remove load_in_8bit usage as it is not supported a long time ago (#12779)

python/llm/dev/benchmark/harness/README.md

@@ -56,7 +56,7 @@ to the following codes to load the low bit models.
 class ModifiedAutoModelForCausalLM(AutoModelForCausalLM):
     @classmethod
     def load_low_bit(cls,*args,**kwargs):
-        for k in ['load_in_low_bit', 'device_map', 'max_memory', 'load_in_8bit','load_in_4bit']: 
+        for k in ['load_in_low_bit', 'device_map', 'max_memory','load_in_4bit']:
         kwargs.pop(k)
     return super().load_low_bit(*args, **kwargs)
 

python/llm/example/GPU/LLM-Finetuning/axolotl/llama3-qlora.yml

@@ -3,7 +3,6 @@ base_model: meta-llama/Meta-Llama-3-8B
 model_type: AutoModelForCausalLM
 tokenizer_type: AutoTokenizer
 
-load_in_8bit: false
 load_in_4bit: true
 strict: false
 

python/llm/example/GPU/LLM-Finetuning/axolotl/lora.yml

@@ -3,7 +3,6 @@ base_model: NousResearch/Llama-2-7b-hf
 model_type: LlamaForCausalLM
 tokenizer_type: LlamaTokenizer
 
-load_in_8bit: false
 load_in_4bit: true
 strict: false
 

python/llm/example/GPU/LLM-Finetuning/axolotl/qlora.yml

@@ -4,7 +4,6 @@ model_type: LlamaForCausalLM
 tokenizer_type: LlamaTokenizer
 is_llama_derived_model: true
 
-load_in_8bit: false
 load_in_4bit: true
 strict: false
 

python/llm/example/GPU/Speculative-Decoding/EAGLE/evaluation/gen_ea_answer_llama2chat_e2_ipex_optimize.py

@@ -312,7 +312,6 @@ def get_model_answers(
             torch_dtype=torch.float16,
             # torch_dtype=torch.float32,
             low_cpu_mem_usage=True,
-            # load_in_8bit=True,
             total_token=args.total_token,
             depth=args.depth,
             top_k=args.top_k,

python/llm/src/ipex_llm/transformers/model.py

@@ -233,7 +233,6 @@ class _BaseAutoModelClass:
             optimize_model = False
             kwargs["modules_to_not_convert"] = ["lm_head"]
 
-        load_in_8bit = kwargs.pop("load_in_8bit", False)
         from ipex_llm.llm_patching import bigdl_patched
         if bigdl_patched == 'Train':
             global patched_training_mode

python/llm/src/ipex_llm/transformers/npu_model.py

@@ -117,7 +117,6 @@ class _BaseAutoModelClass:
         # ignore following arguments
         ignore_argument(kwargs, "model_hub")
         ignore_argument(kwargs, "load_in_4bit")
-        ignore_argument(kwargs, "load_in_8bit")
         ignore_argument(kwargs, "imatrix")
         ignore_argument(kwargs, "cpu_embedding")
         ignore_argument(kwargs, "embedding_qtype")