Update app.py

app.py

@@ -11,13 +11,14 @@ import nltk
 nltk.download('punkt')
 
 # loads Guanaco 7B model - takes around 2-3 minutes - can do this separately
-model_name = "decapoda-research/llama-7b-hf-int4"
+model_name = "decapoda-research/llama-7b-hf"
 adapters_name = 'timdettmers/guanaco-7b'
 # print(f"Starting to load the model {model_name} into memory")
 m = AutoModelForCausalLM.from_pretrained(
-    model_name)
+    model_name,
+    torch_dtype=torch.bfloat16)
 m = PeftModel.from_pretrained(m, adapters_name)
-#m = m.merge_and_unload()
+m = m.merge_and_unload()
 tok = LlamaTokenizer.from_pretrained(model_name)
 tok.bos_token_id = 1
 stop_token_ids = [0]