{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 37.004,
  "global_step": 900,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0,
      "learning_rate": 2e-06,
      "loss": 10.9377,
      "step": 1
    },
    {
      "epoch": 2.0,
      "learning_rate": 9.999999999999999e-05,
      "loss": 9.4009,
      "step": 50
    },
    {
      "epoch": 4.0,
      "learning_rate": 0.00019999999999999998,
      "loss": 7.1915,
      "step": 100
    },
    {
      "epoch": 6.0,
      "learning_rate": 0.0003,
      "loss": 6.4074,
      "step": 150
    },
    {
      "epoch": 8.0,
      "learning_rate": 0.00039999999999999996,
      "loss": 5.6889,
      "step": 200
    },
    {
      "epoch": 10.0,
      "learning_rate": 0.0005,
      "loss": 5.326,
      "step": 250
    },
    {
      "epoch": 12.0,
      "learning_rate": 0.0006,
      "loss": 5.0873,
      "step": 300
    },
    {
      "epoch": 14.0,
      "learning_rate": 0.0005888888888888889,
      "loss": 4.7619,
      "step": 350
    },
    {
      "epoch": 16.01,
      "learning_rate": 0.0005777777777777776,
      "loss": 4.4374,
      "step": 400
    },
    {
      "epoch": 18.01,
      "learning_rate": 0.0005666666666666666,
      "loss": 4.2331,
      "step": 450
    },
    {
      "epoch": 20.01,
      "learning_rate": 0.0005555555555555556,
      "loss": 4.0853,
      "step": 500
    },
    {
      "epoch": 22.01,
      "learning_rate": 0.0005444444444444444,
      "loss": 3.9849,
      "step": 550
    },
    {
      "epoch": 24.01,
      "learning_rate": 0.0005333333333333333,
      "loss": 3.8903,
      "step": 600
    },
    {
      "epoch": 27.0,
      "learning_rate": 0.0005222222222222222,
      "loss": 3.8802,
      "step": 650
    },
    {
      "epoch": 29.0,
      "learning_rate": 0.0005111111111111111,
      "loss": 3.748,
      "step": 700
    },
    {
      "epoch": 31.0,
      "learning_rate": 0.0005,
      "loss": 3.6859,
      "step": 750
    },
    {
      "epoch": 33.0,
      "learning_rate": 0.0004888888888888889,
      "loss": 3.6323,
      "step": 800
    },
    {
      "epoch": 35.0,
      "learning_rate": 0.0004777777777777777,
      "loss": 3.5863,
      "step": 850
    },
    {
      "epoch": 37.0,
      "learning_rate": 0.0004666666666666666,
      "loss": 3.5434,
      "step": 900
    }
  ],
  "max_steps": 3000,
  "num_train_epochs": 9223372036854775807,
  "total_flos": 4.163350221029376e+16,
  "trial_name": null,
  "trial_params": null
}