{
  "best_metric": 1.5155326128005981,
  "best_model_checkpoint": "/temp/mt5_base-qg-af-oficial/checkpoint-5544",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 6930,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.07,
      "learning_rate": 9.855699855699856e-05,
      "loss": 12.1403,
      "step": 100
    },
    {
      "epoch": 0.14,
      "learning_rate": 9.711399711399713e-05,
      "loss": 3.3838,
      "step": 200
    },
    {
      "epoch": 0.22,
      "learning_rate": 9.567099567099568e-05,
      "loss": 2.2366,
      "step": 300
    },
    {
      "epoch": 0.29,
      "learning_rate": 9.422799422799424e-05,
      "loss": 2.1106,
      "step": 400
    },
    {
      "epoch": 0.36,
      "learning_rate": 9.278499278499279e-05,
      "loss": 2.0552,
      "step": 500
    },
    {
      "epoch": 0.43,
      "learning_rate": 9.134199134199136e-05,
      "loss": 2.0231,
      "step": 600
    },
    {
      "epoch": 0.51,
      "learning_rate": 8.98989898989899e-05,
      "loss": 1.974,
      "step": 700
    },
    {
      "epoch": 0.58,
      "learning_rate": 8.845598845598845e-05,
      "loss": 1.9583,
      "step": 800
    },
    {
      "epoch": 0.65,
      "learning_rate": 8.701298701298701e-05,
      "loss": 1.8944,
      "step": 900
    },
    {
      "epoch": 0.72,
      "learning_rate": 8.556998556998557e-05,
      "loss": 1.8719,
      "step": 1000
    },
    {
      "epoch": 0.79,
      "learning_rate": 8.412698412698413e-05,
      "loss": 1.8895,
      "step": 1100
    },
    {
      "epoch": 0.87,
      "learning_rate": 8.268398268398268e-05,
      "loss": 1.8459,
      "step": 1200
    },
    {
      "epoch": 0.94,
      "learning_rate": 8.124098124098124e-05,
      "loss": 1.8203,
      "step": 1300
    },
    {
      "epoch": 1.0,
      "eval_loss": 1.6114802360534668,
      "eval_runtime": 33.1219,
      "eval_samples_per_second": 196.577,
      "eval_steps_per_second": 3.08,
      "step": 1386
    },
    {
      "epoch": 1.01,
      "learning_rate": 7.97979797979798e-05,
      "loss": 1.8007,
      "step": 1400
    },
    {
      "epoch": 1.08,
      "learning_rate": 7.835497835497836e-05,
      "loss": 1.7247,
      "step": 1500
    },
    {
      "epoch": 1.15,
      "learning_rate": 7.691197691197691e-05,
      "loss": 1.7288,
      "step": 1600
    },
    {
      "epoch": 1.23,
      "learning_rate": 7.546897546897548e-05,
      "loss": 1.7174,
      "step": 1700
    },
    {
      "epoch": 1.3,
      "learning_rate": 7.402597402597404e-05,
      "loss": 1.6836,
      "step": 1800
    },
    {
      "epoch": 1.37,
      "learning_rate": 7.258297258297259e-05,
      "loss": 1.7096,
      "step": 1900
    },
    {
      "epoch": 1.44,
      "learning_rate": 7.113997113997114e-05,
      "loss": 1.681,
      "step": 2000
    },
    {
      "epoch": 1.52,
      "learning_rate": 6.96969696969697e-05,
      "loss": 1.6965,
      "step": 2100
    },
    {
      "epoch": 1.59,
      "learning_rate": 6.825396825396825e-05,
      "loss": 1.6696,
      "step": 2200
    },
    {
      "epoch": 1.66,
      "learning_rate": 6.681096681096681e-05,
      "loss": 1.6779,
      "step": 2300
    },
    {
      "epoch": 1.73,
      "learning_rate": 6.536796536796536e-05,
      "loss": 1.6736,
      "step": 2400
    },
    {
      "epoch": 1.8,
