{
  "best_metric": 0.8677751385589866,
  "best_model_checkpoint": "/scratch/mrahma45/pixel/finetuned_models/mbert/mbert-base-finetuned-pos-ud-Tamil-TTB/checkpoint-2000",
  "epoch": 346.15384615384613,
  "global_step": 4500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 7.69,
      "learning_rate": 7.92e-05,
      "loss": 0.9585,
      "step": 100
    },
    {
      "epoch": 15.38,
      "learning_rate": 7.946845637583894e-05,
      "loss": 0.0575,
      "step": 200
    },
    {
      "epoch": 23.08,
      "learning_rate": 7.893691275167786e-05,
      "loss": 0.0242,
      "step": 300
    },
    {
      "epoch": 30.77,
      "learning_rate": 7.840000000000001e-05,
      "loss": 0.0156,
      "step": 400
    },
    {
      "epoch": 38.46,
      "learning_rate": 7.786308724832216e-05,
      "loss": 0.0149,
      "step": 500
    },
    {
      "epoch": 38.46,
      "eval_accuracy": 0.8661916072842438,
      "eval_loss": 0.9775320887565613,
      "eval_runtime": 0.4044,
      "eval_samples_per_second": 197.82,
      "eval_steps_per_second": 24.727,
      "step": 500
    },
    {
      "epoch": 46.15,
      "learning_rate": 7.73261744966443e-05,
      "loss": 0.0111,
      "step": 600
    },
    {
      "epoch": 53.85,
      "learning_rate": 7.678926174496645e-05,
      "loss": 0.0117,
      "step": 700
    },
    {
      "epoch": 61.54,
      "learning_rate": 7.62523489932886e-05,
      "loss": 0.0094,
      "step": 800
    },
    {
      "epoch": 69.23,
      "learning_rate": 7.571543624161075e-05,
      "loss": 0.0111,
      "step": 900
    },
    {
      "epoch": 76.92,
      "learning_rate": 7.51785234899329e-05,
      "loss": 0.0091,
      "step": 1000
    },
    {
      "epoch": 76.92,
      "eval_accuracy": 0.8638163103721298,
      "eval_loss": 0.9691746830940247,
      "eval_runtime": 0.4022,
      "eval_samples_per_second": 198.893,
      "eval_steps_per_second": 24.862,
      "step": 1000
    },
    {
      "epoch": 84.62,
      "learning_rate": 7.464161073825505e-05,
      "loss": 0.0079,
      "step": 1100
    },
    {
      "epoch": 92.31,
      "learning_rate": 7.410469798657718e-05,
      "loss": 0.0065,
      "step": 1200
    },
    {
      "epoch": 100.0,
      "learning_rate": 7.356778523489933e-05,
      "loss": 0.0043,
      "step": 1300
    },
    {
      "epoch": 107.69,
      "learning_rate": 7.303087248322148e-05,
      "loss": 0.0047,
      "step": 1400
    },
    {
      "epoch": 115.38,
      "learning_rate": 7.249395973154363e-05,
      "loss": 0.0055,
      "step": 1500
    },
    {
      "epoch": 115.38,
      "eval_accuracy": 0.8582739509105305,
      "eval_loss": 1.1557462215423584,
      "eval_runtime": 0.4044,
      "eval_samples_per_second": 197.81,
      "eval_steps_per_second": 24.726,
      "step": 1500
    },
    {
      "epoch": 123.08,
      "learning_rate": 7.195704697986577e-05,
      "loss": 0.0051,
      "step": 1600
    },
    {
      "epoch": 130.77,
      "learning_rate": 7.142550335570471e-05,
      "loss": 0.0055,
      "step": 1700
    },
    {
      "epoch": 138.46,
      "learning_rate": 7.088859060402686e-05,
      "loss": 0.0056,
      "step": 1800
    },
    {
      "epoch": 146.15,
      "learning_rate": 7.0351677852349e-05,
      "loss": 0.0055,
      "step": 1900
    },
    {
      "epoch": 153.85,
      "learning_rate": 6.981476510067114e-05,
      "loss": 0.0061,
      "step": 2000
    },
    {
      "epoch": 153.85,
      "eval_accuracy": 0.8677751385589866,
      "eval_loss": 1.1852538585662842,
      "eval_runtime": 0.4041,
      "eval_samples_per_second": 197.965,
      "eval_steps_per_second": 24.746,
      "step": 2000
    },
    {
      "epoch": 161.54,
      "learning_rate": 6.927785234899329e-05,
      "loss": 0.0069,
      "step": 2100
    },
    {
      "epoch": 169.23,
      "learning_rate": 6.874093959731543e-05,
      "loss": 0.0055,
      "step": 2200
    },
    {
      "epoch": 176.92,
      "learning_rate": 6.820402684563758e-05,
      "loss": 0.0043,
      "step": 2300
    },
    {
      "epoch": 184.62,
      "learning_rate": 6.766711409395973e-05,
      "loss": 0.0043,
      "step": 2400
    },
    {
      "epoch": 192.31,
