training_config.json

{
  "model_name": "microsoft/Phi-3.5-mini-instruct",
  "lora_r": 16,
  "lora_alpha": 32,
  "lora_target_modules": [
    "gate_proj",
    "up_proj",
    "down_proj"
  ],
  "num_epochs": 2,
  "batch_size": 2,
  "learning_rate": 0.0002,
  "lambda_router": 1.0,
  "lambda_validator": 1.0
}