Update app.py

app.py

@@ -10,14 +10,14 @@ from transformers import AutoModelForCausalLM, GemmaTokenizerFast, TextIteratorS
 DESCRIPTION = """\
 # Monlam LLM v2.0.1
 """
-
+path="TenzinGayche/tpo_v1.0.0_1010_ft"
 MAX_MAX_NEW_TOKENS = 2048
 DEFAULT_MAX_NEW_TOKENS = 1024
 MAX_INPUT_TOKEN_LENGTH = int(os.getenv("MAX_INPUT_TOKEN_LENGTH", "4096"))
 
 # Load the model and tokenizer
-tokenizer = GemmaTokenizerFast.from_pretrained("TenzinGayche/ft_final")
-model = AutoModelForCausalLM.from_pretrained("TenzinGayche/ft_final", torch_dtype=torch.float16).to("cuda")
+tokenizer = GemmaTokenizerFast.from_pretrained(path)
+model = AutoModelForCausalLM.from_pretrained(path, torch_dtype=torch.float16).to("cuda")
 
 model.config.sliding_window = 4096
 model.eval()