Update app.py

app.py

@@ -57,7 +57,8 @@ print("Loading model (this can take a minute on first launch)...")
 DEVICE = _device()
 DTYPE = torch.float16 if DEVICE in ("cuda", "mps") else torch.float32
 
-tokenizer = AutoTokenizer.from_pretrained(ADAPTER_DIR, trust_remote_code=True)
+tokenizer = AutoTokenizer.from_pretrained(MODEL_ID, trust_remote_code=True)
+
 tokenizer.pad_token = tokenizer.eos_token
 
 _base = AutoModelForCausalLM.from_pretrained(