{
  "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": 0.3689,
  "grad_norm": 0.7466857433319092,
  "learning_rate": 0.0,
  "epoch": 4.948228882833788,
  "current_steps": 1819,
  "train_runtime": 1754.3747,
  "train_samples_per_second": 12.537,
  "train_steps_per_second": 0.259,
  "total_flos": 9.44173788734423e+16,
  "train_loss": 0.5902292704844213
}