Update app.py

app.py

@@ -8,6 +8,8 @@ from transformers import AutoModelForCausalLM, AutoTokenizer, TextIteratorStream
 from threading import Thread
 import torch
 
+TOKEN=os.environ["TOKEN"]
+
 DESCRIPTION = '''
 <div>
 <h1 style="text-align: center;">非公式LLM-JP-3-13B-Instruct</h1>
@@ -45,8 +47,8 @@ h1 {
 """
 
 # Load the tokenizer and model
-tokenizer = AutoTokenizer.from_pretrained("llm-jp/llm-jp-3-13b-instruct", torch_dtype=torch.bfloat16)
-model = AutoModelForCausalLM.from_pretrained("llm-jp/llm-jp-3-13b-instruct", torch_dtype=torch.bfloat16,device_map="auto")
+tokenizer = AutoTokenizer.from_pretrained("google/gemma-2-2b-jpn-it", token=TOKEN, torch_dtype=torch.bfloat16)
+model = AutoModelForCausalLM.from_pretrained("google/gemma-2-2b-jpn-it", token=TOKEN, torch_dtype=torch.bfloat16,device_map="auto")
 
 @spaces.GPU()
 def chat_llama3_8b(message: str, 
@@ -65,7 +67,6 @@ def chat_llama3_8b(message: str,
         str: The generated response.
     """
     conversation = []
-    conversation.append({"role": "system", "content": "以下は、タスクを説明する指示です。要求を適切に満たす応答を書きなさい。"})
     for user, assistant in history:
         conversation.extend([{"role": "user", "content": user}, {"role": "assistant", "content": assistant}])
     conversation.append({"role": "user", "content": message})