{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.0,
  "eval_steps": 1.0,
  "global_step": 14,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.14285714285714285,
      "grad_norm": 203.78257751464844,
      "learning_rate": 4.642857142857143e-05,
      "loss": 0.8088,
      "step": 1
    },
    {
      "epoch": 0.14285714285714285,
      "eval_loss": 1.0832017660140991,
      "eval_runtime": 4.4764,
      "eval_samples_per_second": 2.234,
      "eval_steps_per_second": 1.117,
      "step": 1
    },
    {
      "epoch": 0.2857142857142857,
      "grad_norm": 551.1321411132812,
      "learning_rate": 4.2857142857142856e-05,
      "loss": 1.5034,
      "step": 2
    },
    {
      "epoch": 0.2857142857142857,
      "eval_loss": 0.8266311883926392,
      "eval_runtime": 2.9969,
      "eval_samples_per_second": 3.337,
      "eval_steps_per_second": 1.668,
      "step": 2
    },
    {
      "epoch": 0.42857142857142855,
      "grad_norm": 342.6697998046875,
      "learning_rate": 3.928571428571429e-05,
      "loss": 1.617,
      "step": 3
    },
    {
      "epoch": 0.42857142857142855,
      "eval_loss": 0.7760497331619263,
      "eval_runtime": 2.9732,
      "eval_samples_per_second": 3.363,
      "eval_steps_per_second": 1.682,
      "step": 3
    },
    {
      "epoch": 0.5714285714285714,
      "grad_norm": 792.9654541015625,
      "learning_rate": 3.571428571428572e-05,
      "loss": 2.2375,
      "step": 4
    },
    {
      "epoch": 0.5714285714285714,
      "eval_loss": 0.7384463548660278,
      "eval_runtime": 3.1473,
      "eval_samples_per_second": 3.177,
      "eval_steps_per_second": 1.589,
      "step": 4
    },
    {
      "epoch": 0.7142857142857143,
      "grad_norm": 210.89300537109375,
      "learning_rate": 3.2142857142857144e-05,
      "loss": 1.5411,
      "step": 5
    },
    {
      "epoch": 0.7142857142857143,
      "eval_loss": 0.82770174741745,
      "eval_runtime": 3.0122,
      "eval_samples_per_second": 3.32,
      "eval_steps_per_second": 1.66,
      "step": 5
    },
    {
      "epoch": 0.8571428571428571,
      "grad_norm": 314.6759338378906,
      "learning_rate": 2.857142857142857e-05,
      "loss": 2.4631,
      "step": 6
    },
    {
      "epoch": 0.8571428571428571,
      "eval_loss": 0.44329363107681274,
      "eval_runtime": 4.1331,
      "eval_samples_per_second": 2.42,
      "eval_steps_per_second": 1.21,
      "step": 6
    },
    {
      "epoch": 1.0,
      "grad_norm": 17.239919662475586,
      "learning_rate": 2.5e-05,
      "loss": 0.0217,
      "step": 7
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.9471458196640015,
      "eval_runtime": 3.013,
      "eval_samples_per_second": 3.319,
      "eval_steps_per_second": 1.659,
      "step": 7
    },
    {
      "epoch": 1.1428571428571428,
      "grad_norm": 213.36294555664062,
      "learning_rate": 2.1428571428571428e-05,
      "loss": 2.4217,
      "step": 8
    },
    {
      "epoch": 1.1428571428571428,
      "eval_loss": 0.9182891845703125,
      "eval_runtime": 2.9638,
      "eval_samples_per_second": 3.374,
      "eval_steps_per_second": 1.687,
      "step": 8
    },
    {
      "epoch": 1.2857142857142856,
      "grad_norm": 200.6674346923828,
      "learning_rate": 1.785714285714286e-05,
      "loss": 2.0588,
      "step": 9
    },
    {
      "epoch": 1.2857142857142856,
      "eval_loss": 0.7376034259796143,
      "eval_runtime": 2.958,
      "eval_samples_per_second": 3.381,
      "eval_steps_per_second": 1.69,
      "step": 9
    },
    {
      "epoch": 1.4285714285714286,
      "grad_norm": 215.76296997070312,
      "learning_rate": 1.4285714285714285e-05,
      "loss": 1.6484,
      "step": 10
    },
    {
      "epoch": 1.4285714285714286,
      "eval_loss": 0.5826703906059265,
      "eval_runtime": 2.9863,
      "eval_samples_per_second": 3.349,
      "eval_steps_per_second": 1.674,
      "step": 10
    },
    {
      "epoch": 1.5714285714285714,
      "grad_norm": 122.98780059814453,
      "learning_rate": 1.0714285714285714e-05,
      "loss": 0.9379,
      "step": 11
    },
    {
      "epoch": 1.5714285714285714,
      "eval_loss": 0.5854327082633972,
      "eval_runtime": 2.9738,
      "eval_samples_per_second": 3.363,
      "eval_steps_per_second": 1.681,
      "step": 11
    },
    {
      "epoch": 1.7142857142857144,
      "grad_norm": 110.82337951660156,
      "learning_rate": 7.142857142857143e-06,
      "loss": 0.8608,
      "step": 12
    },
    {
      "epoch": 1.7142857142857144,
      "eval_loss": 0.5832847952842712,
      "eval_runtime": 6.4862,
      "eval_samples_per_second": 1.542,
      "eval_steps_per_second": 0.771,
      "step": 12
    },
    {
      "epoch": 1.8571428571428572,
      "grad_norm": 91.99958801269531,
      "learning_rate": 3.5714285714285714e-06,
      "loss": 0.958,
      "step": 13
    },
    {
      "epoch": 1.8571428571428572,
      "eval_loss": 0.5970481634140015,
      "eval_runtime": 2.9762,
      "eval_samples_per_second": 3.36,
      "eval_steps_per_second": 1.68,
      "step": 13
    },
    {
      "epoch": 2.0,
      "grad_norm": 28.6821346282959,
      "learning_rate": 0.0,
      "loss": 0.0355,
      "step": 14
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.5955778956413269,
      "eval_runtime": 5.3785,
      "eval_samples_per_second": 1.859,
      "eval_steps_per_second": 0.93,
      "step": 14
    },
    {
      "epoch": 2.0,
      "step": 14,
      "total_flos": 5816699796600.0,
      "train_loss": 1.3652541448495217,
      "train_runtime": 216.8469,
      "train_samples_per_second": 0.461,
      "train_steps_per_second": 0.065
    }
  ],
  "logging_steps": 1.0,
  "max_steps": 14,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": false,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 5816699796600.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}