{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 6111,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.25,
      "learning_rate": 1.836360661102929e-05,
      "loss": 3.3023,
      "step": 500
    },
    {
      "epoch": 0.49,
      "learning_rate": 1.6727213222058585e-05,
      "loss": 2.3284,
      "step": 1000
    },
    {
      "epoch": 0.74,
      "learning_rate": 1.5090819833087874e-05,
      "loss": 2.1442,
      "step": 1500
    },
    {
      "epoch": 0.98,
      "learning_rate": 1.3454426444117168e-05,
      "loss": 2.0382,
      "step": 2000
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5937749401436552,
      "eval_loss": 1.8770618438720703,
      "eval_runtime": 302.3493,
      "eval_samples_per_second": 53.875,
      "eval_steps_per_second": 1.687,
      "step": 2037
    },
    {
      "epoch": 1.23,
      "learning_rate": 1.1818033055146459e-05,
      "loss": 1.9449,
      "step": 2500
    },
    {
      "epoch": 1.47,
      "learning_rate": 1.0181639666175749e-05,
      "loss": 1.8855,
      "step": 3000
    },
    {
      "epoch": 1.72,
      "learning_rate": 8.54524627720504e-06,
      "loss": 1.839,
      "step": 3500
    },
    {
      "epoch": 1.96,
      "learning_rate": 6.908852888234332e-06,
      "loss": 1.8027,
      "step": 4000
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6159985266130518,
      "eval_loss": 1.6860090494155884,
      "eval_runtime": 293.1285,
      "eval_samples_per_second": 55.569,
      "eval_steps_per_second": 1.74,
      "step": 4074
    },
    {
      "epoch": 2.21,
      "learning_rate": 5.272459499263623e-06,
      "loss": 1.7814,
      "step": 4500
    },
    {
      "epoch": 2.45,
      "learning_rate": 3.6360661102929145e-06,
      "loss": 1.7355,
      "step": 5000
    },
    {
      "epoch": 2.7,
      "learning_rate": 1.999672721322206e-06,
      "loss": 1.7351,
      "step": 5500
    },
    {
      "epoch": 2.95,
      "learning_rate": 3.6327933235149736e-07,
      "loss": 1.7033,
      "step": 6000
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6246546749340045,
      "eval_loss": 1.6256191730499268,
      "eval_runtime": 298.8328,
      "eval_samples_per_second": 54.509,
      "eval_steps_per_second": 1.707,
      "step": 6111
    },
    {
      "epoch": 3.0,
      "step": 6111,
      "total_flos": 1.9847588302331597e+18,
      "train_loss": 2.015392058142623,
      "train_runtime": 3974.3228,
      "train_samples_per_second": 49.182,
      "train_steps_per_second": 1.538
    }
  ],
  "logging_steps": 500,
  "max_steps": 6111,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 1.9847588302331597e+18,
  "trial_name": null,
  "trial_params": null
}