{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.2222222222222223,
  "eval_steps": 50,
  "global_step": 21,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.14814814814814814,
      "grad_norm": 8.992554664611816,
      "learning_rate": 1.08e-05,
      "loss": 2.2598,
      "step": 1
    },
    {
      "epoch": 0.14814814814814814,
      "eval_loss": 2.293095111846924,
      "eval_runtime": 0.9073,
      "eval_samples_per_second": 99.191,
      "eval_steps_per_second": 3.306,
      "step": 1
    },
    {
      "epoch": 0.2962962962962963,
      "grad_norm": 9.887986183166504,
      "learning_rate": 2.16e-05,
      "loss": 2.3139,
      "step": 2
    },
    {
      "epoch": 0.4444444444444444,
      "grad_norm": 7.760578632354736,
      "learning_rate": 3.2399999999999995e-05,
      "loss": 2.0925,
      "step": 3
    },
    {
      "epoch": 0.5925925925925926,
      "grad_norm": 12.258286476135254,
      "learning_rate": 4.32e-05,
      "loss": 1.8611,
      "step": 4
    },
    {
      "epoch": 0.7407407407407407,
      "grad_norm": 5.926778793334961,
      "learning_rate": 5.4e-05,
      "loss": 1.5496,
      "step": 5
    },
    {
      "epoch": 0.8888888888888888,
      "grad_norm": 4.352044582366943,
      "learning_rate": 6.479999999999999e-05,
      "loss": 1.2611,
      "step": 6
    },
    {
      "epoch": 1.074074074074074,
      "grad_norm": 7.8084306716918945,
      "learning_rate": 7.56e-05,
      "loss": 1.8202,
      "step": 7
    },
    {
      "epoch": 1.2222222222222223,
      "grad_norm": 10.733794212341309,
      "learning_rate": 8.64e-05,
      "loss": 1.2499,
      "step": 8
    },
    {
      "epoch": 1.3703703703703702,
      "grad_norm": 5.377614498138428,
      "learning_rate": 9.72e-05,
      "loss": 1.1389,
      "step": 9
    },
    {
      "epoch": 1.5185185185185186,
      "grad_norm": 4.3859052658081055,
      "learning_rate": 0.000108,
      "loss": 1.0272,
      "step": 10
    },
    {
      "epoch": 1.6666666666666665,
      "grad_norm": 3.651758909225464,
      "learning_rate": 0.00010581262057518286,
      "loss": 1.0901,
      "step": 11
    },
    {
      "epoch": 1.8148148148148149,
      "grad_norm": 1.944640040397644,
      "learning_rate": 9.942769077288378e-05,
      "loss": 0.9815,
      "step": 12
    },
    {
      "epoch": 1.9629629629629628,
      "grad_norm": 2.401359796524048,
      "learning_rate": 8.93624796330454e-05,
      "loss": 1.3839,
      "step": 13
    },
    {
      "epoch": 2.148148148148148,
      "grad_norm": 1.853423833847046,
      "learning_rate": 7.643241070210187e-05,
      "loss": 0.8266,
      "step": 14
    },
    {
      "epoch": 2.2962962962962963,
      "grad_norm": 1.8709155321121216,
      "learning_rate": 6.16850012667574e-05,
      "loss": 0.7687,
      "step": 15
    },
    {
      "epoch": 2.4444444444444446,
      "grad_norm": 1.4593921899795532,
      "learning_rate": 4.63149987332426e-05,
      "loss": 0.7642,
      "step": 16
    },
    {
      "epoch": 2.5925925925925926,
      "grad_norm": 1.3577464818954468,
      "learning_rate": 3.1567589297898136e-05,
      "loss": 0.7629,
      "step": 17
    },
    {
      "epoch": 2.7407407407407405,
      "grad_norm": 1.30984628200531,
      "learning_rate": 1.863752036695461e-05,
      "loss": 0.7169,
      "step": 18
    },
    {
      "epoch": 2.888888888888889,
      "grad_norm": 1.31648588180542,
      "learning_rate": 8.572309227116208e-06,
      "loss": 0.7275,
      "step": 19
    },
    {
      "epoch": 3.074074074074074,
      "grad_norm": 1.8363643884658813,
      "learning_rate": 2.187379424817148e-06,
      "loss": 1.0192,
      "step": 20
    },
    {
      "epoch": 3.2222222222222223,
      "grad_norm": 1.0485384464263916,
      "learning_rate": 0.0,
      "loss": 0.581,
      "step": 21
    }
  ],
  "logging_steps": 1,
  "max_steps": 21,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 50,
  "stateful_callbacks": {
    "EarlyStoppingCallback": {
      "args": {
        "early_stopping_patience": 5,
        "early_stopping_threshold": 0.0
      },
      "attributes": {
        "early_stopping_patience_counter": 0
      }
    },
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 3.58447037684908e+17,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}