trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.6666666666666665,
  "eval_steps": 500,
  "global_step": 6,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.44,
      "grad_norm": 13.352619143268141,
      "learning_rate": 5e-06,
      "loss": 2.0888,
      "step": 1
    },
    {
      "epoch": 0.89,
      "eval_loss": 1.5768516063690186,
      "eval_runtime": 26.6536,
      "eval_samples_per_second": 12.006,
      "eval_steps_per_second": 0.75,
      "step": 2
    },
    {
      "epoch": 1.78,
      "eval_loss": 1.559399962425232,
      "eval_runtime": 26.454,
      "eval_samples_per_second": 12.096,
      "eval_steps_per_second": 0.756,
      "step": 4
    },
    {
      "epoch": 2.22,
      "grad_norm": 9.789598350001413,
      "learning_rate": 4.774575140626317e-07,
      "loss": 1.8724,
      "step": 5
    },
    {
      "epoch": 2.67,
      "eval_loss": 1.554054856300354,
      "eval_runtime": 26.4867,
      "eval_samples_per_second": 12.082,
      "eval_steps_per_second": 0.755,
      "step": 6
    },
    {
      "epoch": 2.67,
      "step": 6,
      "total_flos": 1203933020160.0,
      "train_loss": 1.8568750619888306,
      "train_runtime": 185.6388,
      "train_samples_per_second": 2.101,
      "train_steps_per_second": 0.032
    }
  ],
  "logging_steps": 5,
  "max_steps": 6,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 1203933020160.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}