{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 6411,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.2,
      "grad_norm": 6479046246400.0,
      "learning_rate": 3.3255451713395644e-05,
      "loss": 31699221108.3091,
      "step": 427
    },
    {
      "epoch": 0.4,
      "grad_norm": 8700652158976.0,
      "learning_rate": 4.8162593170393484e-05,
      "loss": 14378762930.6604,
      "step": 854
    },
    {
      "epoch": 0.6,
      "grad_norm": 16.28143310546875,
      "learning_rate": 4.446177847113885e-05,
      "loss": 3635725784.4309,
      "step": 1281
    },
    {
      "epoch": 0.8,
      "grad_norm": 6.706409454345703,
      "learning_rate": 4.076096377188421e-05,
      "loss": 1017393142.4075,
      "step": 1708
    },
    {
      "epoch": 1.0,
      "grad_norm": 9.295268058776855,
      "learning_rate": 3.7060149072629574e-05,
      "loss": 1230655574.3326,
      "step": 2135
    },
    {
      "epoch": 1.2,
      "grad_norm": 7.9164958000183105,
      "learning_rate": 3.335933437337493e-05,
      "loss": 387735187.4848,
      "step": 2562
    },
    {
      "epoch": 1.4,
      "grad_norm": 18.9678897857666,
      "learning_rate": 2.96585196741203e-05,
      "loss": 610998305.5738,
      "step": 2989
    },
    {
      "epoch": 1.6,
      "grad_norm": 22.521636962890625,
      "learning_rate": 2.595770497486566e-05,
      "loss": 286763645.9016,
      "step": 3416
    },
    {
      "epoch": 1.8,
      "grad_norm": 6.2225751876831055,
      "learning_rate": 2.2256890275611026e-05,
      "loss": 146726173.377,
      "step": 3843
    },
    {
      "epoch": 2.0,
      "grad_norm": 4.1397271156311035,
      "learning_rate": 1.8556075576356388e-05,
      "loss": 153057112.1311,
      "step": 4270
    },
    {
      "epoch": 2.2,
      "grad_norm": 20.3540096282959,
      "learning_rate": 1.4855260877101752e-05,
      "loss": 167597749.0585,
      "step": 4697
    },
    {
      "epoch": 2.4,
      "grad_norm": 2.9089903831481934,
      "learning_rate": 1.1154446177847114e-05,
      "loss": 31491796.2342,
      "step": 5124
    },
    {
      "epoch": 2.6,
      "grad_norm": 11.952803611755371,
      "learning_rate": 7.453631478592478e-06,
      "loss": 228096436.459,
      "step": 5551
    },
    {
      "epoch": 2.8,
      "grad_norm": 8.380680084228516,
      "learning_rate": 3.7528167793378402e-06,
      "loss": 170669039.2131,
      "step": 5978
    },
    {
      "epoch": 3.0,
      "grad_norm": 11.055083274841309,
      "learning_rate": 5.2002080083203335e-08,
      "loss": 77525903.2881,
      "step": 6405
    }
  ],
  "logging_steps": 427,
  "max_steps": 6411,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 1.089435891661996e+18,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}