{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 9.652509652509652,
  "eval_steps": 1000,
  "global_step": 5000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.58,
      "learning_rate": 7.239382239382239e-07,
      "loss": 2.3335,
      "step": 300
    },
    {
      "epoch": 1.16,
      "learning_rate": 1.4478764478764478e-06,
      "loss": 2.322,
      "step": 600
    },
    {
      "epoch": 1.74,
      "learning_rate": 2.171814671814672e-06,
      "loss": 2.268,
      "step": 900
    },
    {
      "epoch": 1.93,
      "eval_loss": 2.2363617420196533,
      "eval_runtime": 56.2328,
      "eval_samples_per_second": 51.98,
      "eval_steps_per_second": 3.254,
      "step": 1000
    },
    {
      "epoch": 2.32,
      "learning_rate": 2.8957528957528956e-06,
      "loss": 2.2427,
      "step": 1200
    },
    {
      "epoch": 2.9,
      "learning_rate": 3.61969111969112e-06,
      "loss": 2.2085,
      "step": 1500
    },
    {
      "epoch": 3.47,
      "learning_rate": 4.343629343629344e-06,
      "loss": 2.1902,
      "step": 1800
    },
    {
      "epoch": 3.86,
      "eval_loss": 2.1604719161987305,
      "eval_runtime": 56.4281,
      "eval_samples_per_second": 51.8,
      "eval_steps_per_second": 3.243,
      "step": 2000
    },
    {
      "epoch": 4.05,
      "learning_rate": 5.067567567567568e-06,
      "loss": 2.173,
      "step": 2100
    },
    {
      "epoch": 4.63,
      "learning_rate": 5.791505791505791e-06,
      "loss": 2.1517,
      "step": 2400
    },
    {
      "epoch": 5.21,
      "learning_rate": 6.515444015444016e-06,
      "loss": 2.135,
      "step": 2700
    },
    {
      "epoch": 5.79,
      "learning_rate": 7.23938223938224e-06,
      "loss": 2.1156,
      "step": 3000
    },
    {
      "epoch": 5.79,
      "eval_loss": 2.108267068862915,
      "eval_runtime": 56.4468,
      "eval_samples_per_second": 51.783,
      "eval_steps_per_second": 3.242,
      "step": 3000
    },
    {
      "epoch": 6.37,
      "learning_rate": 7.963320463320465e-06,
      "loss": 2.1038,
      "step": 3300
    },
    {
      "epoch": 6.95,
      "learning_rate": 8.687258687258689e-06,
      "loss": 2.1021,
      "step": 3600
    },
    {
      "epoch": 7.53,
      "learning_rate": 9.41119691119691e-06,
      "loss": 2.0882,
      "step": 3900
    },
    {
      "epoch": 7.72,
      "eval_loss": 2.08085298538208,
      "eval_runtime": 56.4755,
      "eval_samples_per_second": 51.757,
      "eval_steps_per_second": 3.24,
      "step": 4000
    },
    {
      "epoch": 8.11,
      "learning_rate": 9.99549483102341e-06,
      "loss": 2.0834,
      "step": 4200
    },
    {
      "epoch": 8.69,
      "learning_rate": 9.819006680234513e-06,
      "loss": 2.0744,
      "step": 4500
    },
    {
      "epoch": 9.27,
      "learning_rate": 9.394326698914229e-06,
      "loss": 2.0565,
      "step": 4800
    },
    {
      "epoch": 9.65,
      "eval_loss": 2.062009811401367,
      "eval_runtime": 56.2872,
      "eval_samples_per_second": 51.93,
      "eval_steps_per_second": 3.251,
      "step": 5000
    }
  ],
  "logging_steps": 300,
  "max_steps": 8288,
  "num_train_epochs": 16,
  "save_steps": 1000,
  "total_flos": 4.3343543836800614e+17,
  "trial_name": null,
  "trial_params": null
}