{
  "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": 1.3248,
  "grad_norm": 2.7064802646636963,
  "learning_rate": 0.0,
  "epoch": 17.16,
  "current_steps": 479,
  "train_runtime": 121.69,
  "train_samples_per_second": 16.435,
  "train_steps_per_second": 0.986,
  "total_flos": 7447724238569472.0,
  "train_loss": 2.4713871637980143
}