{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 10.0,
  "eval_steps": 500,
  "global_step": 9400,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.1439921722113503,
      "eval_loss": 5.860990047454834,
      "eval_runtime": 4.0326,
      "eval_samples_per_second": 49.595,
      "eval_steps_per_second": 1.736,
      "step": 940
    },
    {
      "epoch": 1.0638297872340425,
      "grad_norm": 11397.4248046875,
      "learning_rate": 0.0005993999999999999,
      "loss": 1.4989,
      "step": 1000
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.15442270058708416,
      "eval_loss": 5.5648722648620605,
      "eval_runtime": 3.5249,
      "eval_samples_per_second": 56.74,
      "eval_steps_per_second": 1.986,
      "step": 1880
    },
    {
      "epoch": 2.127659574468085,
      "grad_norm": 11682.119140625,
      "learning_rate": 0.0005286428571428571,
      "loss": 1.3199,
      "step": 2000
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.17572407045009786,
      "eval_loss": 5.295774936676025,
      "eval_runtime": 3.5315,
      "eval_samples_per_second": 56.633,
      "eval_steps_per_second": 1.982,
      "step": 2820
    },
    {
      "epoch": 3.1914893617021276,
      "grad_norm": 9340.4306640625,
      "learning_rate": 0.0004572142857142857,
      "loss": 1.2518,
      "step": 3000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.19866927592954992,
      "eval_loss": 5.058368682861328,
      "eval_runtime": 3.5342,
      "eval_samples_per_second": 56.589,
      "eval_steps_per_second": 1.981,
      "step": 3760
    },
    {
      "epoch": 4.25531914893617,
      "grad_norm": 9656.10546875,
      "learning_rate": 0.00038578571428571427,
      "loss": 1.1835,
      "step": 4000
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.21291585127201565,
      "eval_loss": 4.927198886871338,
      "eval_runtime": 3.5327,
      "eval_samples_per_second": 56.613,
      "eval_steps_per_second": 1.981,
      "step": 4700
    },
    {
      "epoch": 5.319148936170213,
      "grad_norm": 10298.4814453125,
      "learning_rate": 0.0003143571428571428,
      "loss": 1.1333,
      "step": 5000
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.22777886497064578,
      "eval_loss": 4.809349536895752,
      "eval_runtime": 3.5307,
      "eval_samples_per_second": 56.645,
      "eval_steps_per_second": 1.983,
      "step": 5640
    },
    {
      "epoch": 6.382978723404255,
      "grad_norm": 10808.5107421875,
      "learning_rate": 0.00024292857142857143,
      "loss": 1.095,
      "step": 6000
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.23796477495107632,
      "eval_loss": 4.723742485046387,
      "eval_runtime": 3.5271,
      "eval_samples_per_second": 56.704,
      "eval_steps_per_second": 1.985,
      "step": 6580
    },
    {
      "epoch": 7.446808510638298,
      "grad_norm": 11282.56640625,
      "learning_rate": 0.0001715,
      "loss": 1.0654,
      "step": 7000
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.2473385518590998,
      "eval_loss": 4.654984474182129,
      "eval_runtime": 3.5239,
      "eval_samples_per_second": 56.755,
      "eval_steps_per_second": 1.986,
      "step": 7520
    },
    {
      "epoch": 8.51063829787234,
      "grad_norm": 12032.25,
      "learning_rate": 0.00010007142857142856,
      "loss": 1.0384,
      "step": 8000
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.25424657534246575,
      "eval_loss": 4.611558437347412,
      "eval_runtime": 3.5329,
      "eval_samples_per_second": 56.611,
      "eval_steps_per_second": 1.981,
      "step": 8460
    },
    {
      "epoch": 9.574468085106384,
      "grad_norm": 11857.1240234375,
      "learning_rate": 2.864285714285714e-05,
      "loss": 1.0178,
      "step": 9000
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.25717221135029356,
      "eval_loss": 4.586706638336182,
      "eval_runtime": 3.5354,
      "eval_samples_per_second": 56.57,
      "eval_steps_per_second": 1.98,
      "step": 9400
    },
    {
      "epoch": 10.0,
      "step": 9400,
      "total_flos": 7.852870729728e+16,
      "train_loss": 1.1709923861889129,
      "train_runtime": 8649.5012,
      "train_samples_per_second": 34.747,
      "train_steps_per_second": 1.087
    }
  ],
  "logging_steps": 1000,
  "max_steps": 9400,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 10,
  "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": 7.852870729728e+16,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}