{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.9640591966173364,
  "eval_steps": 500,
  "global_step": 7500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.26,
      "learning_rate": 2.9207188160676534e-05,
      "loss": 0.101,
      "step": 500
    },
    {
      "epoch": 0.53,
      "learning_rate": 2.8414376321353066e-05,
      "loss": 0.0232,
      "step": 1000
    },
    {
      "epoch": 0.79,
      "learning_rate": 2.76215644820296e-05,
      "loss": 0.0138,
      "step": 1500
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.9873417721518988,
      "eval_loss": 0.2694646120071411,
      "eval_runtime": 2.0847,
      "eval_samples_per_second": 227.374,
      "eval_steps_per_second": 9.594,
      "step": 1892
    },
    {
      "epoch": 1.06,
      "learning_rate": 2.6828752642706132e-05,
      "loss": 0.01,
      "step": 2000
    },
    {
      "epoch": 1.32,
      "learning_rate": 2.603594080338266e-05,
      "loss": 0.0056,
      "step": 2500
    },
    {
      "epoch": 1.59,
      "learning_rate": 2.5243128964059198e-05,
      "loss": 0.0065,
      "step": 3000
    },
    {
      "epoch": 1.85,
      "learning_rate": 2.445031712473573e-05,
      "loss": 0.0054,
      "step": 3500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.9957805907172995,
      "eval_loss": 0.30300676822662354,
      "eval_runtime": 2.2025,
      "eval_samples_per_second": 215.207,
      "eval_steps_per_second": 9.08,
      "step": 3784
    },
    {
      "epoch": 2.11,
      "learning_rate": 2.3657505285412263e-05,
      "loss": 0.0054,
      "step": 4000
    },
    {
      "epoch": 2.38,
      "learning_rate": 2.2864693446088796e-05,
      "loss": 0.0073,
      "step": 4500
    },
    {
      "epoch": 2.64,
      "learning_rate": 2.207188160676533e-05,
      "loss": 0.0045,
      "step": 5000
    },
    {
      "epoch": 2.91,
      "learning_rate": 2.1279069767441862e-05,
      "loss": 0.004,
      "step": 5500
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.9936708860759493,
      "eval_loss": 0.3524107336997986,
      "eval_runtime": 2.1316,
      "eval_samples_per_second": 222.363,
      "eval_steps_per_second": 9.382,
      "step": 5676
    },
    {
      "epoch": 3.17,
      "learning_rate": 2.048625792811839e-05,
      "loss": 0.0024,
      "step": 6000
    },
    {
      "epoch": 3.44,
      "learning_rate": 1.9693446088794927e-05,
      "loss": 0.0068,
      "step": 6500
    },
    {
      "epoch": 3.7,
      "learning_rate": 1.890063424947146e-05,
      "loss": 0.0019,
      "step": 7000
    },
    {
      "epoch": 3.96,
      "learning_rate": 1.810782241014799e-05,
      "loss": 0.0018,
      "step": 7500
    }
  ],
  "logging_steps": 500,
  "max_steps": 18920,
  "num_train_epochs": 10,
  "save_steps": 1500,
  "total_flos": 250565766750000.0,
  "trial_name": null,
  "trial_params": null
}