{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.9333333333333336,
  "eval_steps": 500,
  "global_step": 90,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.17,
      "learning_rate": 0.00011111111111111112,
      "loss": 1.195,
      "step": 5
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.00019753086419753085,
      "loss": 1.0234,
      "step": 10
    },
    {
      "epoch": 0.5,
      "learning_rate": 0.0001851851851851852,
      "loss": 0.9328,
      "step": 15
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.0001728395061728395,
      "loss": 0.8382,
      "step": 20
    },
    {
      "epoch": 0.83,
      "learning_rate": 0.00016296296296296295,
      "loss": 0.7861,
      "step": 25
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.0001506172839506173,
      "loss": 0.7378,
      "step": 30
    },
    {
      "epoch": 1.13,
      "learning_rate": 0.0001382716049382716,
      "loss": 0.6849,
      "step": 35
    },
    {
      "epoch": 1.3,
      "learning_rate": 0.00012592592592592592,
      "loss": 0.6841,
      "step": 40
    },
    {
      "epoch": 1.47,
      "learning_rate": 0.00011358024691358025,
      "loss": 0.5974,
      "step": 45
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.00010123456790123458,
      "loss": 0.6252,
      "step": 50
    },
    {
      "epoch": 1.8,
      "learning_rate": 8.888888888888889e-05,
      "loss": 0.5852,
      "step": 55
    },
    {
      "epoch": 1.97,
      "learning_rate": 7.65432098765432e-05,
      "loss": 0.589,
      "step": 60
    },
    {
      "epoch": 2.1,
      "learning_rate": 6.419753086419753e-05,
      "loss": 0.5679,
      "step": 65
    },
    {
      "epoch": 2.27,
      "learning_rate": 5.185185185185185e-05,
      "loss": 0.5437,
      "step": 70
    },
    {
      "epoch": 2.43,
      "learning_rate": 3.950617283950617e-05,
      "loss": 0.5003,
      "step": 75
    },
    {
      "epoch": 2.6,
      "learning_rate": 2.7160493827160493e-05,
      "loss": 0.5405,
      "step": 80
    },
    {
      "epoch": 2.77,
      "learning_rate": 1.4814814814814815e-05,
      "loss": 0.5395,
      "step": 85
    },
    {
      "epoch": 2.93,
      "learning_rate": 2.469135802469136e-06,
      "loss": 0.5096,
      "step": 90
    }
  ],
  "logging_steps": 5,
  "max_steps": 90,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 1.57427177619456e+16,
  "trial_name": null,
  "trial_params": null
}