{
  "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": 3.3121,
  "grad_norm": 1.0046179294586182,
  "learning_rate": 0.0,
  "epoch": 4.32,
  "current_steps": 119,
  "train_runtime": 30.59,
  "train_samples_per_second": 16.345,
  "train_steps_per_second": 0.981,
  "total_flos": 1874951556562944.0,
  "train_loss": 3.729141934712728
}