trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.999958205046044,
  "global_step": 13458,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.11,
      "learning_rate": 0.0001,
      "loss": 2.0593,
      "step": 500
    },
    {
      "epoch": 0.22,
      "learning_rate": 0.0001,
      "loss": 2.021,
      "step": 1000
    },
    {
      "epoch": 0.33,
      "learning_rate": 0.0001,
      "loss": 2.0008,
      "step": 1500
    },
    {
      "epoch": 0.45,
      "learning_rate": 0.0001,
      "loss": 1.9989,
      "step": 2000
    },
    {
      "epoch": 0.56,
      "learning_rate": 0.0001,
      "loss": 1.9951,
      "step": 2500
    },
    {
      "epoch": 0.67,
      "learning_rate": 0.0001,
      "loss": 1.9894,
      "step": 3000
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.0001,
      "loss": 1.9806,
      "step": 3500
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.0001,
      "loss": 1.9778,
      "step": 4000
    },
    {
      "epoch": 1.0,
      "learning_rate": 0.0001,
      "loss": 1.9716,
      "step": 4500
    },
    {
      "epoch": 1.11,
      "learning_rate": 0.0001,
      "loss": 1.9545,
      "step": 5000
    },
    {
      "epoch": 1.23,
      "learning_rate": 0.0001,
      "loss": 1.9532,
      "step": 5500
    },
    {
      "epoch": 1.34,
      "learning_rate": 0.0001,
      "loss": 1.9484,
      "step": 6000
    },
    {
      "epoch": 1.45,
      "learning_rate": 0.0001,
      "loss": 1.9464,
      "step": 6500
    },
    {
      "epoch": 1.56,
      "learning_rate": 0.0001,
      "loss": 1.9441,
      "step": 7000
    },
    {
      "epoch": 1.67,
      "learning_rate": 0.0001,
      "loss": 1.9399,
      "step": 7500
    },
    {
      "epoch": 1.78,
      "learning_rate": 0.0001,
      "loss": 1.9398,
      "step": 8000
    },
    {
      "epoch": 1.89,
      "learning_rate": 0.0001,
      "loss": 1.9377,
      "step": 8500
    },
    {
      "epoch": 2.01,
      "learning_rate": 0.0001,
      "loss": 1.933,
      "step": 9000
    },
    {
      "epoch": 2.12,
      "learning_rate": 0.0001,
      "loss": 1.9219,
      "step": 9500
    },
    {
      "epoch": 2.23,
      "learning_rate": 0.0001,
      "loss": 1.9213,
      "step": 10000
    },
    {
      "epoch": 2.34,
      "learning_rate": 0.0001,
      "loss": 1.9168,
      "step": 10500
    },
    {
      "epoch": 2.45,
      "learning_rate": 0.0001,
      "loss": 1.9097,
      "step": 11000
    },
    {
      "epoch": 2.56,
      "learning_rate": 0.0001,
      "loss": 1.9154,
      "step": 11500
    },
    {
      "epoch": 2.67,
      "learning_rate": 0.0001,
      "loss": 1.9129,
      "step": 12000
    },
    {
      "epoch": 2.79,
      "learning_rate": 0.0001,
      "loss": 1.9179,
      "step": 12500
    },
    {
      "epoch": 2.9,
      "learning_rate": 0.0001,
      "loss": 1.9131,
      "step": 13000
    },
    {
      "epoch": 3.0,
      "step": 13458,
      "total_flos": 2.775475250397266e+17,
      "train_loss": 1.9529757999279007,
      "train_runtime": 29729.9646,
      "train_samples_per_second": 28.972,
      "train_steps_per_second": 0.453
    }
  ],
  "max_steps": 13458,
  "num_train_epochs": 3,
  "total_flos": 2.775475250397266e+17,
  "trial_name": null,
  "trial_params": null
}