{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.0,
  "global_step": 4936,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.41,
      "learning_rate": 8.987034035656403e-06,
      "loss": 1.5994,
      "step": 500
    },
    {
      "epoch": 0.81,
      "learning_rate": 7.974068071312805e-06,
      "loss": 0.8358,
      "step": 1000
    },
    {
      "epoch": 1.22,
      "learning_rate": 6.961102106969206e-06,
      "loss": 0.7185,
      "step": 1500
    },
    {
      "epoch": 1.62,
      "learning_rate": 5.948136142625608e-06,
      "loss": 0.6419,
      "step": 2000
    },
    {
      "epoch": 2.03,
      "learning_rate": 4.93517017828201e-06,
      "loss": 0.6228,
      "step": 2500
    },
    {
      "epoch": 2.43,
      "learning_rate": 3.922204213938412e-06,
      "loss": 0.5474,
      "step": 3000
    },
    {
      "epoch": 2.84,
      "learning_rate": 2.9092382495948142e-06,
      "loss": 0.5291,
      "step": 3500
    },
    {
      "epoch": 3.24,
      "learning_rate": 1.8962722852512159e-06,
      "loss": 0.5007,
      "step": 4000
    },
    {
      "epoch": 3.65,
      "learning_rate": 8.833063209076175e-07,
      "loss": 0.4824,
      "step": 4500
    },
    {
      "epoch": 4.0,
      "step": 4936,
      "total_flos": 2.7156576624344064e+17,
      "train_loss": 0.6981605795639261,
      "train_runtime": 13226.0556,
      "train_samples_per_second": 26.867,
      "train_steps_per_second": 0.373
    }
  ],
  "max_steps": 4936,
  "num_train_epochs": 4,
  "total_flos": 2.7156576624344064e+17,
  "trial_name": null,
  "trial_params": null
}