{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 9750,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.2564102564102564,
      "grad_norm": 0.3748186528682709,
      "learning_rate": 0.0018976410256410258,
      "loss": 0.4066,
      "step": 500
    },
    {
      "epoch": 0.5128205128205128,
      "grad_norm": 0.34374505281448364,
      "learning_rate": 0.001795076923076923,
      "loss": 0.1902,
      "step": 1000
    },
    {
      "epoch": 0.7692307692307693,
      "grad_norm": 0.29863688349723816,
      "learning_rate": 0.0016925128205128206,
      "loss": 0.1362,
      "step": 1500
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.0691862627863884,
      "eval_runtime": 22.7429,
      "eval_samples_per_second": 304.799,
      "eval_steps_per_second": 9.541,
      "step": 1950
    },
    {
      "epoch": 1.0256410256410255,
      "grad_norm": 0.25807419419288635,
      "learning_rate": 0.001589948717948718,
      "loss": 0.1005,
      "step": 2000
    },
    {
      "epoch": 1.282051282051282,
      "grad_norm": 0.24320101737976074,
      "learning_rate": 0.0014873846153846152,
      "loss": 0.0615,
      "step": 2500
    },
    {
      "epoch": 1.5384615384615383,
      "grad_norm": 0.24260133504867554,
      "learning_rate": 0.001384820512820513,
      "loss": 0.0562,
      "step": 3000
    },
    {
      "epoch": 1.7948717948717947,
      "grad_norm": 0.4375219941139221,
      "learning_rate": 0.0012822564102564103,
      "loss": 0.0466,
      "step": 3500
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.03345639631152153,
      "eval_runtime": 23.0664,
      "eval_samples_per_second": 300.523,
      "eval_steps_per_second": 9.408,
      "step": 3900
    },
    {
      "epoch": 2.051282051282051,
      "grad_norm": 0.20671938359737396,
      "learning_rate": 0.0011796923076923076,
      "loss": 0.038,
      "step": 4000
    },
    {
      "epoch": 2.3076923076923075,
      "grad_norm": 0.16543850302696228,
      "learning_rate": 0.0010771282051282051,
      "loss": 0.0222,
      "step": 4500
    },
    {
      "epoch": 2.564102564102564,
      "grad_norm": 0.12343248724937439,
      "learning_rate": 0.0009745641025641025,
      "loss": 0.0223,
      "step": 5000
    },
    {
      "epoch": 2.8205128205128203,
      "grad_norm": 0.11194447427988052,
      "learning_rate": 0.000872,
      "loss": 0.0194,
      "step": 5500
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.020821645855903625,
      "eval_runtime": 22.9804,
      "eval_samples_per_second": 301.649,
      "eval_steps_per_second": 9.443,
      "step": 5850
    },
    {
      "epoch": 3.076923076923077,
      "grad_norm": 0.07755649834871292,
      "learning_rate": 0.0007694358974358975,
      "loss": 0.0145,
      "step": 6000
    },
    {
      "epoch": 3.3333333333333335,
      "grad_norm": 0.0282865259796381,
      "learning_rate": 0.0006668717948717949,
      "loss": 0.0087,
      "step": 6500
    },
    {
      "epoch": 3.58974358974359,
      "grad_norm": 0.18944522738456726,
      "learning_rate": 0.0005643076923076923,
      "loss": 0.0076,
      "step": 7000
    },
    {
      "epoch": 3.8461538461538463,
      "grad_norm": 0.13857877254486084,
      "learning_rate": 0.00046174358974358977,
      "loss": 0.0073,
      "step": 7500
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.016552278771996498,
      "eval_runtime": 22.8045,
      "eval_samples_per_second": 303.975,
      "eval_steps_per_second": 9.516,
      "step": 7800
    },
    {
      "epoch": 4.102564102564102,
      "grad_norm": 0.1588769406080246,
      "learning_rate": 0.0003591794871794872,
      "loss": 0.0049,
      "step": 8000
    },
    {
      "epoch": 4.358974358974359,
      "grad_norm": 0.13791902363300323,
      "learning_rate": 0.00025661538461538465,
      "loss": 0.0028,
      "step": 8500
    },
    {
      "epoch": 4.615384615384615,
      "grad_norm": 0.1163698062300682,
      "learning_rate": 0.00015405128205128206,
      "loss": 0.002,
      "step": 9000
    },
    {
      "epoch": 4.871794871794872,
      "grad_norm": 0.023280449211597443,
      "learning_rate": 5.148717948717948e-05,
      "loss": 0.0017,
      "step": 9500
    },
    {
      "epoch": 5.0,
      "eval_loss": 0.015809964388608932,
      "eval_runtime": 22.9082,
      "eval_samples_per_second": 302.599,
      "eval_steps_per_second": 9.473,
      "step": 9750
    }
  ],
  "logging_steps": 500,
  "max_steps": 9750,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 7.18585886063616e+16,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}