{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.9984168865435357,
  "global_step": 946,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.21,
      "learning_rate": 3.58139534883721e-05,
      "loss": 2.303,
      "step": 100
    },
    {
      "epoch": 0.42,
      "learning_rate": 3.158562367864693e-05,
      "loss": 1.8006,
      "step": 200
    },
    {
      "epoch": 0.63,
      "learning_rate": 2.739957716701903e-05,
      "loss": 1.7525,
      "step": 300
    },
    {
      "epoch": 0.84,
      "learning_rate": 2.317124735729387e-05,
      "loss": 1.6911,
      "step": 400
    },
    {
      "epoch": 1.06,
      "learning_rate": 1.8942917547568712e-05,
      "loss": 1.6314,
      "step": 500
    },
    {
      "epoch": 1.27,
      "learning_rate": 1.4714587737843554e-05,
      "loss": 1.5195,
      "step": 600
    },
    {
      "epoch": 1.48,
      "learning_rate": 1.0486257928118395e-05,
      "loss": 1.5048,
      "step": 700
    },
    {
      "epoch": 1.69,
      "learning_rate": 6.257928118393235e-06,
      "loss": 1.4919,
      "step": 800
    },
    {
      "epoch": 1.9,
      "learning_rate": 2.0295983086680763e-06,
      "loss": 1.5073,
      "step": 900
    },
    {
      "epoch": 2.0,
      "step": 946,
      "total_flos": 1.834379077681152e+16,
      "train_loss": 1.6799574886265567,
      "train_runtime": 1313.2029,
      "train_samples_per_second": 46.174,
      "train_steps_per_second": 0.72
    },
    {
      "epoch": 2.0,
      "eval_bleu": 12.619547537319557,
      "eval_loss": 1.4679548740386963,
      "eval_runtime": 33.7947,
      "eval_samples_per_second": 29.59,
      "eval_steps_per_second": 1.864,
      "step": 946
    }
  ],
  "max_steps": 946,
  "num_train_epochs": 2,
  "total_flos": 1.834379077681152e+16,
  "trial_name": null,
  "trial_params": null
}