{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 100.0,
  "eval_steps": 500,
  "global_step": 15700,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 3.18,
      "learning_rate": 4.840764331210191e-05,
      "loss": 1.2813,
      "step": 500
    },
    {
      "epoch": 6.37,
      "learning_rate": 4.681528662420383e-05,
      "loss": 0.2415,
      "step": 1000
    },
    {
      "epoch": 9.55,
      "learning_rate": 4.522292993630574e-05,
      "loss": 0.1252,
      "step": 1500
    },
    {
      "epoch": 12.74,
      "learning_rate": 4.3630573248407646e-05,
      "loss": 0.0815,
      "step": 2000
    },
    {
      "epoch": 15.92,
      "learning_rate": 4.2038216560509556e-05,
      "loss": 0.0684,
      "step": 2500
    },
    {
      "epoch": 19.11,
      "learning_rate": 4.044585987261147e-05,
      "loss": 0.0612,
      "step": 3000
    },
    {
      "epoch": 22.29,
      "learning_rate": 3.885350318471338e-05,
      "loss": 0.0629,
      "step": 3500
    },
    {
      "epoch": 25.48,
      "learning_rate": 3.7261146496815283e-05,
      "loss": 0.0734,
      "step": 4000
    },
    {
      "epoch": 28.66,
      "learning_rate": 3.56687898089172e-05,
      "loss": 0.0968,
      "step": 4500
    },
    {
      "epoch": 31.85,
      "learning_rate": 3.407643312101911e-05,
      "loss": 0.1412,
      "step": 5000
    },
    {
      "epoch": 35.03,
      "learning_rate": 3.248407643312102e-05,
      "loss": 0.2021,
      "step": 5500
    },
    {
      "epoch": 38.22,
      "learning_rate": 3.089171974522293e-05,
      "loss": 0.2834,
      "step": 6000
    },
    {
      "epoch": 41.4,
      "learning_rate": 2.929936305732484e-05,
      "loss": 0.3706,
      "step": 6500
    },
    {
      "epoch": 44.59,
      "learning_rate": 2.7707006369426753e-05,
      "loss": 0.3453,
      "step": 7000
    },
    {
      "epoch": 47.77,
      "learning_rate": 2.6114649681528662e-05,
      "loss": 0.2912,
      "step": 7500
    },
    {
      "epoch": 50.96,
      "learning_rate": 2.4522292993630575e-05,
      "loss": 0.2182,
      "step": 8000
    },
    {
      "epoch": 54.14,
      "learning_rate": 2.2929936305732484e-05,
      "loss": 0.1055,
      "step": 8500
    },
    {
      "epoch": 57.32,
      "learning_rate": 2.1337579617834397e-05,
      "loss": 0.0526,
      "step": 9000
    },
    {
      "epoch": 60.51,
      "learning_rate": 1.974522292993631e-05,
      "loss": 0.03,
      "step": 9500
    },
    {
      "epoch": 63.69,
      "learning_rate": 1.8152866242038215e-05,
      "loss": 0.0182,
      "step": 10000
    },
    {
      "epoch": 66.88,
      "learning_rate": 1.6560509554140128e-05,
      "loss": 0.0138,
      "step": 10500
    },
    {
      "epoch": 70.06,
      "learning_rate": 1.4968152866242039e-05,
      "loss": 0.0105,
      "step": 11000
    },
    {
      "epoch": 73.25,
      "learning_rate": 1.337579617834395e-05,
      "loss": 0.0113,
      "step": 11500
    },
    {
      "epoch": 76.43,
      "learning_rate": 1.178343949044586e-05,
      "loss": 0.0086,
      "step": 12000
    },
    {
      "epoch": 79.62,
      "learning_rate": 1.0191082802547772e-05,
      "loss": 0.0073,
      "step": 12500
    },
    {
      "epoch": 82.8,
      "learning_rate": 8.598726114649681e-06,
      "loss": 0.007,
      "step": 13000
    },
    {
      "epoch": 85.99,
      "learning_rate": 7.006369426751593e-06,
      "loss": 0.0049,
      "step": 13500
    },
    {
      "epoch": 89.17,
      "learning_rate": 5.414012738853504e-06,
      "loss": 0.0047,
      "step": 14000
    },
    {
      "epoch": 92.36,
      "learning_rate": 3.821656050955414e-06,
      "loss": 0.0031,
      "step": 14500
    },
    {
      "epoch": 95.54,
      "learning_rate": 2.229299363057325e-06,
      "loss": 0.0043,
      "step": 15000
    },
    {
      "epoch": 98.73,
      "learning_rate": 6.369426751592357e-07,
      "loss": 0.0035,
      "step": 15500
    },
    {
      "epoch": 100.0,
      "step": 15700,
      "total_flos": 9.7986283776e+16,
      "train_loss": 0.13472519155520543,
      "train_runtime": 4354.3545,
      "train_samples_per_second": 114.828,
      "train_steps_per_second": 3.606
    }
  ],
  "logging_steps": 500,
  "max_steps": 15700,
  "num_train_epochs": 100,
  "save_steps": 500,
  "total_flos": 9.7986283776e+16,
  "trial_name": null,
  "trial_params": null
}