{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 810,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.37037037037037035,
      "grad_norm": 7.28477144241333,
      "learning_rate": 4.388888888888889e-05,
      "loss": 0.5289,
      "step": 100
    },
    {
      "epoch": 0.7407407407407407,
      "grad_norm": 1.0851131677627563,
      "learning_rate": 3.771604938271605e-05,
      "loss": 0.0582,
      "step": 200
    },
    {
      "epoch": 1.1111111111111112,
      "grad_norm": 3.6556451320648193,
      "learning_rate": 3.1543209876543214e-05,
      "loss": 0.0499,
      "step": 300
    },
    {
      "epoch": 1.4814814814814814,
      "grad_norm": 1.600995659828186,
      "learning_rate": 2.5370370370370372e-05,
      "loss": 0.0436,
      "step": 400
    },
    {
      "epoch": 1.8518518518518519,
      "grad_norm": 1.711982011795044,
      "learning_rate": 1.9197530864197534e-05,
      "loss": 0.045,
      "step": 500
    },
    {
      "epoch": 2.2222222222222223,
      "grad_norm": 0.9681500792503357,
      "learning_rate": 1.3024691358024694e-05,
      "loss": 0.0407,
      "step": 600
    },
    {
      "epoch": 2.5925925925925926,
      "grad_norm": 1.1934666633605957,
      "learning_rate": 6.851851851851852e-06,
      "loss": 0.0346,
      "step": 700
    },
    {
      "epoch": 2.962962962962963,
      "grad_norm": 1.1116873025894165,
      "learning_rate": 6.790123456790124e-07,
      "loss": 0.0334,
      "step": 800
    }
  ],
  "logging_steps": 100,
  "max_steps": 810,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 200,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 211577748848640.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}