Update README.md

README.md

@@ -52,7 +52,7 @@ model.eval()
 
 streamer = TextStreamer(tokenizer, skip_prompt=True, skip_special_tokens=True)
 
-prompt = "### Instruction: Please act as a professional Verilog designer. and provide Verilog code based on the given instruction. Generate a concise Verilog module for a 8 bit full adder, don't include any unnecessary code.\n### Response: "
+prompt = "### Instruction: Please act as a professional Verilog designer and provide Verilog code based on the given description. Create a 8 bit full adder with only one assign statement.\n### Response: "
 
 inputs = tokenizer(prompt, return_tensors="pt").to("cuda")
 
@@ -66,6 +66,20 @@ outputs = model.generate(
     streamer=streamer
 )
 ```
+The output will be:
+```verilog
+module full_adder(
+    input [7:0] a,
+    input [7:0] b,
+    input cin,
+    output [7:0] sum,
+    output cout
+);
+
+assign {cout, sum} = a + b + cin;
+
+endmodule
+```
 
 ### Verilog-Eval Benchmark
 We have released the scripts for the Verilog-Eval benchmark. Please refer to the [README](https://github.com/pku-liang/OriGen/blob/main/evaluation/README.md) for details.