{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 1000,
  "global_step": 3867,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.0002585983966899405,
      "grad_norm": 6.478072643280029,
      "learning_rate": 0.0,
      "loss": 3.5134,
      "step": 1
    },
    {
      "epoch": 0.2585983966899405,
      "grad_norm": 3.769953966140747,
      "learning_rate": 1.9372979961215257e-05,
      "loss": 2.5294,
      "step": 1000
    },
    {
      "epoch": 0.2585983966899405,
      "eval_cosine_accuracy": 0.9476074576377869,
      "eval_loss": 1.1219995021820068,
      "eval_runtime": 26.9302,
      "eval_samples_per_second": 351.539,
      "eval_steps_per_second": 1.374,
      "step": 1000
    },
    {
      "epoch": 0.517196793379881,
      "grad_norm": 3.698908567428589,
      "learning_rate": 2.90259320451117e-05,
      "loss": 1.84,
      "step": 2000
    },
    {
      "epoch": 0.517196793379881,
      "eval_cosine_accuracy": 0.9596493244171143,
      "eval_loss": 1.035656452178955,
      "eval_runtime": 25.5286,
      "eval_samples_per_second": 370.839,
      "eval_steps_per_second": 1.449,
      "step": 2000
    },
    {
      "epoch": 0.7757951900698216,
      "grad_norm": 3.6514859199523926,
      "learning_rate": 2.6870914445801306e-05,
      "loss": 1.6007,
      "step": 3000
    },
    {
      "epoch": 0.7757951900698216,
      "eval_cosine_accuracy": 0.9655646085739136,
      "eval_loss": 0.969273030757904,
      "eval_runtime": 26.2028,
      "eval_samples_per_second": 361.297,
      "eval_steps_per_second": 1.412,
      "step": 3000
    }
  ],
  "logging_steps": 1000,
  "max_steps": 15468,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "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": 0.0,
  "train_batch_size": 256,
  "trial_name": null,
  "trial_params": null
}