{
  "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": 0.3006,
  "grad_norm": 9.823419570922852,
  "learning_rate": 1.4285714285714284e-05,
  "epoch": 1.4444444444444444,
  "current_steps": 15,
  "train_runtime": 16.6599,
  "train_samples_per_second": 11.645,
  "train_steps_per_second": 0.24,
  "total_flos": 628813151600640.0,
  "train_loss": 2.6954402923583984
}