{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 13.88888888888889,
  "eval_steps": 500,
  "global_step": 500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_bleu": 32.3172,
      "eval_gen_len": 18.7687,
      "eval_loss": 0.4578506052494049,
      "eval_meteor": 0.5435,
      "eval_runtime": 9.7073,
      "eval_samples_per_second": 15.143,
      "eval_steps_per_second": 0.515,
      "step": 36
    },
    {
      "epoch": 2.0,
      "eval_bleu": 39.1064,
      "eval_gen_len": 18.8776,
      "eval_loss": 0.2628231346607208,
      "eval_meteor": 0.6002,
      "eval_runtime": 4.5901,
      "eval_samples_per_second": 32.026,
      "eval_steps_per_second": 1.089,
      "step": 72
    },
    {
      "epoch": 3.0,
      "eval_bleu": 41.6145,
      "eval_gen_len": 18.9524,
      "eval_loss": 0.1903754472732544,
      "eval_meteor": 0.6203,
      "eval_runtime": 4.7069,
      "eval_samples_per_second": 31.231,
      "eval_steps_per_second": 1.062,
      "step": 108
    },
    {
      "epoch": 4.0,
      "eval_bleu": 42.3964,
      "eval_gen_len": 18.9592,
      "eval_loss": 0.15086665749549866,
      "eval_meteor": 0.6265,
      "eval_runtime": 4.8165,
      "eval_samples_per_second": 30.52,
      "eval_steps_per_second": 1.038,
      "step": 144
    },
    {
      "epoch": 5.0,
      "eval_bleu": 44.1755,
      "eval_gen_len": 18.9524,
      "eval_loss": 0.12055599689483643,
      "eval_meteor": 0.6429,
      "eval_runtime": 4.7165,
      "eval_samples_per_second": 31.167,
      "eval_steps_per_second": 1.06,
      "step": 180
    },
    {
      "epoch": 6.0,
      "eval_bleu": 44.9309,
      "eval_gen_len": 18.966,
      "eval_loss": 0.09637363255023956,
      "eval_meteor": 0.6459,
      "eval_runtime": 4.8039,
      "eval_samples_per_second": 30.6,
      "eval_steps_per_second": 1.041,
      "step": 216
    },
    {
      "epoch": 7.0,
      "eval_bleu": 45.7689,
      "eval_gen_len": 18.966,
      "eval_loss": 0.08098377287387848,
      "eval_meteor": 0.653,
      "eval_runtime": 4.7281,
      "eval_samples_per_second": 31.091,
      "eval_steps_per_second": 1.058,
      "step": 252
    },
    {
      "epoch": 8.0,
      "eval_bleu": 45.9332,
      "eval_gen_len": 18.966,
      "eval_loss": 0.06647928804159164,
      "eval_meteor": 0.6519,
      "eval_runtime": 4.8185,
      "eval_samples_per_second": 30.507,
      "eval_steps_per_second": 1.038,
      "step": 288
    },
    {
      "epoch": 9.0,
      "eval_bleu": 46.4058,
      "eval_gen_len": 18.966,
      "eval_loss": 0.05435887724161148,
      "eval_meteor": 0.6531,
      "eval_runtime": 4.7173,
      "eval_samples_per_second": 31.162,
      "eval_steps_per_second": 1.06,
      "step": 324
    },
    {
      "epoch": 10.0,
      "eval_bleu": 46.8506,
      "eval_gen_len": 18.966,
      "eval_loss": 0.04958011582493782,
      "eval_meteor": 0.6616,
      "eval_runtime": 4.7859,
      "eval_samples_per_second": 30.716,
      "eval_steps_per_second": 1.045,
      "step": 360
    },
    {
      "epoch": 11.0,
      "eval_bleu": 46.9661,
      "eval_gen_len": 18.9524,
      "eval_loss": 0.03974379226565361,
      "eval_meteor": 0.6633,
      "eval_runtime": 4.8008,
      "eval_samples_per_second": 30.62,
      "eval_steps_per_second": 1.041,
      "step": 396
    },
    {
      "epoch": 12.0,
      "eval_bleu": 47.1602,
      "eval_gen_len": 18.9524,
      "eval_loss": 0.03513666242361069,
      "eval_meteor": 0.6674,
      "eval_runtime": 4.7634,
      "eval_samples_per_second": 30.861,
      "eval_steps_per_second": 1.05,
      "step": 432
    },
    {
      "epoch": 13.0,
      "eval_bleu": 47.3154,
      "eval_gen_len": 18.966,
      "eval_loss": 0.03375198319554329,
      "eval_meteor": 0.6668,
      "eval_runtime": 4.8164,
      "eval_samples_per_second": 30.52,
      "eval_steps_per_second": 1.038,
      "step": 468
    },
    {
      "epoch": 13.89,
      "learning_rate": 8.888888888888888e-06,
      "loss": 0.327,
      "step": 500
    }
  ],
  "logging_steps": 500,
  "max_steps": 900,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 25,
  "save_steps": 500,
  "total_flos": 2008343123066880.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}