      "learning_rate": 6.392496392496393e-05,
      "loss": 1.6729,
      "step": 2500
    },
    {
      "epoch": 1.88,
      "learning_rate": 6.248196248196248e-05,
      "loss": 1.6795,
      "step": 2600
    },
    {
      "epoch": 1.95,
      "learning_rate": 6.103896103896104e-05,
      "loss": 1.6624,
      "step": 2700
    },
    {
      "epoch": 2.0,
      "eval_loss": 1.5542315244674683,
      "eval_runtime": 33.2024,
      "eval_samples_per_second": 196.101,
      "eval_steps_per_second": 3.072,
      "step": 2772
    },
    {
      "epoch": 2.02,
      "learning_rate": 5.959595959595959e-05,
      "loss": 1.6222,
      "step": 2800
    },
    {
      "epoch": 2.09,
      "learning_rate": 5.815295815295816e-05,
      "loss": 1.5908,
      "step": 2900
    },
    {
      "epoch": 2.16,
      "learning_rate": 5.6709956709956715e-05,
      "loss": 1.5692,
      "step": 3000
    },
    {
      "epoch": 2.24,
      "learning_rate": 5.526695526695527e-05,
      "loss": 1.5753,
      "step": 3100
    },
    {
      "epoch": 2.31,
      "learning_rate": 5.382395382395382e-05,
      "loss": 1.5802,
      "step": 3200
    },
    {
      "epoch": 2.38,
      "learning_rate": 5.2380952380952384e-05,
      "loss": 1.5546,
      "step": 3300
    },
    {
      "epoch": 2.45,
      "learning_rate": 5.093795093795094e-05,
      "loss": 1.5665,
      "step": 3400
    },
    {
      "epoch": 2.53,
      "learning_rate": 4.94949494949495e-05,
      "loss": 1.5381,
      "step": 3500
    },
    {
      "epoch": 2.6,
      "learning_rate": 4.8051948051948054e-05,
      "loss": 1.5634,
      "step": 3600
    },
    {
      "epoch": 2.67,
      "learning_rate": 4.6608946608946615e-05,
      "loss": 1.5554,
      "step": 3700
    },
    {
      "epoch": 2.74,
      "learning_rate": 4.516594516594517e-05,
      "loss": 1.5645,
      "step": 3800
    },
    {
      "epoch": 2.81,
      "learning_rate": 4.3722943722943724e-05,
      "loss": 1.5652,
      "step": 3900
    },
    {
      "epoch": 2.89,
      "learning_rate": 4.227994227994228e-05,
      "loss": 1.5615,
      "step": 4000
    },
    {
      "epoch": 2.96,
      "learning_rate": 4.083694083694084e-05,
      "loss": 1.5489,
      "step": 4100
    },
    {
      "epoch": 3.0,
      "eval_loss": 1.5325491428375244,
      "eval_runtime": 33.1544,
      "eval_samples_per_second": 196.384,
      "eval_steps_per_second": 3.077,
      "step": 4158
    },
    {
      "epoch": 3.03,
      "learning_rate": 3.939393939393939e-05,
      "loss": 1.5307,
      "step": 4200
    },
    {
      "epoch": 3.1,
      "learning_rate": 3.7950937950937954e-05,
      "loss": 1.4781,
      "step": 4300
    },
    {
      "epoch": 3.17,
      "learning_rate": 3.650793650793651e-05,
      "loss": 1.4875,
      "step": 4400
    },
    {
      "epoch": 3.25,
      "learning_rate": 3.506493506493507e-05,
      "loss": 1.4927,
      "step": 4500
    },
    {
      "epoch": 3.32,
      "learning_rate": 3.3621933621933624e-05,
      "loss": 1.4904,
      "step": 4600
    },
    {
      "epoch": 3.39,
      "learning_rate": 3.217893217893218e-05,
      "loss": 1.5048,
      "step": 4700
    },
    {
      "epoch": 3.46,
      "learning_rate": 3.073593073593073e-05,
      "loss": 1.4868,
      "step": 4800
    },
    {
      "epoch": 3.54,
