{
  "best_metric": 42.9908,
  "best_model_checkpoint": "opus_big_fine_tfidf_wce_unsampled/checkpoint-32000",
  "epoch": 0.7886629698089956,
  "global_step": 32000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.2,
      "learning_rate": 1.9753635243376465e-05,
      "loss": 0.2036,
      "step": 8000
    },
    {
      "epoch": 0.2,
      "eval_bleu": 41.9332,
      "eval_gen_len": 33.7565,
      "eval_loss": 0.09845346212387085,
      "eval_runtime": 98.5797,
      "eval_samples_per_second": 10.58,
      "eval_steps_per_second": 0.67,
      "step": 8000
    },
    {
      "epoch": 0.39,
      "learning_rate": 1.950723967960567e-05,
      "loss": 0.1717,
      "step": 16000
    },
    {
      "epoch": 0.39,
      "eval_bleu": 42.466,
      "eval_gen_len": 34.3557,
      "eval_loss": 0.09779991954565048,
      "eval_runtime": 101.8795,
      "eval_samples_per_second": 10.238,
      "eval_steps_per_second": 0.648,
      "step": 16000
    },
    {
      "epoch": 0.59,
      "learning_rate": 1.926090573012939e-05,
      "loss": 0.1668,
      "step": 24000
    },
    {
      "epoch": 0.59,
      "eval_bleu": 42.674,
      "eval_gen_len": 33.7651,
      "eval_loss": 0.09813275933265686,
      "eval_runtime": 147.0398,
      "eval_samples_per_second": 7.093,
      "eval_steps_per_second": 0.449,
      "step": 24000
    },
    {
      "epoch": 0.79,
      "learning_rate": 1.9014540973505854e-05,
      "loss": 0.1623,
      "step": 32000
    },
    {
      "epoch": 0.79,
      "eval_bleu": 42.9908,
      "eval_gen_len": 33.8389,
      "eval_loss": 0.09677895158529282,
      "eval_runtime": 81.8819,
      "eval_samples_per_second": 12.738,
      "eval_steps_per_second": 0.806,
      "step": 32000
    }
  ],
  "max_steps": 649200,
  "num_train_epochs": 16,
  "total_flos": 1.0342418777269862e+17,
  "trial_name": null,
  "trial_params": null
}