checkpoint-12/trainer_state.json

{
  "best_metric": 0.7733290791511536,
  "best_model_checkpoint": "ctllama7b/checkpoint-12",
  "epoch": 5.086956521739131,
  "eval_steps": 500,
  "global_step": 12,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.09,
      "eval_loss": 0.9089534878730774,
      "eval_runtime": 8.5765,
      "eval_samples_per_second": 10.027,
      "eval_steps_per_second": 0.7,
      "step": 2
    },
    {
      "epoch": 1.09,
      "eval_loss": 0.9039565920829773,
      "eval_runtime": 8.6271,
      "eval_samples_per_second": 9.969,
      "eval_steps_per_second": 0.695,
      "step": 4
    },
    {
      "epoch": 2.04,
      "learning_rate": 7.1428571428571436e-06,
      "loss": 0.8946,
      "step": 5
    },
    {
      "epoch": 2.09,
      "eval_loss": 0.8744738101959229,
      "eval_runtime": 8.6363,
      "eval_samples_per_second": 9.958,
      "eval_steps_per_second": 0.695,
      "step": 6
    },
    {
      "epoch": 3.09,
      "eval_loss": 0.8518250584602356,
      "eval_runtime": 8.6389,
      "eval_samples_per_second": 9.955,
      "eval_steps_per_second": 0.695,
      "step": 8
    },
    {
      "epoch": 4.09,
      "learning_rate": 1.4285714285714287e-05,
      "loss": 0.844,
      "step": 10
    },
    {
      "epoch": 4.09,
      "eval_loss": 0.7924608588218689,
      "eval_runtime": 8.6417,
      "eval_samples_per_second": 9.952,
      "eval_steps_per_second": 0.694,
      "step": 10
    },
    {
      "epoch": 5.09,
      "eval_loss": 0.7733290791511536,
      "eval_runtime": 8.6368,
      "eval_samples_per_second": 9.957,
      "eval_steps_per_second": 0.695,
      "step": 12
    }
  ],
  "logging_steps": 5,
  "max_steps": 138,
  "num_train_epochs": 6,
  "save_steps": 500,
  "total_flos": 1.2482646458236928e+16,
  "trial_name": null,
  "trial_params": null
}