trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.8888888888888888,
  "eval_steps": 500,
  "global_step": 7,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.13,
      "learning_rate": 0.0,
      "loss": 1.2981,
      "step": 1
    },
    {
      "epoch": 0.25,
      "learning_rate": 0.0002,
      "loss": 1.0019,
      "step": 2
    },
    {
      "epoch": 0.38,
      "learning_rate": 0.0002,
      "loss": 1.0169,
      "step": 3
    },
    {
      "epoch": 0.51,
      "learning_rate": 0.0002,
      "loss": 0.9584,
      "step": 4
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.0002,
      "loss": 0.9591,
      "step": 5
    },
    {
      "epoch": 0.76,
      "learning_rate": 0.0002,
      "loss": 0.8598,
      "step": 6
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.0002,
      "loss": 0.8748,
      "step": 7
    },
    {
      "epoch": 0.89,
      "step": 7,
      "total_flos": 10822128107520.0,
      "train_loss": 0.9955644607543945,
      "train_runtime": 509.7363,
      "train_samples_per_second": 1.972,
      "train_steps_per_second": 0.014
    }
  ],
  "logging_steps": 1.0,
  "max_steps": 7,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 50000,
  "total_flos": 10822128107520.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}