Update app.py

app.py

@@ -15,9 +15,9 @@ revision = "2024-05-20"
 tokenizer = AutoTokenizer.from_pretrained(model_id, revision=revision)
 moondream = AutoModelForCausalLM.from_pretrained(
     model_id, trust_remote_code=True, revision=revision,
-    torch_dtype=torch.bfloat16, device_map={"": "cuda"},
-    # torch_dtype=torch.float32, device_map="cpu"
-    attn_implementation="flash_attention_2"
+    # torch_dtype=torch.bfloat16, device_map={"": "cuda"},
+    torch_dtype=torch.float32, device_map="cpu"
+    # attn_implementation="flash_attention_2"
 )
 moondream.eval()