{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 7.861635220125786,
  "eval_steps": 500,
  "global_step": 2500,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5829032258064516,
      "eval_loss": 0.1985393762588501,
      "eval_runtime": 5.6203,
      "eval_samples_per_second": 551.572,
      "eval_steps_per_second": 11.565,
      "step": 318
    },
    {
      "epoch": 1.5723270440251573,
      "grad_norm": 0.5288612246513367,
      "learning_rate": 1.606918238993711e-05,
      "loss": 0.3184,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8174193548387096,
      "eval_loss": 0.09726663678884506,
      "eval_runtime": 6.0438,
      "eval_samples_per_second": 512.92,
      "eval_steps_per_second": 10.755,
      "step": 636
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.8764516129032258,
      "eval_loss": 0.06577406078577042,
      "eval_runtime": 5.6348,
      "eval_samples_per_second": 550.155,
      "eval_steps_per_second": 11.535,
      "step": 954
    },
    {
      "epoch": 3.1446540880503147,
      "grad_norm": 0.44873112440109253,
      "learning_rate": 1.2138364779874214e-05,
      "loss": 0.1127,
      "step": 1000
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.8964516129032258,
      "eval_loss": 0.05149334296584129,
      "eval_runtime": 5.7578,
      "eval_samples_per_second": 538.401,
      "eval_steps_per_second": 11.289,
      "step": 1272
    },
    {
      "epoch": 4.716981132075472,
      "grad_norm": 0.30973368883132935,
      "learning_rate": 8.207547169811321e-06,
      "loss": 0.0733,
      "step": 1500
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9051612903225806,
      "eval_loss": 0.04290274158120155,
      "eval_runtime": 5.9031,
      "eval_samples_per_second": 525.15,
      "eval_steps_per_second": 11.011,
      "step": 1590
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.9093548387096774,
      "eval_loss": 0.038180265575647354,
      "eval_runtime": 5.6503,
      "eval_samples_per_second": 548.642,
      "eval_steps_per_second": 11.504,
      "step": 1908
    },
    {
      "epoch": 6.289308176100629,
      "grad_norm": 0.27582159638404846,
      "learning_rate": 4.276729559748428e-06,
      "loss": 0.0587,
      "step": 2000
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.9141935483870968,
      "eval_loss": 0.03587044030427933,
      "eval_runtime": 5.962,
      "eval_samples_per_second": 519.958,
      "eval_steps_per_second": 10.902,
      "step": 2226
    },
    {
      "epoch": 7.861635220125786,
      "grad_norm": 0.275245726108551,
      "learning_rate": 3.459119496855346e-07,
      "loss": 0.0528,
      "step": 2500
    }
  ],
  "logging_steps": 500,
  "max_steps": 2544,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": false
      },
      "attributes": {}
    }
  },
  "total_flos": 651155886807636.0,
  "train_batch_size": 48,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.819089504077056,
    "num_train_epochs": 8,
    "temperature": 12
  }
}