{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.02717391304347826,
  "eval_steps": 500,
  "global_step": 25,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.002173913043478261,
      "grad_norm": 0.02604757621884346,
      "learning_rate": 4.966666666666667e-06,
      "loss": 0.0029,
      "on_policy_loss": 0.0029,
      "step": 2,
      "unsup_pseudo_label_acc": 1.0
    },
    {
      "epoch": 0.004347826086956522,
      "grad_norm": 0.028299972414970398,
      "learning_rate": 4.9000000000000005e-06,
      "loss": 0.0019,
      "on_policy_loss": 0.0019,
      "step": 4,
      "unsup_pseudo_label_acc": 1.0
    },
    {
      "epoch": 0.006521739130434782,
      "grad_norm": 0.04308891296386719,
      "learning_rate": 4.833333333333333e-06,
      "loss": 0.0043,
      "on_policy_loss": 0.0043,
      "step": 6,
      "unsup_pseudo_label_acc": 0.8571428571428571
    },
    {
      "epoch": 0.008695652173913044,
      "grad_norm": 0.024277612566947937,
      "learning_rate": 4.766666666666667e-06,
      "loss": 0.0022,
      "on_policy_loss": 0.0022,
      "step": 8,
      "unsup_pseudo_label_acc": 1.0
    },
    {
      "epoch": 0.010869565217391304,
      "grad_norm": 0.014457030221819878,
      "learning_rate": 4.7e-06,
      "loss": 0.0025,
      "on_policy_loss": 0.0025,
      "step": 10,
      "unsup_pseudo_label_acc": 1.0
    },
    {
      "epoch": 0.013043478260869565,
      "grad_norm": 0.017821121960878372,
      "learning_rate": 4.633333333333334e-06,
      "loss": 0.0028,
      "on_policy_loss": 0.0028,
      "step": 12,
      "unsup_pseudo_label_acc": 1.0
    },
    {
      "epoch": 0.015217391304347827,
      "grad_norm": 0.01567935012280941,
      "learning_rate": 4.566666666666667e-06,
      "loss": 0.0012,
      "on_policy_loss": 0.0012,
      "step": 14,
      "unsup_pseudo_label_acc": 1.0
    },
    {
      "epoch": 0.017391304347826087,
      "grad_norm": 0.01740153506398201,
      "learning_rate": 4.5e-06,
      "loss": -0.0001,
      "on_policy_loss": -0.0001,
      "step": 16,
      "unsup_pseudo_label_acc": 0.8
    },
    {
      "epoch": 0.01956521739130435,
      "grad_norm": 0.018150754272937775,
      "learning_rate": 4.433333333333334e-06,
      "loss": 0.0019,
      "on_policy_loss": 0.0019,
      "step": 18,
      "unsup_pseudo_label_acc": 0.8571428571428571
    },
    {
      "epoch": 0.021739130434782608,
      "grad_norm": 0.019861530512571335,
      "learning_rate": 4.366666666666667e-06,
      "loss": -0.0006,
      "on_policy_loss": -0.0006,
      "step": 20,
      "unsup_pseudo_label_acc": 1.0
    },
    {
      "epoch": 0.02391304347826087,
      "grad_norm": 0.01661345362663269,
      "learning_rate": 4.3e-06,
      "loss": 0.0006,
      "on_policy_loss": 0.0006,
      "step": 22,
      "unsup_pseudo_label_acc": 1.0
    },
    {
      "epoch": 0.02608695652173913,
      "grad_norm": 0.014850025996565819,
      "learning_rate": 4.233333333333334e-06,
      "loss": -0.0026,
      "on_policy_loss": -0.0026,
      "step": 24,
      "unsup_pseudo_label_acc": 1.0
    }
  ],
  "logging_steps": 2,
  "max_steps": 150,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 25,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 0.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}