{
  "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": 1.1025,
  "grad_norm": 0.39419829845428467,
  "learning_rate": 0.0003,
  "epoch": 0.9174311926605505,
  "current_steps": 99,
  "train_runtime": 43.0179,
  "train_samples_per_second": 8.415,
  "train_steps_per_second": 0.186,
  "total_flos": 1569864557789184.0,
  "train_loss": 0.9896926879882812
}