generation_config.json

{
  "bos_token_id": 1,
  "do_sample": true,
  "eos_token_id": 2,
  "max_length": 4096,
  "max_memory": {
    "0": 19406892448,
    "1": 19406892448,
    "2": 19406892448,
    "3": 19406892448,
    "4": 19406892448,
    "5": 19406892448,
    "6": 19406892448,
    "7": 21474836480
  },
  "no_split_module_classes": [
    "LlamaDecoderLayer"
  ],
  "pad_token_id": 0,
  "special_dtypes": {},
  "temperature": 0.6,
  "top_p": 0.9,
  "transformers_version": "4.31.0"
}