{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.684563758389262,
  "eval_steps": 500,
  "global_step": 2000,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.67,
      "grad_norm": 10.663328170776367,
      "learning_rate": 4.161073825503356e-05,
      "loss": 0.5114,
      "step": 500
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.813953488372093,
      "eval_f1": 0.7058823529411765,
      "eval_loss": 0.4199533462524414,
      "eval_precision": 0.631578947368421,
      "eval_recall": 0.8,
      "eval_runtime": 1.2208,
      "eval_samples_per_second": 704.445,
      "eval_steps_per_second": 44.233,
      "step": 745
    },
    {
      "epoch": 1.34,
      "grad_norm": 3.2087831497192383,
      "learning_rate": 3.3221476510067115e-05,
      "loss": 0.4622,
      "step": 1000
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8383720930232558,
      "eval_f1": 0.702355460385439,
      "eval_loss": 0.37572425603866577,
      "eval_precision": 0.7224669603524229,
      "eval_recall": 0.6833333333333333,
      "eval_runtime": 1.2295,
      "eval_samples_per_second": 699.468,
      "eval_steps_per_second": 43.92,
      "step": 1490
    },
    {
      "epoch": 2.01,
      "grad_norm": 3.846531867980957,
      "learning_rate": 2.4832214765100674e-05,
      "loss": 0.4133,
      "step": 1500
    },
    {
      "epoch": 2.68,
      "grad_norm": 33.30527877807617,
      "learning_rate": 1.644295302013423e-05,
      "loss": 0.3408,
      "step": 2000
    }
  ],
  "logging_steps": 500,
  "max_steps": 2980,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 500,
  "total_flos": 1171467029553360.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}