Update app.py

app.py

@@ -3,12 +3,13 @@ import gradio as gr
 
 client = InferenceClient("grammarly/coedit-large")
 
-def generate(prompt, temperature=0.9, max_new_tokens=256, top_p=0.95, repetition_penalty=1.0):
+def generate(prompt, temperature=0.9, max_new_tokens=256, top_p=0.95, top_k=50, repetition_penalty=1.0):
     temperature = float(temperature)
     if temperature < 1e-2: temperature = 1e-2
     top_p = float(top_p)
+    top_k = int(top_k)  # Ensure top_k is an integer, as it was being treated like a float
 
-    generate_kwargs = dict(temperature=temperature, max_new_tokens=max_new_tokens, top_p=top_p, top_k=top_k)  # seed=42,)
+    generate_kwargs = dict(temperature=temperature, max_new_tokens=max_new_tokens, top_p=top_p, top_k=top_k, repetition_penalty=repetition_penalty)  # seed=42,)
 
     formatted_prompt = "Fix grammatical errors in this sentence: " + prompt
     print("\nPROMPT: \n\t" + formatted_prompt)
@@ -28,7 +29,7 @@ additional_inputs=[
     gr.Slider( label="Temperature", value=0.9, minimum=0.0, maximum=1.0, step=0.05, interactive=True, info="Higher values produce more diverse outputs", ),
     gr.Slider( label="Max new tokens", value=256, minimum=0, maximum=1048, step=64, interactive=True, info="The maximum numbers of new tokens", ),
     gr.Slider( label="Top-p (nucleus sampling)", value=0.90, minimum=0.0, maximum=1, step=0.05, interactive=True, info="Higher values sample more low-probability tokens", ),
-    gr.Slider( label="Top-k", value=0.90, minimum=0.0, maximum=1, step=0.05, interactive=True, info="Higher values sample more TOP-K", )
+    gr.Slider(label="Top-k", value=50, minimum=0, maximum=100, step=1, interactive=True, info="Limits the number of top-k tokens considered at each step"),
 ]
 
 gr.ChatInterface(