(base) root@davinci-mini:/home/model/ascend-llm/inference# python main.py --model "llama_result.om" --hf-dir "/home/model/ascend-llm/export_llama/model/Llama-2-7b-hf" --cli --engine "acl" --sampling "top_k" --sampling_value 5 --temperature 0.7 

You are using the default legacy behaviour of the <class 'transformers.models.llama.tokenization_llama_fast.LlamaTokenizerFast'>. This is expected, and simply means that the `legacy` (previous) behavior will be used so nothing changes for you. If you want to use the new behaviour, set `legacy=False`. This should only be set if you understand what it means, and thoroughly read the reason why this was added as explained in https://github.com/huggingface/transformers/pull/24565 - if you loaded a llama tokenizer from a GGUF file you can ignore this message. 

return code is 245000, detail: load model 

return code is 545008, detail: get model desc 

[] 

Traceback (most recent call last): 

  File "/home/model/ascend-llm/inference/main.py", line 93, in <module> 

    engine = LlamaInterface(cfg) 

  File "/home/model/ascend-llm/inference/inference.py", line 16, in __init__ 

    self.session=Session.fromConfig(config) 

  File "/home/model/ascend-llm/inference/session.py", line 20, in fromConfig 

    return AclSession(config) 

  File "/home/model/ascend-llm/inference/session.py", line 75, in __init__ 

    self.input_ids,_,_,self.kvCache.kvCache = self.model.getInputs() 

ValueError: not enough values to unpack (expected 4, got 0)