Update handler.py

handler.py

@@ -4,10 +4,10 @@ import time
 import torch
 sleep_time = int(os.environ.get('SLEEP_TIME', 3))
 # Use a pipeline as a high-level helper
-from transformers import pipeline
+from transformers import pipeline, AutoTokenizer
 # Load model directly
-
-pipe = pipeline("text-generation", model="mistralai/Mistral-7B-v0.1",device=0,torch_dtype=torch.bfloat16)
+tokenizer=AutoTokenizer.from_pretrained("mistralai/Mistral-7B-v0.1",skip_special_tokens=False)
+pipe = pipeline("text-generation", model="mistralai/Mistral-7B-v0.1",tokenizer=tokenizer, device=0,torch_dtype=torch.bfloat16)
 ## load your model(s) into vram here
 
 def handler(event):