{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.025220680958385876,
  "eval_steps": 100,
  "global_step": 10,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0025220680958385876,
      "grad_norm": 119.76318359375,
      "kl_loss": -1.1687562835330993e-15,
      "kl_weight": 1000000.0,
      "learning_rate": 0.0,
      "loss": 2.6394360065460205,
      "step": 1,
      "total_loss": 2.6394360065460205
    },
    {
      "epoch": 0.005044136191677175,
      "grad_norm": 116.01831817626953,
      "kl_loss": -1.280914393650412e-14,
      "kl_weight": 1000000.0,
      "learning_rate": 0.0001,
      "loss": 3.2936160564422607,
      "step": 2,
      "total_loss": 3.2936160564422607
    },
    {
      "epoch": 0.007566204287515763,
      "grad_norm": 104.04817962646484,
      "kl_loss": 7.10318071028837e-09,
      "kl_weight": 1000000.0,
      "learning_rate": 0.0002,
      "loss": 3.084439992904663,
      "step": 3,
      "total_loss": 3.091543197631836
    },
    {
      "epoch": 0.01008827238335435,
      "grad_norm": 68.36679077148438,
      "kl_loss": 2.8489626657801637e-08,
      "kl_weight": 1000000.0,
      "learning_rate": 0.0003,
      "loss": 3.105210304260254,
      "step": 4,
      "total_loss": 3.133699893951416
    },
    {
      "epoch": 0.012610340479192938,
      "grad_norm": 61.00284957885742,
      "kl_loss": 4.923957774849441e-08,
      "kl_weight": 1000000.0,
      "learning_rate": 0.0004,
      "loss": 3.345022678375244,
      "step": 5,
      "total_loss": 3.3942623138427734
    },
    {
      "epoch": 0.015132408575031526,
      "grad_norm": 65.48960876464844,
      "kl_loss": 1.43211394743048e-07,
      "kl_weight": 1000000.0,
      "learning_rate": 0.0005,
      "loss": 2.3467514514923096,
      "step": 6,
      "total_loss": 2.4899628162384033
    },
    {
      "epoch": 0.017654476670870115,
      "grad_norm": 63.001102447509766,
      "kl_loss": 9.109995602329946e-08,
      "kl_weight": 1000000.0,
      "learning_rate": 0.0004993662864385298,
      "loss": 2.5077083110809326,
      "step": 7,
      "total_loss": 2.5988082885742188
    },
    {
      "epoch": 0.0201765447667087,
      "grad_norm": 58.6073112487793,
      "kl_loss": 2.3511624647198914e-07,
      "kl_weight": 1000000.0,
      "learning_rate": 0.0004987325728770596,
      "loss": 2.2668278217315674,
      "step": 8,
      "total_loss": 2.501944065093994
    },
    {
      "epoch": 0.02269861286254729,
      "grad_norm": 97.743896484375,
      "kl_loss": 2.1175161180053692e-07,
      "kl_weight": 1000000.0,
      "learning_rate": 0.0004980988593155894,
      "loss": 2.352029800415039,
      "step": 9,
      "total_loss": 2.563781499862671
    },
    {
      "epoch": 0.025220680958385876,
      "grad_norm": 60.91500473022461,
      "kl_loss": 1.2846226127294358e-07,
      "kl_weight": 1000000.0,
      "learning_rate": 0.0004974651457541192,
      "loss": 2.2376697063446045,
      "step": 10,
      "total_loss": 2.3661320209503174
    }
  ],
  "logging_steps": 1,
  "max_steps": 794,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 5,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": false,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 3438280704000000.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}