      "learning_rate": 2.9292929292929294e-05,
      "loss": 1.4631,
      "step": 4900
    },
    {
      "epoch": 3.61,
      "learning_rate": 2.7849927849927855e-05,
      "loss": 1.4957,
      "step": 5000
    },
    {
      "epoch": 3.68,
      "learning_rate": 2.640692640692641e-05,
      "loss": 1.488,
      "step": 5100
    },
    {
      "epoch": 3.75,
      "learning_rate": 2.4963924963924963e-05,
      "loss": 1.4898,
      "step": 5200
    },
    {
      "epoch": 3.82,
      "learning_rate": 2.352092352092352e-05,
      "loss": 1.4818,
      "step": 5300
    },
    {
      "epoch": 3.9,
      "learning_rate": 2.207792207792208e-05,
      "loss": 1.4967,
      "step": 5400
    },
    {
      "epoch": 3.97,
      "learning_rate": 2.0634920634920636e-05,
      "loss": 1.4926,
      "step": 5500
    },
    {
      "epoch": 4.0,
      "eval_loss": 1.5155326128005981,
      "eval_runtime": 33.2094,
      "eval_samples_per_second": 196.059,
      "eval_steps_per_second": 3.071,
      "step": 5544
    },
    {
      "epoch": 4.04,
      "learning_rate": 1.919191919191919e-05,
      "loss": 1.4528,
      "step": 5600
    },
    {
      "epoch": 4.11,
      "learning_rate": 1.7748917748917752e-05,
      "loss": 1.4396,
      "step": 5700
    },
    {
      "epoch": 4.18,
      "learning_rate": 1.630591630591631e-05,
      "loss": 1.4457,
      "step": 5800
    },
    {
      "epoch": 4.26,
      "learning_rate": 1.4862914862914865e-05,
      "loss": 1.4471,
      "step": 5900
    },
    {
      "epoch": 4.33,
      "learning_rate": 1.3419913419913421e-05,
      "loss": 1.4543,
      "step": 6000
    },
    {
      "epoch": 4.4,
      "learning_rate": 1.1976911976911977e-05,
      "loss": 1.4612,
      "step": 6100
    },
    {
      "epoch": 4.47,
      "learning_rate": 1.0533910533910535e-05,
      "loss": 1.4536,
      "step": 6200
    },
    {
      "epoch": 4.55,
      "learning_rate": 9.090909090909091e-06,
      "loss": 1.4451,
      "step": 6300
    },
    {
      "epoch": 4.62,
      "learning_rate": 7.647907647907649e-06,
      "loss": 1.4491,
      "step": 6400
    },
    {
      "epoch": 4.69,
      "learning_rate": 6.204906204906205e-06,
      "loss": 1.4394,
      "step": 6500
    },
    {
      "epoch": 4.76,
      "learning_rate": 4.7619047619047615e-06,
      "loss": 1.4367,
      "step": 6600
    },
    {
      "epoch": 4.83,
      "learning_rate": 3.318903318903319e-06,
      "loss": 1.445,
      "step": 6700
    },
    {
      "epoch": 4.91,
      "learning_rate": 1.875901875901876e-06,
      "loss": 1.4478,
      "step": 6800
    },
    {
      "epoch": 4.98,
      "learning_rate": 4.329004329004329e-07,
      "loss": 1.4687,
      "step": 6900
    },
    {
      "epoch": 5.0,
      "eval_loss": 1.5195423364639282,
      "eval_runtime": 33.3056,
      "eval_samples_per_second": 195.493,
      "eval_steps_per_second": 3.063,
      "step": 6930
    },
    {
      "epoch": 5.0,
      "step": 6930,
      "total_flos": 2.6579255601659904e+17,
      "train_loss": 1.7969476851205977,
      "train_runtime": 4653.9626,
      "train_samples_per_second": 47.63,
      "train_steps_per_second": 1.489
    }
  ],
  "logging_steps": 100,
  "max_steps": 6930,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 2.6579255601659904e+17,
  "trial_name": null,
  "trial_params": null
}