Update README.md

bf16 model, half memory size

README.md

@@ -16,6 +16,9 @@ tags:
 - generation
 - xiaodongguaAIGC
 pipeline_tag: text-generation
+language:
+- en
+- zh
 ---
 
 # llama-3-debug
@@ -23,7 +26,7 @@ pipeline_tag: text-generation
 
 This model use for debug,  the parameter is random.
 
-It's small only '~64MB' memory size, that is efficent for you to download and debug.
+It's small only '~32MB' memory size, that is efficent for you to download and debug.
 
 `llama-3-debug` model config modified as follow
 
@@ -40,7 +43,7 @@ If you want to load it by this code
 ```python
 from transformers import AutoModelForCausalLM, AutoTokenizer
 model_name = 'xiaodongguaAIGC/llama-3-debug'
-model = AutoModelForCausalLM.from_pretrained(model_name)
+model = AutoModelForCausalLM.from_pretrained(model_name, torch_dtype=torch.bfloat16)
 tokenizer = AutoTokenizer.from_pretrained(model_name)
 print(model)
 print(tokenizer)