{
  "base_model_name": "meta-llama_Llama-3.2-3B",
  "base_model_class": "LlamaForCausalLM",
  "base_loaded_in_4bit": false,
  "base_loaded_in_8bit": false,
  "projections": "q, v",
  "loss": 0.5418,
  "grad_norm": 0.8030437231063843,
  "learning_rate": 0.00027325905292479105,
  "epoch": 2.0217983651226157,
  "current_steps": 746,
  "train_runtime": 713.4792,
  "train_samples_per_second": 123.311,
  "train_steps_per_second": 2.551,
  "total_flos": 3.867417714465178e+16,
  "train_loss": 0.8046114883756125
}