{
  "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": 0.1617,
  "grad_norm": 8.796697616577148,
  "learning_rate": 1.4285714285714284e-05,
  "epoch": 15.444444444444445,
  "current_steps": 183,
  "train_runtime": 128.9091,
  "train_samples_per_second": 17.842,
  "train_steps_per_second": 0.357,
  "total_flos": 6713122473639936.0,
  "train_loss": 0.7870955677784007
}