{
  "best_metric": 0.8730670103092784,
  "best_model_checkpoint": "bert_trainer/run-3/checkpoint-2000",
  "epoch": 1.6118633139909737,
  "eval_steps": 250,
  "global_step": 2500,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02,
      "learning_rate": 1.3880483293934951e-05,
      "loss": 1.3534,
      "step": 25
    },
    {
      "epoch": 0.03,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 1.0289,
      "step": 50
    },
    {
      "epoch": 0.05,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.8774,
      "step": 75
    },
    {
      "epoch": 0.06,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.7025,
      "step": 100
    },
    {
      "epoch": 0.08,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.8032,
      "step": 125
    },
    {
      "epoch": 0.1,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.808,
      "step": 150
    },
    {
      "epoch": 0.11,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.651,
      "step": 175
    },
    {
      "epoch": 0.13,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.6946,
      "step": 200
    },
    {
      "epoch": 0.15,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.6587,
      "step": 225
    },
    {
      "epoch": 0.16,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.5824,
      "step": 250
    },
    {
      "epoch": 0.16,
      "eval_accuracy": 0.8112113402061856,
      "eval_loss": 0.5165922045707703,
      "eval_runtime": 30.3844,
      "eval_samples_per_second": 102.158,
      "eval_steps_per_second": 6.385,
      "step": 250
    },
    {
      "epoch": 0.18,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.546,
      "step": 275
    },
    {
      "epoch": 0.19,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.5713,
      "step": 300
    },
    {
      "epoch": 0.21,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.4633,
      "step": 325
    },
    {
      "epoch": 0.23,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.5839,
      "step": 350
    },
    {
      "epoch": 0.24,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.4685,
      "step": 375
    },
    {
      "epoch": 0.26,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.5632,
      "step": 400
    },
    {
      "epoch": 0.27,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.52,
      "step": 425
    },
    {
      "epoch": 0.29,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.5565,
      "step": 450
    },
    {
      "epoch": 0.31,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.5394,
      "step": 475
    },
    {
      "epoch": 0.32,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.4944,
      "step": 500
    },
    {
      "epoch": 0.32,
      "eval_accuracy": 0.8244201030927835,
      "eval_loss": 0.4521254003047943,
      "eval_runtime": 30.375,
      "eval_samples_per_second": 102.189,
      "eval_steps_per_second": 6.387,
      "step": 500
    },
    {
      "epoch": 0.34,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.5233,
      "step": 525
    },
    {
      "epoch": 0.35,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.475,
      "step": 550
    },
    {
      "epoch": 0.37,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.4169,
      "step": 575
    },
    {
      "epoch": 0.39,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.4961,
      "step": 600
    },
    {
      "epoch": 0.4,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.5372,
      "step": 625
    },
    {
      "epoch": 0.42,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.5151,
      "step": 650
    },
    {
      "epoch": 0.44,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.5491,
      "step": 675
    },
    {
      "epoch": 0.45,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.4596,
      "step": 700
    },
    {
      "epoch": 0.47,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.4722,
      "step": 725
    },
    {
      "epoch": 0.48,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.3632,
      "step": 750
    },
    {
      "epoch": 0.48,
      "eval_accuracy": 0.8344072164948454,
      "eval_loss": 0.48342078924179077,
      "eval_runtime": 30.4129,
      "eval_samples_per_second": 102.062,
      "eval_steps_per_second": 6.379,
      "step": 750
    },
    {
      "epoch": 0.5,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.5665,
      "step": 775
    },
    {
      "epoch": 0.52,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.3748,
      "step": 800
    },
    {
      "epoch": 0.53,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.4397,
      "step": 825
    },
    {
      "epoch": 0.55,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.4141,
      "step": 850
    },
    {
      "epoch": 0.56,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.4029,
      "step": 875
    },
    {
      "epoch": 0.58,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.4191,
      "step": 900
    },
    {
      "epoch": 0.6,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.5094,
      "step": 925
    },
    {
      "epoch": 0.61,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.4308,
      "step": 950
    },
    {
      "epoch": 0.63,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.3921,
      "step": 975
    },
    {
      "epoch": 0.64,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.4381,
      "step": 1000
    },
    {
      "epoch": 0.64,
      "eval_accuracy": 0.8588917525773195,
      "eval_loss": 0.4446374177932739,
      "eval_runtime": 29.9606,
      "eval_samples_per_second": 103.603,
      "eval_steps_per_second": 6.475,
      "step": 1000
    },
    {
      "epoch": 0.66,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.4766,
      "step": 1025
    },
    {
      "epoch": 0.68,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.4957,
      "step": 1050
    },
    {
      "epoch": 0.69,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.4285,
      "step": 1075
    },
    {
      "epoch": 0.71,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.3494,
      "step": 1100
    },
    {
      "epoch": 0.73,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.4066,
      "step": 1125
    },
    {
      "epoch": 0.74,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.3833,
      "step": 1150
    },
    {
      "epoch": 0.76,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.4553,
      "step": 1175
    },
    {
      "epoch": 0.77,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.4185,
      "step": 1200
    },
    {
      "epoch": 0.79,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.3519,
      "step": 1225
    },
    {
      "epoch": 0.81,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.4848,
      "step": 1250
    },
    {
      "epoch": 0.81,
      "eval_accuracy": 0.8582474226804123,
      "eval_loss": 0.408128023147583,
      "eval_runtime": 29.8798,
      "eval_samples_per_second": 103.883,
      "eval_steps_per_second": 6.493,
      "step": 1250
    },
    {
      "epoch": 0.82,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.539,
