llama-3.2-1b-code-instruct/generation_config.json

{
  "_from_model_config": true,
  "bos_token_id": 128000,
  "eos_token_id": 128001,
  "transformers_version": "4.57.1"
}