{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.2941626306398164,
  "eval_steps": 500,
  "global_step": 9000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.1274534794799898,
      "grad_norm": 855.9707641601562,
      "learning_rate": 9.999156208166614e-06,
      "loss": 74.3469,
      "step": 500
    },
    {
      "epoch": 0.2549069589599796,
      "grad_norm": 302.50543212890625,
      "learning_rate": 9.89824533543787e-06,
      "loss": 61.1889,
      "step": 1000
    },
    {
      "epoch": 0.3823604384399694,
      "grad_norm": 495.7685546875,
      "learning_rate": 9.632470336074009e-06,
      "loss": 57.9995,
      "step": 1500
    },
    {
      "epoch": 0.5098139179199592,
      "grad_norm": 265.8482666015625,
      "learning_rate": 9.21077660993783e-06,
      "loss": 57.4618,
      "step": 2000
    },
    {
      "epoch": 0.637267397399949,
      "grad_norm": 1070.9898681640625,
      "learning_rate": 8.647357437093104e-06,
      "loss": 53.0734,
      "step": 2500
    },
    {
      "epoch": 0.7647208768799388,
      "grad_norm": 607.1848754882812,
      "learning_rate": 7.961176263324902e-06,
      "loss": 54.122,
      "step": 3000
    },
    {
      "epoch": 0.8921743563599286,
      "grad_norm": 299.444091796875,
      "learning_rate": 7.175328432410367e-06,
      "loss": 55.9574,
      "step": 3500
    },
    {
      "epoch": 1.0,
      "eval_loss": 2435.821044921875,
      "eval_runtime": 13.0829,
      "eval_samples_per_second": 266.532,
      "eval_steps_per_second": 33.326,
      "step": 3923
    },
    {
      "epoch": 1.0196278358399185,
      "grad_norm": 313.2464294433594,
      "learning_rate": 6.3162638478229965e-06,
      "loss": 52.9767,
      "step": 4000
    },
    {
      "epoch": 1.1470813153199082,
      "grad_norm": 406.19097900390625,
      "learning_rate": 5.412896727361663e-06,
      "loss": 49.7189,
      "step": 4500
    },
    {
      "epoch": 1.274534794799898,
      "grad_norm": 227.44325256347656,
      "learning_rate": 4.495632414398659e-06,
      "loss": 48.3298,
      "step": 5000
    },
    {
      "epoch": 1.401988274279888,
      "grad_norm": 292.7037353515625,
      "learning_rate": 3.595344001132154e-06,
      "loss": 50.4245,
      "step": 5500
    },
    {
      "epoch": 1.5294417537598777,
      "grad_norm": 522.2669677734375,
      "learning_rate": 2.7423332084455543e-06,
      "loss": 49.8634,
      "step": 6000
    },
    {
      "epoch": 1.6568952332398674,
      "grad_norm": 408.0614929199219,
      "learning_rate": 1.965310496864217e-06,
      "loss": 49.0125,
      "step": 6500
    },
    {
      "epoch": 1.7843487127198574,
      "grad_norm": 722.4154052734375,
      "learning_rate": 1.290428735823593e-06,
      "loss": 44.8823,
      "step": 7000
    },
    {
      "epoch": 1.911802192199847,
      "grad_norm": 383.4438171386719,
      "learning_rate": 7.404029558083653e-07,
      "loss": 47.8235,
      "step": 7500
    },
    {
      "epoch": 2.0,
      "eval_loss": 2325.86083984375,
      "eval_runtime": 13.1669,
      "eval_samples_per_second": 264.831,
      "eval_steps_per_second": 33.113,
      "step": 7846
    },
    {
      "epoch": 2.039255671679837,
      "grad_norm": 408.5860290527344,
      "learning_rate": 3.337458105627145e-07,
      "loss": 44.1611,
      "step": 8000
    },
    {
      "epoch": 2.166709151159827,
      "grad_norm": 252.09278869628906,
      "learning_rate": 8.414448202622494e-08,
      "loss": 45.9323,
      "step": 8500
    },
    {
      "epoch": 2.2941626306398164,
      "grad_norm": 174.15538024902344,
      "learning_rate": 0.0,
      "loss": 47.7162,
      "step": 9000
    }
  ],
  "logging_steps": 500,
  "max_steps": 9000,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 3000,
  "total_flos": 0.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}