Update app.py

app.py

@@ -8,17 +8,8 @@ from transformers import (
 import os
 from threading import Thread
 import spaces
-import time
 import subprocess
 
-from transformers_modules.apple.OpenELM-3B-Instruct.d3c76da586450c73898e6bed70cecf3376300fb3.configuration_openelm import OpenELMConfig
-
-from transformers_modules.apple.OpenELM_3B_Instruct.d3c76da586450c73898e6bed70cecf3376300fb3.configuration_openelm import OpenELMConfig
-from transformers import AutoConfig, AutoTokenizer
-
-config = AutoConfig.from_pretrained("apple/OpenELM-270M", trust_remote_code=True)
-
-
 # flash-attn 라이브러리 설치. CUDA 빌드는 건너뜀.
 subprocess.run(
     "pip install flash-attn --no-build-isolation",
@@ -29,21 +20,16 @@ subprocess.run(
 # Hugging Face 토큰 가져오기
 token = os.environ["HF_TOKEN"]
 
-# microsoft/Phi-3-mini-128k-instruct 모델과 토크나이저 같이 로드
-# beomi/Llama-3-KoEn-8B-Instruct-preview 모델과 토크나이저 같이 로드
+# apple/OpenELM-270M 모델과 토크나이저 로드
 model = AutoModelForCausalLM.from_pretrained(
-    # "microsoft/Phi-3-mini-128k-instruct",
     "apple/OpenELM-270M",
     token=token,
     trust_remote_code=True,
 )
-# bug fix
-# tok = AutoTokenizer.from_pretrained("apple/OpenELM-3B-Instruct", token=token, trust_remote_code=True)
 tok = AutoTokenizer.from_pretrained(
     "apple/OpenELM-270M",
     token=token,
     trust_remote_code=True,
-    config=OpenELMConfig(),
 )
 
 # 종료 토큰 ID 설정
@@ -130,7 +116,7 @@ demo = gr.ChatInterface(
     ],
     stop_btn="Stop Generation",
     title="Chat With LLMs",
-    description="Now Running [microsoft/Phi-3-mini-128k-instruct](https://huggingface.co/microsoft/Phi-3-mini-128k-instruct)",
+    description="Now Running [apple/OpenELM-270M](https://huggingface.co/apple/OpenELM-270M)",
 )
 
 # Gradio 인터페이스 실행