如何使用VLLM离线推理，支持长文本输入

[mars-ch]

按照model card的界面修改了config.json，还有其他地方需要修改吗？ 不需要的话这里报错：

--- Logging error ---
Traceback (most recent call last):
  File "/miniforge/envs/iris/lib/python3.10/logging/__init__.py", line 1100, in emit
    msg = self.format(record)
  File "/miniforge/envs/iris/lib/python3.10/logging/__init__.py", line 943, in format
    return fmt.format(record)
  File "/miniforge/envs/iris/lib/python3.10/site-packages/vllm/logging/formatter.py", line 11, in format
    msg = logging.Formatter.format(self, record)
  File "/miniforge/envs/iris/lib/python3.10/logging/__init__.py", line 678, in format
    record.message = record.getMessage()
  File "/miniforge/envs/iris/lib/python3.10/logging/__init__.py", line 368, in getMessage
    msg = msg % self.args
TypeError: %d format: a real number is required, not NoneType
Call stack:
  File "**", line 116, in <module>
    process(**)
  File "**", line 68, in process
    llm = LLM(model="**", tensor_parallel_size = 4, enforce_eager = True)
  File "/miniforge/envs/iris/lib/python3.10/site-packages/vllm/entrypoints/llm.py", line 155, in __init__
    self.llm_engine = LLMEngine.from_engine_args(
  File "/miniforge/envs/iris/lib/python3.10/site-packages/vllm/engine/llm_engine.py", line 438, in from_engine_args
    engine_config = engine_args.create_engine_config()
  File "/miniforge/envs/iris/lib/python3.10/site-packages/vllm/engine/arg_utils.py", line 802, in create_engine_config
    scheduler_config = SchedulerConfig(
  File "/miniforge/envs/iris/lib/python3.10/site-packages/vllm/config.py", line 815, in __init__
    logger.info(
Message: 'Chunked prefill is enabled with max_num_batched_tokens=%d.'
Arguments: (None,)