{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 16.611295681063122,
  "eval_steps": 1000,
  "global_step": 5000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "learning_rate": 9.966777408637875e-07,
      "loss": 2.3466,
      "step": 300
    },
    {
      "epoch": 1.99,
      "learning_rate": 1.993355481727575e-06,
      "loss": 2.3107,
      "step": 600
    },
    {
      "epoch": 2.99,
      "learning_rate": 2.9900332225913626e-06,
      "loss": 2.2553,
      "step": 900
    },
    {
      "epoch": 3.32,
      "eval_loss": 2.210308074951172,
      "eval_runtime": 21.2106,
      "eval_samples_per_second": 80.007,
      "eval_steps_per_second": 5.045,
      "step": 1000
    },
    {
      "epoch": 3.99,
      "learning_rate": 3.98671096345515e-06,
      "loss": 2.2071,
      "step": 1200
    },
    {
      "epoch": 4.98,
      "learning_rate": 4.983388704318937e-06,
      "loss": 2.1719,
      "step": 1500
    },
    {
      "epoch": 5.98,
      "learning_rate": 5.980066445182725e-06,
      "loss": 2.1433,
      "step": 1800
    },
    {
      "epoch": 6.64,
      "eval_loss": 2.1161131858825684,
      "eval_runtime": 21.3287,
      "eval_samples_per_second": 79.564,
      "eval_steps_per_second": 5.017,
      "step": 2000
    },
    {
      "epoch": 6.98,
      "learning_rate": 6.976744186046513e-06,
      "loss": 2.1165,
      "step": 2100
    },
    {
      "epoch": 7.97,
      "learning_rate": 7.9734219269103e-06,
      "loss": 2.0987,
      "step": 2400
    },
    {
      "epoch": 8.97,
      "learning_rate": 8.970099667774087e-06,
      "loss": 2.0829,
      "step": 2700
    },
    {
      "epoch": 9.97,
      "learning_rate": 9.966777408637874e-06,
      "loss": 2.0627,
      "step": 3000
    },
    {
      "epoch": 9.97,
      "eval_loss": 2.0649547576904297,
      "eval_runtime": 21.2155,
      "eval_samples_per_second": 79.989,
      "eval_steps_per_second": 5.043,
      "step": 3000
    },
    {
      "epoch": 10.96,
      "learning_rate": 9.772707766890726e-06,
      "loss": 2.0473,
      "step": 3300
    },
    {
      "epoch": 11.96,
      "learning_rate": 9.081575685366919e-06,
      "loss": 2.0326,
      "step": 3600
    },
    {
      "epoch": 12.96,
      "learning_rate": 7.993539143316044e-06,
      "loss": 2.0225,
      "step": 3900
    },
    {
      "epoch": 13.29,
      "eval_loss": 2.0337860584259033,
      "eval_runtime": 21.3584,
      "eval_samples_per_second": 79.453,
      "eval_steps_per_second": 5.01,
      "step": 4000
    },
    {
      "epoch": 13.95,
      "learning_rate": 6.614402023857231e-06,
      "loss": 2.0126,
      "step": 4200
    },
    {
      "epoch": 14.95,
      "learning_rate": 5.078275689020129e-06,
      "loss": 2.0037,
      "step": 4500
    },
    {
      "epoch": 15.95,
      "learning_rate": 3.534537595264944e-06,
      "loss": 1.995,
      "step": 4800
    },
    {
      "epoch": 16.61,
      "eval_loss": 2.0199732780456543,
      "eval_runtime": 21.332,
      "eval_samples_per_second": 79.552,
      "eval_steps_per_second": 5.016,
      "step": 5000
    }
  ],
  "logging_steps": 300,
  "max_steps": 6020,
  "num_train_epochs": 20,
  "save_steps": 1000,
  "total_flos": 2.583141509776343e+17,
  "trial_name": null,
  "trial_params": null
}