      "step": 1275
    },
    {
      "epoch": 0.84,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.3632,
      "step": 1300
    },
    {
      "epoch": 0.85,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.3861,
      "step": 1325
    },
    {
      "epoch": 0.87,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.5181,
      "step": 1350
    },
    {
      "epoch": 0.89,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.4158,
      "step": 1375
    },
    {
      "epoch": 0.9,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.3584,
      "step": 1400
    },
    {
      "epoch": 0.92,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.4162,
      "step": 1425
    },
    {
      "epoch": 0.93,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.4607,
      "step": 1450
    },
    {
      "epoch": 0.95,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.4146,
      "step": 1475
    },
    {
      "epoch": 0.97,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.3954,
      "step": 1500
    },
    {
      "epoch": 0.97,
      "eval_accuracy": 0.8714561855670103,
      "eval_loss": 0.38059666752815247,
      "eval_runtime": 29.8772,
      "eval_samples_per_second": 103.892,
      "eval_steps_per_second": 6.493,
      "step": 1500
    },
    {
      "epoch": 0.98,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.3881,
      "step": 1525
    },
    {
      "epoch": 1.0,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.3825,
      "step": 1550
    },
    {
      "epoch": 1.02,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.3082,
      "step": 1575
    },
    {
      "epoch": 1.03,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.3367,
      "step": 1600
    },
    {
      "epoch": 1.05,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.3191,
      "step": 1625
    },
    {
      "epoch": 1.06,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.3298,
      "step": 1650
    },
    {
      "epoch": 1.08,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.2631,
      "step": 1675
    },
    {
      "epoch": 1.1,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.1897,
      "step": 1700
    },
    {
      "epoch": 1.11,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.3901,
      "step": 1725
    },
    {
      "epoch": 1.13,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.3268,
      "step": 1750
    },
    {
      "epoch": 1.13,
      "eval_accuracy": 0.8714561855670103,
      "eval_loss": 0.5426503419876099,
      "eval_runtime": 29.922,
      "eval_samples_per_second": 103.737,
      "eval_steps_per_second": 6.484,
      "step": 1750
    },
    {
      "epoch": 1.14,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.2771,
      "step": 1775
    },
    {
      "epoch": 1.16,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.3346,
      "step": 1800
    },
    {
      "epoch": 1.18,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.312,
      "step": 1825
    },
    {
      "epoch": 1.19,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.3482,
      "step": 1850
    },
    {
      "epoch": 1.21,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.3186,
      "step": 1875
    },
    {
      "epoch": 1.23,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.317,
      "step": 1900
    },
    {
      "epoch": 1.24,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.3214,
      "step": 1925
    },
    {
      "epoch": 1.26,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.2073,
      "step": 1950
    },
    {
      "epoch": 1.27,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.3709,
      "step": 1975
    },
    {
      "epoch": 1.29,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.2492,
      "step": 2000
    },
    {
      "epoch": 1.29,
      "eval_accuracy": 0.8730670103092784,
      "eval_loss": 0.4012953042984009,
      "eval_runtime": 29.8764,
      "eval_samples_per_second": 103.895,
      "eval_steps_per_second": 6.493,
      "step": 2000
    },
    {
      "epoch": 1.31,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.2831,
      "step": 2025
    },
    {
      "epoch": 1.32,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.2479,
      "step": 2050
    },
    {
      "epoch": 1.34,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.2467,
      "step": 2075
    },
    {
      "epoch": 1.35,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.332,
      "step": 2100
    },
    {
      "epoch": 1.37,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.2565,
      "step": 2125
    },
    {
      "epoch": 1.39,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.4016,
      "step": 2150
    },
    {
      "epoch": 1.4,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.1485,
      "step": 2175
    },
    {
      "epoch": 1.42,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.3333,
      "step": 2200
    },
    {
      "epoch": 1.43,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.3541,
      "step": 2225
    },
    {
      "epoch": 1.45,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.1829,
      "step": 2250
    },
    {
      "epoch": 1.45,
      "eval_accuracy": 0.8704896907216495,
      "eval_loss": 0.49169790744781494,
      "eval_runtime": 29.9217,
      "eval_samples_per_second": 103.737,
      "eval_steps_per_second": 6.484,
      "step": 2250
    },
    {
      "epoch": 1.47,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.2888,
      "step": 2275
    },
    {
      "epoch": 1.48,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.2472,
      "step": 2300
    },
    {
      "epoch": 1.5,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.3547,
      "step": 2325
    },
    {
      "epoch": 1.52,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.1869,
      "step": 2350
    },
    {
      "epoch": 1.53,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.3094,
      "step": 2375
    },
    {
      "epoch": 1.55,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.314,
      "step": 2400
    },
    {
      "epoch": 1.56,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.3161,
      "step": 2425
    },
    {
      "epoch": 1.58,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.3445,
      "step": 2450
    },
    {
      "epoch": 1.6,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.4138,
      "step": 2475
    },
    {
      "epoch": 1.61,
      "learning_rate": 2.7760966587869903e-05,
      "loss": 0.2214,
      "step": 2500
    },
    {
      "epoch": 1.61,
      "eval_accuracy": 0.852770618556701,
      "eval_loss": 0.5358509421348572,
      "eval_runtime": 29.9717,
      "eval_samples_per_second": 103.564,
      "eval_steps_per_second": 6.473,
      "step": 2500
    }
  ],
  "logging_steps": 25,
  "max_steps": 3000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 250,
  "total_flos": 5261836612995072.0,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 2.7760966587869903e-05,
    "per_device_eval_batch_size": 16,
    "per_device_train_batch_size": 8
  }
}