": 0.0259,
      "step": 900
    },
    {
      "epoch": 8.0,
      "learning_rate": 7.517315436241611e-05,
      "loss": 0.0214,
      "step": 1000
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.9563294637921503,
      "eval_loss": 0.21138951182365417,
      "eval_runtime": 2.5897,
      "eval_samples_per_second": 193.075,
      "eval_steps_per_second": 24.327,
      "step": 1000
    },
    {
      "epoch": 8.8,
      "learning_rate": 7.463624161073826e-05,
      "loss": 0.0184,
      "step": 1100
    },
    {
      "epoch": 9.6,
      "learning_rate": 7.409932885906041e-05,
      "loss": 0.0176,
      "step": 1200
    },
    {
      "epoch": 10.4,
      "learning_rate": 7.356241610738256e-05,
      "loss": 0.017,
      "step": 1300
    },
    {
      "epoch": 11.2,
      "learning_rate": 7.30255033557047e-05,
      "loss": 0.0139,
      "step": 1400
    },
    {
      "epoch": 12.0,
      "learning_rate": 7.248859060402685e-05,
      "loss": 0.0143,
      "step": 1500
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.9564084340203743,
      "eval_loss": 0.22227974236011505,
      "eval_runtime": 2.5849,
      "eval_samples_per_second": 193.435,
      "eval_steps_per_second": 24.373,
      "step": 1500
    },
    {
      "epoch": 12.8,
      "learning_rate": 7.1951677852349e-05,
      "loss": 0.0128,
      "step": 1600
    },
    {
      "epoch": 13.6,
      "learning_rate": 7.141476510067115e-05,
      "loss": 0.0112,
      "step": 1700
    },
    {
      "epoch": 14.4,
      "learning_rate": 7.08778523489933e-05,
      "loss": 0.0118,
      "step": 1800
    },
    {
      "epoch": 15.2,
      "learning_rate": 7.034093959731545e-05,
      "loss": 0.0118,
      "step": 1900
    },
    {
      "epoch": 16.0,
      "learning_rate": 6.98040268456376e-05,
      "loss": 0.0099,
      "step": 2000
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.9560135828792545,
      "eval_loss": 0.24710656702518463,
      "eval_runtime": 2.5783,
      "eval_samples_per_second": 193.93,
      "eval_steps_per_second": 24.435,
      "step": 2000
    },
    {
      "epoch": 16.8,
      "learning_rate": 6.926711409395974e-05,
      "loss": 0.0087,
      "step": 2100
    },
    {
      "epoch": 17.6,
      "learning_rate": 6.873020134228189e-05,
      "loss": 0.0101,
      "step": 2200
    },
    {
      "epoch": 18.4,
      "learning_rate": 6.819328859060404e-05,
      "loss": 0.0094,
      "step": 2300
    },
    {
      "epoch": 19.2,
      "learning_rate": 6.765637583892619e-05,
      "loss": 0.0095,
      "step": 2400
    },
    {
      "epoch": 20.0,
      "learning_rate": 6.711946308724834e-05,
      "loss": 0.0086,
      "step": 2500
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.9556977019663587,
      "eval_loss": 0.248493954539299,
      "eval_runtime": 2.5942,
      "eval_samples_per_second": 192.736,
      "eval_steps_per_second": 24.285,
      "step": 2500
    },
    {
      "epoch": 20.8,
      "learning_rate": 6.658255033557048e-05,
      "loss": 0.009,
      "step": 2600
    },
    {
      "epoch": 21.6,
      "learning_rate": 6.604563758389262e-05,
      "loss": 0.0088,
      "step": 2700
    },
    {
      "epoch": 22.4,
      "learning_rate": 6.550872483221477e-05,
      "loss": 0.0087,
      "step": 2800
    },
    {
      "epoch": 23.2,
      "learning_rate": 6.497181208053692e-05,
      "loss": 0.0073,
      "step": 2900
    },
    {
      "epoch": 24.0,
      "learning_rate": 6.443489932885906e-05,
      "loss": 0.0083,
      "step": 3000
    },
    {
      "epoch": 24.0,
      "eval_accuracy": 0.9560925531074784,
      "eval_loss": 0.2502564489841461,
      "eval_runtime": 2.5819,
      "eval_samples_per_second": 193.657,
      "eval_steps_per_second": 24.401,
      "step": 3000
    },
    {
      "epoch": 24.8,
      "learning_rate": 6.389798657718121e-05,
      "loss": 0.0063,
      "step": 3100
    },
    {
      "epoch": 25.6,
      "learning_rate": 6.336107382550336e-05,
      "loss": 0.0064,
      "step": 3200
    },
    {
      "epoch": 26.4,
      "learning_rate": 6.282416107382551e-05,
      "loss": 0.0069,
      "step": 3300
    },
    {
      "epoch": 27.2,
      "learning_rate": 6.228724832214766e-05,
      "loss": 0.0071,
      "step": 3400
    },
    {
      "epoch": 28.0,
      "learning_rate": 6.17503355704698e-05,
      "loss": 0.0052,
      "step": 3500
    },
    {
      "epoch": 28.0,
      "eval_accuracy": 0.9552238805970149,
      "eval_loss": 0.26657333970069885,
      "eval_runtime": 2.6131,
      "eval_samples_per_second": 191.344,
      "eval_steps_per_second": 24.109,
      "step": 3500
    },
    {
      "epoch": 28.8,
      "learning_rate": 6.121342281879195e-05,
      "loss": 0.0053,
      "step": 3600
    },
    {
      "epoch": 29.6,
      "learning_rate": 6.06765100671141e-05,
      "loss": 0.006,
      "step": 3700
    },
    {
      "epoch": 30.4,
      "learning_rate": 6.013959731543624e-05,
      "loss": 0.0058,
      "step": 3800
    },
    {
      "epoch": 31.2,
      "learning_rate": 5.960268456375839e-05,
      "loss": 0.006,
      "step": 3900
    },
    {
      "epoch": 32.0,
      "learning_rate": 5.906577181208054e-05,
      "loss": 0.0065,
      "step": 4000
    },
    {
      "epoch": 32.0,
      "eval_accuracy": 0.9546710889994472,
      "eval_loss": 0.27891796827316284,
      "eval_runtime": 2.5975,
      "eval_samples_per_second": 192.494,
      "eval_steps_per_second": 24.254,
      "step": 4000
    },
    {
      "epoch": 32.0,
      "step": 4000,
      "total_flos": 1.6712716359155712e+16,
      "train_loss": 0.04717882344126701,
      "train_runtime": 1107.6488,
      "train_samples_per_second": 433.35,
      "train_steps_per_second": 13.542
    }
  ],
  "max_steps": 15000,
  "num_train_epochs": 120,
  "total_flos": 1.6712716359155712e+16,
  "trial_name": null,
  "trial_params": null
}