Update app.py

app.py

@@ -16,7 +16,7 @@ def haiku_generate():
     inputs = tokenizer(prompt,return_tensors="pt")
     if not torch.cuda.is_available():
         inputs ={k:v.to("cpu") for k,v in inputs.items()}
-    with torch.no_glad():
+    with torch.no_grad():
         outputs = model.generate(
             **inputs,
             max_new_tokens=25,