{
  "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": 2.3772,
  "grad_norm": 0.5868062973022461,
  "learning_rate": 6.521739130434782e-06,
  "epoch": 0.997867803837953,
  "current_steps": 467,
  "train_runtime": 130.8265,
  "train_samples_per_second": 14.324,
  "train_steps_per_second": 0.894,
  "total_flos": 8118194619285504.0,
  "train_loss": 2.482547796689547
}