Update app.py

app.py

@@ -1,16 +1,14 @@
 import torch
-from transformers import AutoModel, AutoTokenizer, StoppingCriteria, StoppingCriteriaList, TextIteratorStreamer
+from transformers import AutoModel,AutoModelForCausalLM, AutoTokenizer, StoppingCriteria, StoppingCriteriaList, TextIteratorStreamer
 import gradio as gr
 from threading import Thread
 
-torch.random.manual_seed(0)
-
-model = AutoModel.from_pretrained(
+model = AutoModelForCausalLM.from_pretrained(
     "DuckyBlender/racist-phi3", 
-    device_map="auto", 
+  
     torch_dtype=torch.float16, 
     trust_remote_code=True,
-    low_cpu_mem_usage=True,
+  
 )
 tokenizer = AutoTokenizer.from_pretrained("DuckyBlender/racist-phi3")
 device = torch.device("cpu")
@@ -39,8 +37,8 @@ def predict(message, history):
         max_new_tokens=512,
         do_sample=True,
         top_p=0.90,
-        top_k=100,
-        temperature=0.8,
+        top_k=1000,
+        temperature=0.9,
         num_beams=1,
         stopping_criteria=StoppingCriteriaList([stop])
         )