{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.8849557522123894,
  "eval_steps": 5,
  "global_step": 50,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.017699115044247787,
      "eval_loss": 0.3882482945919037,
      "eval_runtime": 8.7717,
      "eval_samples_per_second": 10.83,
      "eval_steps_per_second": 2.736,
      "step": 1
    },
    {
      "epoch": 0.05309734513274336,
      "grad_norm": 19.522727966308594,
      "learning_rate": 3e-05,
      "loss": 3.5442,
      "step": 3
    },
    {
      "epoch": 0.08849557522123894,
      "eval_loss": 0.2214079350233078,
      "eval_runtime": 8.9304,
      "eval_samples_per_second": 10.638,
      "eval_steps_per_second": 2.687,
      "step": 5
    },
    {
      "epoch": 0.10619469026548672,
      "grad_norm": 12.816184043884277,
      "learning_rate": 6e-05,
      "loss": 2.3401,
      "step": 6
    },
    {
      "epoch": 0.1592920353982301,
      "grad_norm": 11.983319282531738,
      "learning_rate": 9e-05,
      "loss": 1.7236,
      "step": 9
    },
    {
      "epoch": 0.17699115044247787,
      "eval_loss": 0.17182576656341553,
      "eval_runtime": 8.9512,
      "eval_samples_per_second": 10.613,
      "eval_steps_per_second": 2.681,
      "step": 10
    },
    {
      "epoch": 0.21238938053097345,
      "grad_norm": 140.21737670898438,
      "learning_rate": 9.938441702975689e-05,
      "loss": 1.6192,
      "step": 12
    },
    {
      "epoch": 0.26548672566371684,
      "grad_norm": 11.418569564819336,
      "learning_rate": 9.619397662556435e-05,
      "loss": 2.1148,
      "step": 15
    },
    {
      "epoch": 0.26548672566371684,
      "eval_loss": 0.16701287031173706,
      "eval_runtime": 8.9818,
      "eval_samples_per_second": 10.577,
      "eval_steps_per_second": 2.672,
      "step": 15
    },
    {
      "epoch": 0.3185840707964602,
      "grad_norm": 10.450840950012207,
      "learning_rate": 9.045084971874738e-05,
      "loss": 1.6178,
      "step": 18
    },
    {
      "epoch": 0.35398230088495575,
      "eval_loss": 0.15620334446430206,
      "eval_runtime": 8.9863,
      "eval_samples_per_second": 10.572,
      "eval_steps_per_second": 2.671,
      "step": 20
    },
    {
      "epoch": 0.37168141592920356,
      "grad_norm": 9.89293384552002,
      "learning_rate": 8.247240241650918e-05,
      "loss": 1.3005,
      "step": 21
    },
    {
      "epoch": 0.4247787610619469,
      "grad_norm": 7.906158924102783,
      "learning_rate": 7.269952498697734e-05,
      "loss": 1.1924,
      "step": 24
    },
    {
      "epoch": 0.4424778761061947,
      "eval_loss": 0.15097488462924957,
      "eval_runtime": 8.9887,
      "eval_samples_per_second": 10.569,
      "eval_steps_per_second": 2.67,
      "step": 25
    },
    {
      "epoch": 0.4778761061946903,
      "grad_norm": 6.850850582122803,
      "learning_rate": 6.167226819279528e-05,
      "loss": 1.0073,
      "step": 27
    },
    {
      "epoch": 0.5309734513274337,
      "grad_norm": 6.192421913146973,
      "learning_rate": 5e-05,
      "loss": 1.3851,
      "step": 30
    },
    {
      "epoch": 0.5309734513274337,
      "eval_loss": 0.14557085931301117,
      "eval_runtime": 8.9912,
      "eval_samples_per_second": 10.566,
      "eval_steps_per_second": 2.669,
      "step": 30
    },
    {
      "epoch": 0.584070796460177,
      "grad_norm": 6.561943531036377,
      "learning_rate": 3.832773180720475e-05,
      "loss": 1.2746,
      "step": 33
    },
    {
      "epoch": 0.6194690265486725,
      "eval_loss": 0.1403820514678955,
      "eval_runtime": 8.9853,
      "eval_samples_per_second": 10.573,
      "eval_steps_per_second": 2.671,
      "step": 35
    },
    {
      "epoch": 0.6371681415929203,
      "grad_norm": 7.959310531616211,
      "learning_rate": 2.7300475013022663e-05,
      "loss": 1.4708,
      "step": 36
    },
    {
      "epoch": 0.6902654867256637,
      "grad_norm": 6.8480024337768555,
      "learning_rate": 1.7527597583490822e-05,
      "loss": 1.3729,
      "step": 39
    },
    {
      "epoch": 0.7079646017699115,
      "eval_loss": 0.1385464370250702,
      "eval_runtime": 8.9767,
      "eval_samples_per_second": 10.583,
      "eval_steps_per_second": 2.674,
      "step": 40
    },
    {
      "epoch": 0.7433628318584071,
      "grad_norm": 8.225177764892578,
      "learning_rate": 9.549150281252633e-06,
      "loss": 1.3741,
      "step": 42
    },
    {
      "epoch": 0.7964601769911505,
      "grad_norm": 9.350586891174316,
      "learning_rate": 3.8060233744356633e-06,
      "loss": 1.6437,
      "step": 45
    },
    {
      "epoch": 0.7964601769911505,
      "eval_loss": 0.1372378021478653,
      "eval_runtime": 8.9857,
      "eval_samples_per_second": 10.572,
      "eval_steps_per_second": 2.671,
      "step": 45
    },
    {
      "epoch": 0.8495575221238938,
      "grad_norm": 6.208302974700928,
      "learning_rate": 6.15582970243117e-07,
      "loss": 1.2495,
      "step": 48
    },
    {
      "epoch": 0.8849557522123894,
      "eval_loss": 0.13698044419288635,
      "eval_runtime": 8.9913,
      "eval_samples_per_second": 10.566,
      "eval_steps_per_second": 2.669,
      "step": 50
    }
  ],
  "logging_steps": 3,
  "max_steps": 50,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 5,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 7.15499004493824e+16,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": null
}