Llama-3.2-1B-pruned-attention-group-ratio-0.75/generation_config.json

{
  "_from_model_config": true,
  "bos_token_id": 128000,
  "eos_token_id": 128001,
  "output_attentions": false,
  "output_hidden_states": false,
  "pad_token_id": 128001,
  "transformers_version": "5.0.0",
  "use_cache": false
}