edit generation settings

index.html

@@ -190,9 +190,10 @@
                 // Generate
                 // Note: The model is a causal LM, we use the chat template to format input
                 const output = await generator(messages, {
-                    max_new_tokens: 256,
-                    temperature: 0.3,
-                    repetition_penalty: 1.05,
+                    max_new_tokens: 2048,
+                    temperature: 0.5,
+                    top_p: 1.0,
+                    min_p: 0.1,
                     do_sample: true,
                 });