      "learning_rate": 6.713020134228188e-05,
      "loss": 0.0041,
      "step": 2500
    },
    {
      "epoch": 192.31,
      "eval_accuracy": 0.8622327790973872,
      "eval_loss": 1.1900646686553955,
      "eval_runtime": 0.4087,
      "eval_samples_per_second": 195.729,
      "eval_steps_per_second": 24.466,
      "step": 2500
    },
    {
      "epoch": 200.0,
      "learning_rate": 6.659328859060403e-05,
      "loss": 0.0029,
      "step": 2600
    },
    {
      "epoch": 207.69,
      "learning_rate": 6.605637583892618e-05,
      "loss": 0.0021,
      "step": 2700
    },
    {
      "epoch": 215.38,
      "learning_rate": 6.551946308724832e-05,
      "loss": 0.0047,
      "step": 2800
    },
    {
      "epoch": 223.08,
      "learning_rate": 6.498255033557047e-05,
      "loss": 0.0058,
      "step": 2900
    },
    {
      "epoch": 230.77,
      "learning_rate": 6.444563758389262e-05,
      "loss": 0.0045,
      "step": 3000
    },
    {
      "epoch": 230.77,
      "eval_accuracy": 0.8566904196357878,
      "eval_loss": 1.263396978378296,
      "eval_runtime": 0.4055,
      "eval_samples_per_second": 197.27,
      "eval_steps_per_second": 24.659,
      "step": 3000
    },
    {
      "epoch": 238.46,
      "learning_rate": 6.390872483221477e-05,
      "loss": 0.0043,
      "step": 3100
    },
    {
      "epoch": 246.15,
      "learning_rate": 6.337181208053692e-05,
      "loss": 0.0028,
      "step": 3200
    },
    {
      "epoch": 253.85,
      "learning_rate": 6.283489932885907e-05,
      "loss": 0.0025,
      "step": 3300
    },
    {
      "epoch": 261.54,
      "learning_rate": 6.229798657718121e-05,
      "loss": 0.0055,
      "step": 3400
    },
    {
      "epoch": 269.23,
      "learning_rate": 6.176107382550336e-05,
      "loss": 0.0031,
      "step": 3500
    },
    {
      "epoch": 269.23,
      "eval_accuracy": 0.8574821852731591,
      "eval_loss": 1.2649881839752197,
      "eval_runtime": 0.4055,
      "eval_samples_per_second": 197.278,
      "eval_steps_per_second": 24.66,
      "step": 3500
    },
    {
      "epoch": 276.92,
      "learning_rate": 6.122416107382551e-05,
      "loss": 0.0057,
      "step": 3600
    },
    {
      "epoch": 284.62,
      "learning_rate": 6.068724832214766e-05,
      "loss": 0.0036,
      "step": 3700
    },
    {
      "epoch": 292.31,
      "learning_rate": 6.015033557046981e-05,
      "loss": 0.0017,
      "step": 3800
    },
    {
      "epoch": 300.0,
      "learning_rate": 5.9613422818791955e-05,
      "loss": 0.003,
      "step": 3900
    },
    {
      "epoch": 307.69,
      "learning_rate": 5.90765100671141e-05,
      "loss": 0.0052,
      "step": 4000
    },
    {
      "epoch": 307.69,
      "eval_accuracy": 0.8661916072842438,
      "eval_loss": 1.126146674156189,
      "eval_runtime": 0.4039,
      "eval_samples_per_second": 198.092,
      "eval_steps_per_second": 24.761,
      "step": 4000
    },
    {
      "epoch": 315.38,
      "learning_rate": 5.853959731543625e-05,
      "loss": 0.0029,
      "step": 4100
    },
    {
      "epoch": 323.08,
      "learning_rate": 5.80026845637584e-05,
      "loss": 0.0048,
      "step": 4200
    },
    {
      "epoch": 330.77,
      "learning_rate": 5.7465771812080534e-05,
      "loss": 0.0037,
      "step": 4300
    },
    {
      "epoch": 338.46,
      "learning_rate": 5.692885906040268e-05,
      "loss": 0.0018,
      "step": 4400
    },
    {
      "epoch": 346.15,
      "learning_rate": 5.6397315436241616e-05,
      "loss": 0.0021,
      "step": 4500
    },
    {
      "epoch": 346.15,
      "eval_accuracy": 0.8630245447347585,
      "eval_loss": 1.288573980331421,
      "eval_runtime": 0.4065,
      "eval_samples_per_second": 196.825,
      "eval_steps_per_second": 24.603,
      "step": 4500
    },
    {
      "epoch": 346.15,
      "step": 4500,
      "total_flos": 1.809255033426739e+16,
      "train_loss": 0.028387814574771457,
      "train_runtime": 1178.9995,
      "train_samples_per_second": 407.125,
      "train_steps_per_second": 12.723
    }
  ],
  "max_steps": 15000,
  "num_train_epochs": 1154,
  "total_flos": 1.809255033426739e+16,
  "trial_name": null,
  "trial_params": null
}