{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.2631578947368421,
  "eval_steps": 4,
  "global_step": 25,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.010526315789473684,
      "grad_norm": 29.17266082763672,
      "learning_rate": 1.0000000000000002e-06,
      "loss": 5.424,
      "step": 1
    },
    {
      "epoch": 0.010526315789473684,
      "eval_loss": NaN,
      "eval_runtime": 5.3901,
      "eval_samples_per_second": 7.421,
      "eval_steps_per_second": 1.855,
      "step": 1
    },
    {
      "epoch": 0.021052631578947368,
      "grad_norm": 35.16109848022461,
      "learning_rate": 2.0000000000000003e-06,
      "loss": 4.8392,
      "step": 2
    },
    {
      "epoch": 0.031578947368421054,
      "grad_norm": 26.968463897705078,
      "learning_rate": 3e-06,
      "loss": 5.4673,
      "step": 3
    },
    {
      "epoch": 0.042105263157894736,
      "grad_norm": 32.266021728515625,
      "learning_rate": 4.000000000000001e-06,
      "loss": 5.3509,
      "step": 4
    },
    {
      "epoch": 0.042105263157894736,
      "eval_loss": NaN,
      "eval_runtime": 5.0573,
      "eval_samples_per_second": 7.909,
      "eval_steps_per_second": 1.977,
      "step": 4
    },
    {
      "epoch": 0.05263157894736842,
      "grad_norm": 39.015323638916016,
      "learning_rate": 5e-06,
      "loss": 4.2175,
      "step": 5
    },
    {
      "epoch": 0.06315789473684211,
      "grad_norm": 27.262535095214844,
      "learning_rate": 6e-06,
      "loss": 5.0985,
      "step": 6
    },
    {
      "epoch": 0.07368421052631578,
      "grad_norm": 31.20633316040039,
      "learning_rate": 7e-06,
      "loss": 5.267,
      "step": 7
    },
    {
      "epoch": 0.08421052631578947,
      "grad_norm": 31.28679084777832,
      "learning_rate": 8.000000000000001e-06,
      "loss": 4.7477,
      "step": 8
    },
    {
      "epoch": 0.08421052631578947,
      "eval_loss": NaN,
      "eval_runtime": 5.0562,
      "eval_samples_per_second": 7.911,
      "eval_steps_per_second": 1.978,
      "step": 8
    },
    {
      "epoch": 0.09473684210526316,
      "grad_norm": 37.779701232910156,
      "learning_rate": 9e-06,
      "loss": 5.0317,
      "step": 9
    },
    {
      "epoch": 0.10526315789473684,
      "grad_norm": 31.49554443359375,
      "learning_rate": 1e-05,
      "loss": 4.7971,
      "step": 10
    },
    {
      "epoch": 0.11578947368421053,
      "grad_norm": 30.3826847076416,
      "learning_rate": 9.890738003669029e-06,
      "loss": 4.3297,
      "step": 11
    },
    {
      "epoch": 0.12631578947368421,
      "grad_norm": 29.516464233398438,
      "learning_rate": 9.567727288213005e-06,
      "loss": 4.4154,
      "step": 12
    },
    {
      "epoch": 0.12631578947368421,
      "eval_loss": NaN,
      "eval_runtime": 5.0722,
      "eval_samples_per_second": 7.886,
      "eval_steps_per_second": 1.972,
      "step": 12
    },
    {
      "epoch": 0.1368421052631579,
      "grad_norm": 28.98780632019043,
      "learning_rate": 9.045084971874738e-06,
      "loss": 3.7288,
      "step": 13
    },
    {
      "epoch": 0.14736842105263157,
      "grad_norm": 40.040767669677734,
      "learning_rate": 8.345653031794292e-06,
      "loss": 3.8781,
      "step": 14
    },
    {
      "epoch": 0.15789473684210525,
      "grad_norm": 25.301918029785156,
      "learning_rate": 7.500000000000001e-06,
      "loss": 3.185,
      "step": 15
    },
    {
      "epoch": 0.16842105263157894,
      "grad_norm": 26.731529235839844,
      "learning_rate": 6.545084971874738e-06,
      "loss": 2.6103,
      "step": 16
    },
    {
      "epoch": 0.16842105263157894,
      "eval_loss": NaN,
      "eval_runtime": 5.1077,
      "eval_samples_per_second": 7.831,
      "eval_steps_per_second": 1.958,
      "step": 16
    },
    {
      "epoch": 0.17894736842105263,
      "grad_norm": 31.212656021118164,
      "learning_rate": 5.522642316338268e-06,
      "loss": 2.5839,
      "step": 17
    },
    {
      "epoch": 0.18947368421052632,
      "grad_norm": 19.863901138305664,
      "learning_rate": 4.477357683661734e-06,
      "loss": 2.1497,
      "step": 18
    },
    {
      "epoch": 0.2,
      "grad_norm": 19.72088050842285,
      "learning_rate": 3.4549150281252635e-06,
      "loss": 1.8183,
      "step": 19
    },
    {
      "epoch": 0.21052631578947367,
      "grad_norm": 20.8995361328125,
      "learning_rate": 2.5000000000000015e-06,
      "loss": 1.6411,
      "step": 20
    },
    {
      "epoch": 0.21052631578947367,
      "eval_loss": NaN,
      "eval_runtime": 5.1024,
      "eval_samples_per_second": 7.839,
      "eval_steps_per_second": 1.96,
      "step": 20
    },
    {
      "epoch": 0.22105263157894736,
      "grad_norm": 19.47292137145996,
      "learning_rate": 1.6543469682057105e-06,
      "loss": 1.6535,
      "step": 21
    },
    {
      "epoch": 0.23157894736842105,
      "grad_norm": 16.744619369506836,
      "learning_rate": 9.549150281252633e-07,
      "loss": 1.3746,
      "step": 22
    },
    {
      "epoch": 0.24210526315789474,
      "grad_norm": 16.703203201293945,
      "learning_rate": 4.322727117869951e-07,
      "loss": 1.4706,
      "step": 23
    },
    {
      "epoch": 0.25263157894736843,
      "grad_norm": 16.96294403076172,
      "learning_rate": 1.0926199633097156e-07,
      "loss": 1.4195,
      "step": 24
    },
    {
      "epoch": 0.25263157894736843,
      "eval_loss": NaN,
      "eval_runtime": 5.0979,
      "eval_samples_per_second": 7.846,
      "eval_steps_per_second": 1.962,
      "step": 24
    },
    {
      "epoch": 0.2631578947368421,
      "grad_norm": 15.989789009094238,
      "learning_rate": 0.0,
      "loss": 1.4447,
      "step": 25
    }
  ],
  "logging_steps": 1,
  "max_steps": 25,
  "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": true
      },
      "attributes": {}
    }
  },
  "total_flos": 9273591044505600.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}