{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.8,
  "eval_steps": 500,
  "global_step": 7000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.2,
      "grad_norm": 3517696.25,
      "learning_rate": 1.866666666666667e-05,
      "loss": 2.2035,
      "step": 500
    },
    {
      "epoch": 0.4,
      "grad_norm": 1682748.5,
      "learning_rate": 1.7333333333333336e-05,
      "loss": 1.9143,
      "step": 1000
    },
    {
      "epoch": 0.6,
      "grad_norm": 1393452.75,
      "learning_rate": 1.6000000000000003e-05,
      "loss": 1.7796,
      "step": 1500
    },
    {
      "epoch": 0.8,
      "grad_norm": 1795642.875,
      "learning_rate": 1.4666666666666666e-05,
      "loss": 1.6879,
      "step": 2000
    },
    {
      "epoch": 1.0,
      "grad_norm": 1723541.625,
      "learning_rate": 1.3333333333333333e-05,
      "loss": 1.5931,
      "step": 2500
    },
    {
      "epoch": 1.0,
      "eval_gen_len": 18.6233,
      "eval_loss": 1.3491483926773071,
      "eval_rouge1": 0.4823,
      "eval_rouge2": 0.2715,
      "eval_rougeL": 0.3708,
      "eval_rougeLsum": 0.371,
      "eval_runtime": 2274.1943,
      "eval_samples_per_second": 4.397,
      "eval_steps_per_second": 1.099,
      "step": 2500
    },
    {
      "epoch": 1.2,
      "grad_norm": 1764786.5,
      "learning_rate": 1.2e-05,
      "loss": 1.4312,
      "step": 3000
    },
    {
      "epoch": 1.4,
      "grad_norm": 1982422.5,
      "learning_rate": 1.0666666666666667e-05,
      "loss": 1.4074,
      "step": 3500
    },
    {
      "epoch": 1.6,
      "grad_norm": 1838183.875,
      "learning_rate": 9.333333333333334e-06,
      "loss": 1.3547,
      "step": 4000
    },
    {
      "epoch": 1.8,
      "grad_norm": 1721691.875,
      "learning_rate": 8.000000000000001e-06,
      "loss": 1.315,
      "step": 4500
    },
    {
      "epoch": 2.0,
      "grad_norm": 1726840.875,
      "learning_rate": 6.666666666666667e-06,
      "loss": 1.2782,
      "step": 5000
    },
    {
      "epoch": 2.0,
      "eval_gen_len": 18.7402,
      "eval_loss": 1.0953999757766724,
      "eval_rouge1": 0.5034,
      "eval_rouge2": 0.3127,
      "eval_rougeL": 0.4011,
      "eval_rougeLsum": 0.4013,
      "eval_runtime": 2277.7908,
      "eval_samples_per_second": 4.39,
      "eval_steps_per_second": 1.098,
      "step": 5000
    },
    {
      "epoch": 2.2,
      "grad_norm": 2554161.25,
      "learning_rate": 5.333333333333334e-06,
      "loss": 1.2057,
      "step": 5500
    },
    {
      "epoch": 2.4,
      "grad_norm": 1766171.0,
      "learning_rate": 4.000000000000001e-06,
      "loss": 1.1856,
      "step": 6000
    },
    {
      "epoch": 2.6,
      "grad_norm": 1563001.875,
      "learning_rate": 2.666666666666667e-06,
      "loss": 1.1493,
      "step": 6500
    },
    {
      "epoch": 2.8,
      "grad_norm": 1845935.625,
      "learning_rate": 1.3333333333333334e-06,
      "loss": 1.1331,
      "step": 7000
    }
  ],
  "logging_steps": 500,
  "max_steps": 7500,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 9.744998028668928e+16,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}