trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 10767,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.14,
      "learning_rate": 0.0001,
      "loss": 1.703,
      "step": 500
    },
    {
      "epoch": 0.28,
      "learning_rate": 0.0001,
      "loss": 1.5833,
      "step": 1000
    },
    {
      "epoch": 0.42,
      "learning_rate": 0.0001,
      "loss": 1.5444,
      "step": 1500
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0001,
      "loss": 1.5199,
      "step": 2000
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.0001,
      "loss": 1.4972,
      "step": 2500
    },
    {
      "epoch": 0.84,
      "learning_rate": 0.0001,
      "loss": 1.4789,
      "step": 3000
    },
    {
      "epoch": 0.98,
      "learning_rate": 0.0001,
      "loss": 1.448,
      "step": 3500
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.0001,
      "loss": 1.3726,
      "step": 4000
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.0001,
      "loss": 1.3587,
      "step": 4500
    },
    {
      "epoch": 1.39,
      "learning_rate": 0.0001,
      "loss": 1.3301,
      "step": 5000
    },
    {
      "epoch": 1.53,
      "learning_rate": 0.0001,
      "loss": 1.3619,
      "step": 5500
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.0001,
      "loss": 1.3321,
      "step": 6000
    },
    {
      "epoch": 1.81,
      "learning_rate": 0.0001,
      "loss": 1.3311,
      "step": 6500
    },
    {
      "epoch": 1.95,
      "learning_rate": 0.0001,
      "loss": 1.328,
      "step": 7000
    },
    {
      "epoch": 2.09,
      "learning_rate": 0.0001,
      "loss": 1.2657,
      "step": 7500
    },
    {
      "epoch": 2.23,
      "learning_rate": 0.0001,
      "loss": 1.2526,
      "step": 8000
    },
    {
      "epoch": 2.37,
      "learning_rate": 0.0001,
      "loss": 1.2357,
      "step": 8500
    },
    {
      "epoch": 2.51,
      "learning_rate": 0.0001,
      "loss": 1.234,
      "step": 9000
    },
    {
      "epoch": 2.65,
      "learning_rate": 0.0001,
      "loss": 1.2335,
      "step": 9500
    },
    {
      "epoch": 2.79,
      "learning_rate": 0.0001,
      "loss": 1.2372,
      "step": 10000
    },
    {
      "epoch": 2.93,
      "learning_rate": 0.0001,
      "loss": 1.2362,
      "step": 10500
    }
  ],
  "logging_steps": 500,
  "max_steps": 10767,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 6298513440768.0,
  "trial_name": null,
  "trial_params": null
}