last-checkpoint/trainer_state.json

{
  "best_metric": 2.096945285797119,
  "best_model_checkpoint": "dq158/pingusPongus/checkpoint-790",
  "epoch": 2.998102466793169,
  "eval_steps": 500,
  "global_step": 2370,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.63,
      "learning_rate": 4.6266135493489015e-05,
      "loss": 2.2387,
      "step": 500
    },
    {
      "epoch": 1.0,
      "eval_bleu": 1.0,
      "eval_brevity_penalty": 1.0,
      "eval_length_ratio": 1.0,
      "eval_loss": 2.096945285797119,
      "eval_precisions": [
        1.0,
        1.0,
        1.0,
        1.0
      ],
      "eval_reference_length": 1439232,
      "eval_runtime": 879.4262,
      "eval_samples_per_second": 3.196,
      "eval_steps_per_second": 0.2,
      "eval_translation_length": 1439232,
      "step": 790
    },
    {
      "epoch": 1.27,
      "learning_rate": 3.2988191110253866e-05,
      "loss": 2.241,
      "step": 1000
    },
    {
      "epoch": 1.9,
      "learning_rate": 1.6035418002300977e-05,
      "loss": 2.2435,
      "step": 1500
    },
    {
      "epoch": 2.0,
      "eval_bleu": 1.0,
      "eval_brevity_penalty": 1.0,
      "eval_length_ratio": 1.0,
      "eval_loss": 2.096945285797119,
      "eval_precisions": [
        1.0,
        1.0,
        1.0,
        1.0
      ],
      "eval_reference_length": 1439232,
      "eval_runtime": 877.8823,
      "eval_samples_per_second": 3.202,
      "eval_steps_per_second": 0.2,
      "eval_translation_length": 1439232,
      "step": 1581
    },
    {
      "epoch": 2.53,
      "learning_rate": 3.206645306399442e-06,
      "loss": 2.2391,
      "step": 2000
    },
    {
      "epoch": 3.0,
      "eval_bleu": 1.0,
      "eval_brevity_penalty": 1.0,
      "eval_length_ratio": 1.0,
      "eval_loss": 2.096945285797119,
      "eval_precisions": [
        1.0,
        1.0,
        1.0,
        1.0
      ],
      "eval_reference_length": 1439232,
      "eval_runtime": 865.4618,
      "eval_samples_per_second": 3.248,
      "eval_steps_per_second": 0.203,
      "eval_translation_length": 1439232,
      "step": 2370
    }
  ],
  "logging_steps": 500,
  "max_steps": 2370,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 6.496213080106598e+17,
  "trial_name": null,
  "trial_params": null
}