trainer_state.json

{
  "best_metric": 2.5536394119262695,
  "best_model_checkpoint": "shawgpt-ft/checkpoint-13",
  "epoch": 4.0,
  "eval_steps": 500,
  "global_step": 13,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.9230769230769231,
      "grad_norm": 2.1487667560577393,
      "learning_rate": 0.00019285714285714286,
      "loss": 4.5923,
      "step": 3
    },
    {
      "epoch": 0.9230769230769231,
      "eval_loss": 3.957515239715576,
      "eval_runtime": 5.3861,
      "eval_samples_per_second": 1.671,
      "eval_steps_per_second": 0.557,
      "step": 3
    },
    {
      "epoch": 1.8461538461538463,
      "grad_norm": 2.26593279838562,
      "learning_rate": 0.00017142857142857143,
      "loss": 4.0416,
      "step": 6
    },
    {
      "epoch": 1.8461538461538463,
      "eval_loss": 3.429060697555542,
      "eval_runtime": 5.4348,
      "eval_samples_per_second": 1.656,
      "eval_steps_per_second": 0.552,
      "step": 6
    },
    {
      "epoch": 2.769230769230769,
      "grad_norm": 2.352947950363159,
      "learning_rate": 0.00015000000000000001,
      "loss": 3.4652,
      "step": 9
    },
    {
      "epoch": 2.769230769230769,
      "eval_loss": 2.976196527481079,
      "eval_runtime": 5.4009,
      "eval_samples_per_second": 1.666,
      "eval_steps_per_second": 0.555,
      "step": 9
    },
    {
      "epoch": 4.0,
      "grad_norm": 2.3249692916870117,
      "learning_rate": 0.00012142857142857143,
      "loss": 2.2554,
      "step": 13
    },
    {
      "epoch": 4.0,
      "eval_loss": 2.5536394119262695,
      "eval_runtime": 5.4046,
      "eval_samples_per_second": 1.665,
      "eval_steps_per_second": 0.555,
      "step": 13
    }
  ],
  "logging_steps": 500,
  "max_steps": 30,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 43039318818816.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}