trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.9997255224875508,
  "global_step": 9561,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.16,
      "learning_rate": 0.0001,
      "loss": 2.0436,
      "step": 500
    },
    {
      "epoch": 0.31,
      "learning_rate": 0.0001,
      "loss": 1.9352,
      "step": 1000
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.0001,
      "loss": 1.8866,
      "step": 1500
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.0001,
      "loss": 1.8498,
      "step": 2000
    },
    {
      "epoch": 0.78,
      "learning_rate": 0.0001,
      "loss": 1.8294,
      "step": 2500
    },
    {
      "epoch": 0.94,
      "learning_rate": 0.0001,
      "loss": 1.8093,
      "step": 3000
    },
    {
      "epoch": 1.1,
      "learning_rate": 0.0001,
      "loss": 1.7401,
      "step": 3500
    },
    {
      "epoch": 1.26,
      "learning_rate": 0.0001,
      "loss": 1.6925,
      "step": 4000
    },
    {
      "epoch": 1.41,
      "learning_rate": 0.0001,
      "loss": 1.6858,
      "step": 4500
    },
    {
      "epoch": 1.57,
      "learning_rate": 0.0001,
      "loss": 1.6807,
      "step": 5000
    },
    {
      "epoch": 1.73,
      "learning_rate": 0.0001,
      "loss": 1.6756,
      "step": 5500
    },
    {
      "epoch": 1.88,
      "learning_rate": 0.0001,
      "loss": 1.6704,
      "step": 6000
    },
    {
      "epoch": 2.04,
      "learning_rate": 0.0001,
      "loss": 1.6396,
      "step": 6500
    },
    {
      "epoch": 2.2,
      "learning_rate": 0.0001,
      "loss": 1.5626,
      "step": 7000
    },
    {
      "epoch": 2.35,
      "learning_rate": 0.0001,
      "loss": 1.5695,
      "step": 7500
    },
    {
      "epoch": 2.51,
      "learning_rate": 0.0001,
      "loss": 1.5607,
      "step": 8000
    },
    {
      "epoch": 2.67,
      "learning_rate": 0.0001,
      "loss": 1.5643,
      "step": 8500
    },
    {
      "epoch": 2.82,
      "learning_rate": 0.0001,
      "loss": 1.571,
      "step": 9000
    },
    {
      "epoch": 2.98,
      "learning_rate": 0.0001,
      "loss": 1.5614,
      "step": 9500
    },
    {
      "epoch": 3.0,
      "step": 9561,
      "total_flos": 2.1033365013137326e+18,
      "train_loss": 1.7111171247220764,
      "train_runtime": 116164.4127,
      "train_samples_per_second": 5.269,
      "train_steps_per_second": 0.082
    }
  ],
  "max_steps": 9561,
  "num_train_epochs": 3,
  "total_flos": 2.1033365013137326e+18,
  "trial_name": null,
  "trial_params": null
}