{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 10.0,
  "eval_steps": 500,
  "global_step": 5530,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.9,
      "grad_norm": 0.1664367914199829,
      "learning_rate": 0.0009095840867992767,
      "loss": 0.4003,
      "step": 500
    },
    {
      "epoch": 0.9,
      "eval_loss": 0.162927508354187,
      "eval_runtime": 3.0318,
      "eval_samples_per_second": 2591.854,
      "eval_steps_per_second": 20.45,
      "step": 500
    },
    {
      "epoch": 1.81,
      "grad_norm": 0.1386815905570984,
      "learning_rate": 0.0008191681735985533,
      "loss": 0.2314,
      "step": 1000
    },
    {
      "epoch": 1.81,
      "eval_loss": 0.1386137157678604,
      "eval_runtime": 3.0434,
      "eval_samples_per_second": 2582.007,
      "eval_steps_per_second": 20.372,
      "step": 1000
    },
    {
      "epoch": 2.71,
      "grad_norm": 0.1781063824892044,
      "learning_rate": 0.0007287522603978301,
      "loss": 0.2065,
      "step": 1500
    },
    {
      "epoch": 2.71,
      "eval_loss": 0.1289130598306656,
      "eval_runtime": 3.0677,
      "eval_samples_per_second": 2561.542,
      "eval_steps_per_second": 20.211,
      "step": 1500
    },
    {
      "epoch": 3.62,
      "grad_norm": 0.15570929646492004,
      "learning_rate": 0.0006383363471971068,
      "loss": 0.187,
      "step": 2000
    },
    {
      "epoch": 3.62,
      "eval_loss": 0.12326223403215408,
      "eval_runtime": 3.0605,
      "eval_samples_per_second": 2567.579,
      "eval_steps_per_second": 20.258,
      "step": 2000
    },
    {
      "epoch": 4.52,
      "grad_norm": 0.16776247322559357,
      "learning_rate": 0.0005479204339963833,
      "loss": 0.1791,
      "step": 2500
    },
    {
      "epoch": 4.52,
      "eval_loss": 0.1168670803308487,
      "eval_runtime": 3.0473,
      "eval_samples_per_second": 2578.705,
      "eval_steps_per_second": 20.346,
      "step": 2500
    },
    {
      "epoch": 5.42,
      "grad_norm": 0.1355486512184143,
      "learning_rate": 0.0004575045207956601,
      "loss": 0.1713,
      "step": 3000
    },
    {
      "epoch": 5.42,
      "eval_loss": 0.11528698354959488,
      "eval_runtime": 3.0013,
      "eval_samples_per_second": 2618.163,
      "eval_steps_per_second": 20.657,
      "step": 3000
    },
    {
      "epoch": 6.33,
      "grad_norm": 0.16372531652450562,
      "learning_rate": 0.0003670886075949367,
      "loss": 0.1661,
      "step": 3500
    },
    {
      "epoch": 6.33,
      "eval_loss": 0.11218289285898209,
      "eval_runtime": 2.9586,
      "eval_samples_per_second": 2655.959,
      "eval_steps_per_second": 20.956,
      "step": 3500
    },
    {
      "epoch": 7.23,
      "grad_norm": 0.1596778929233551,
      "learning_rate": 0.0002766726943942134,
      "loss": 0.1604,
      "step": 4000
    },
    {
      "epoch": 7.23,
      "eval_loss": 0.1085081547498703,
      "eval_runtime": 2.9539,
      "eval_samples_per_second": 2660.243,
      "eval_steps_per_second": 20.989,
      "step": 4000
    },
    {
      "epoch": 8.14,
      "grad_norm": 0.15582768619060516,
      "learning_rate": 0.00018625678119349006,
      "loss": 0.1574,
      "step": 4500
    },
    {
      "epoch": 8.14,
      "eval_loss": 0.1098729744553566,
      "eval_runtime": 2.9739,
      "eval_samples_per_second": 2642.311,
      "eval_steps_per_second": 20.848,
      "step": 4500
    },
    {
      "epoch": 9.04,
      "grad_norm": 0.15063905715942383,
      "learning_rate": 9.584086799276672e-05,
      "loss": 0.1541,
      "step": 5000
    },
    {
      "epoch": 9.04,
      "eval_loss": 0.10638037323951721,
      "eval_runtime": 3.0665,
      "eval_samples_per_second": 2562.534,
      "eval_steps_per_second": 20.219,
      "step": 5000
    },
    {
      "epoch": 9.95,
      "grad_norm": 0.14130930602550507,
      "learning_rate": 5.4249547920433995e-06,
      "loss": 0.1521,
      "step": 5500
    },
    {
      "epoch": 9.95,
      "eval_loss": 0.1071261540055275,
      "eval_runtime": 3.024,
      "eval_samples_per_second": 2598.51,
      "eval_steps_per_second": 20.502,
      "step": 5500
    },
    {
      "epoch": 10.0,
      "step": 5530,
      "total_flos": 2.2872619342626816e+16,
      "train_loss": 0.19658087959772425,
      "train_runtime": 755.9435,
      "train_samples_per_second": 935.506,
      "train_steps_per_second": 7.315
    }
  ],
  "logging_steps": 500,
  "max_steps": 5530,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "save_steps": 500,
  "total_flos": 2.2872619342626816e+16,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}