last-checkpoint/trainer_state.json

{
  "best_metric": 2.240180015563965,
  "best_model_checkpoint": "dq158/pingusPongus/checkpoint-1581",
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 1581,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.32,
      "learning_rate": 0.0005,
      "loss": 2.8358,
      "step": 500
    },
    {
      "epoch": 0.63,
      "learning_rate": 0.0004943963563220183,
      "loss": 2.4569,
      "step": 1000
    },
    {
      "epoch": 0.95,
      "learning_rate": 0.0004778366318678313,
      "loss": 2.3821,
      "step": 1500
    },
    {
      "epoch": 1.0,
      "eval_bleu": 1.0,
      "eval_brevity_penalty": 1.0,
      "eval_length_ratio": 1.0,
      "eval_loss": 2.240180015563965,
      "eval_precisions": [
        1.0,
        1.0,
        1.0,
        1.0
      ],
      "eval_reference_length": 1439232,
      "eval_runtime": 880.2333,
      "eval_samples_per_second": 3.193,
      "eval_steps_per_second": 0.2,
      "eval_translation_length": 1439232,
      "step": 1581
    }
  ],
  "logging_steps": 500,
  "max_steps": 7905,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 2.1666322696686797e+17,
  "trial_name": null,
  "trial_params": null
}