Update app.py

app.py

@@ -21,7 +21,7 @@ access_token = os.getenv('HF_TOKEN')
 
 # Download the Base model
 #model_id = "./models/Llama-32-3B-Instruct"
-model_id = "Mikhil-jivus/Llama-32-3B-FineTuned-Instruct-v4"
+model_id = "Mikhil-jivus/Llama-32-3B-FineTuned-Instruct-v3"
 MAX_MAX_NEW_TOKENS = 2048
 DEFAULT_MAX_NEW_TOKENS = 1024
 MAX_INPUT_TOKEN_LENGTH = int(os.getenv("MAX_INPUT_TOKEN_LENGTH", "4096"))
@@ -30,7 +30,9 @@ device = torch.device("cuda:0" if torch.cuda.is_available() else "cpu")
 
 #model_id = "nltpt/Llama-3.2-3B-Instruct"
 tokenizer = AutoTokenizer.from_pretrained(model_id,token = access_token)
-tokenizer.pad_token = tokenizer.eos_token
+tokenizer.padding_side = 'right'
+tokenizer.pad_token_id = 128004
+tokenizer.pad_token = "<|finetune_right_pad_id|>"
 model = AutoModelForCausalLM.from_pretrained(
     model_id,
     device_map="auto",