{
  "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": 1.0177,
  "grad_norm": 0.35818982124328613,
  "learning_rate": 0.00018095238095238093,
  "epoch": 1.1123595505617978,
  "current_steps": 199,
  "train_runtime": 290.2172,
  "train_samples_per_second": 8.979,
  "train_steps_per_second": 0.186,
  "total_flos": 1.1062774825746432e+16,
  "train_loss": 1.0735347535875108
}