last-checkpoint/trainer_state.json

{
  "best_metric": 1.0537036657333374,
  "best_model_checkpoint": "./outputs/checkpoint-300",
  "epoch": 0.4,
  "eval_steps": 100,
  "global_step": 300,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.13,
      "learning_rate": 0.0002,
      "loss": 1.0003,
      "step": 100
    },
    {
      "epoch": 0.13,
      "eval_loss": 1.09341299533844,
      "eval_runtime": 882.5981,
      "eval_samples_per_second": 2.187,
      "eval_steps_per_second": 0.274,
      "step": 100
    },
    {
      "epoch": 0.27,
      "learning_rate": 0.0002,
      "loss": 0.8971,
      "step": 200
    },
    {
      "epoch": 0.27,
      "eval_loss": 1.0693918466567993,
      "eval_runtime": 888.706,
      "eval_samples_per_second": 2.172,
      "eval_steps_per_second": 0.272,
      "step": 200
    },
    {
      "epoch": 0.4,
      "learning_rate": 0.0002,
      "loss": 0.8816,
      "step": 300
    },
    {
      "epoch": 0.4,
      "eval_loss": 1.0537036657333374,
      "eval_runtime": 878.8304,
      "eval_samples_per_second": 2.196,
      "eval_steps_per_second": 0.275,
      "step": 300
    }
  ],
  "logging_steps": 100,
  "max_steps": 2250,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 100,
  "total_flos": 7.772891970699264e+16,
  "trial_name": null,
  "trial_params": null
}