{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.11152872422292362,
  "global_step": 10000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.01,
      "learning_rate": 1.4999999999999998e-06,
      "loss": 0.9292,
      "step": 500
    },
    {
      "epoch": 0.01,
      "learning_rate": 2.9999999999999997e-06,
      "loss": 0.7143,
      "step": 1000
    },
    {
      "epoch": 0.01,
      "eval_loss": 0.6849376559257507,
      "eval_runtime": 2.3913,
      "eval_samples_per_second": 960.554,
      "eval_steps_per_second": 15.054,
      "step": 1000
    },
    {
      "epoch": 0.02,
      "learning_rate": 4.499999999999999e-06,
      "loss": 0.6841,
      "step": 1500
    },
    {
      "epoch": 0.02,
      "learning_rate": 5.999999999999999e-06,
      "loss": 0.6817,
      "step": 2000
    },
    {
      "epoch": 0.02,
      "eval_loss": 0.678993284702301,
      "eval_runtime": 2.3276,
      "eval_samples_per_second": 986.872,
      "eval_steps_per_second": 15.467,
      "step": 2000
    },
    {
      "epoch": 0.03,
      "learning_rate": 7.499999999999999e-06,
      "loss": 0.6813,
      "step": 2500
    },
    {
      "epoch": 0.03,
      "learning_rate": 8.999999999999999e-06,
      "loss": 0.6806,
      "step": 3000
    },
    {
      "epoch": 0.03,
      "eval_loss": 0.6789333820343018,
      "eval_runtime": 2.3788,
      "eval_samples_per_second": 965.594,
      "eval_steps_per_second": 15.133,
      "step": 3000
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.05e-05,
      "loss": 0.6806,
      "step": 3500
    },
    {
      "epoch": 0.04,
      "learning_rate": 1.1999999999999999e-05,
      "loss": 0.6804,
      "step": 4000
    },
    {
      "epoch": 0.04,
      "eval_loss": 0.6781470775604248,
      "eval_runtime": 2.452,
      "eval_samples_per_second": 936.772,
      "eval_steps_per_second": 14.682,
      "step": 4000
    },
    {
      "epoch": 0.05,
      "learning_rate": 1.3499999999999998e-05,
      "loss": 0.6799,
      "step": 4500
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.4999999999999999e-05,
      "loss": 0.6797,
      "step": 5000
    },
    {
      "epoch": 0.06,
      "eval_loss": 0.6779675483703613,
      "eval_runtime": 2.3912,
      "eval_samples_per_second": 960.599,
      "eval_steps_per_second": 15.055,
      "step": 5000
    },
    {
      "epoch": 0.06,
      "learning_rate": 1.6499999999999998e-05,
      "loss": 0.6796,
      "step": 5500
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.7999999999999997e-05,
      "loss": 0.6793,
      "step": 6000
    },
    {
      "epoch": 0.07,
      "eval_loss": 0.677739143371582,
      "eval_runtime": 2.3548,
      "eval_samples_per_second": 975.453,
      "eval_steps_per_second": 15.288,
      "step": 6000
    },
    {
      "epoch": 0.07,
      "learning_rate": 1.95e-05,
      "loss": 0.6791,
      "step": 6500
    },
    {
      "epoch": 0.08,
      "learning_rate": 2.1e-05,
      "loss": 0.6787,
      "step": 7000
    },
    {
      "epoch": 0.08,
      "eval_loss": 0.6776532530784607,
      "eval_runtime": 2.3673,
      "eval_samples_per_second": 970.284,
      "eval_steps_per_second": 15.207,
      "step": 7000
    },
    {
      "epoch": 0.08,
      "learning_rate": 2.2499999999999998e-05,
      "loss": 0.6784,
      "step": 7500
    },
    {
      "epoch": 0.09,
      "learning_rate": 2.3999999999999997e-05,
      "loss": 0.6775,
      "step": 8000
    },
    {
      "epoch": 0.09,
      "eval_loss": 0.6731997728347778,
      "eval_runtime": 2.3959,
      "eval_samples_per_second": 958.715,
      "eval_steps_per_second": 15.026,
      "step": 8000
    },
    {
      "epoch": 0.09,
      "learning_rate": 2.55e-05,
      "loss": 0.6689,
      "step": 8500
    },
    {
      "epoch": 0.1,
      "learning_rate": 2.6999999999999996e-05,
      "loss": 0.6553,
      "step": 9000
    },
    {
      "epoch": 0.1,
      "eval_loss": 0.645298957824707,
      "eval_runtime": 2.3807,
      "eval_samples_per_second": 964.84,
      "eval_steps_per_second": 15.122,
      "step": 9000
    },
    {
      "epoch": 0.11,
      "learning_rate": 2.8499999999999998e-05,
      "loss": 0.6445,
      "step": 9500
    },
    {
      "epoch": 0.11,
      "learning_rate": 2.9999999999999997e-05,
      "loss": 0.6351,
      "step": 10000
    },
    {
      "epoch": 0.11,
      "eval_loss": 0.6254131197929382,
      "eval_runtime": 2.3999,
      "eval_samples_per_second": 957.11,
      "eval_steps_per_second": 15.0,
      "step": 10000
    }
  ],
  "max_steps": 1000000,
  "num_train_epochs": 12,
  "total_flos": 7.010026103034715e+20,
  "trial_name": null,
  "trial_params": null
}