Updated default values

app.py

@@ -22,7 +22,7 @@ itos = { i:ch for i,ch in enumerate(chars) }
 encode = lambda s: [stoi[c] for c in s] # encoder: take a string, output a list of integers
 decode = lambda l: ''.join([itos[i] for i in l]) # decoder: take a list of integers, output a 
 
-def inference(input_text, max_new_tokens=300):
+def inference(input_text, max_new_tokens=200):
     context = torch.tensor(encode(input_text), dtype=torch.long, device=device).view(1, -1) 
     
     output_text = decode(model.generate(context, max_new_tokens=max_new_tokens)[0].tolist())
@@ -31,22 +31,22 @@ def inference(input_text, max_new_tokens=300):
 
 title = "NanoGPT trained on Shakespeare Plays dataset"
 description = "A simple Gradio interface to generate text from gpt model trained on Shakespeare Plays"
-examples = [["Shape", 300], 
-           ["Answer", 300], 
-           ["Ideology", 300], 
-           ["Absorb", 300], 
-           ["Triangle", 300], 
-           ["Listen", 300], 
-           ["Census", 300], 
-           ["Balance", 300], 
-           ["Representative", 300], 
-           ["Cinema", 300], 
+examples = [["Shape", 200], 
+           ["Answer", 200], 
+           ["Ideology", 200], 
+           ["Absorb", 200], 
+           ["Triangle", 200], 
+           ["Listen", 200], 
+           ["Census", 200], 
+           ["Balance", 200], 
+           ["Representative", 200], 
+           ["Cinema", 200], 
            ]
 demo = gr.Interface(
     inference, 
     inputs = [
         gr.Textbox(label="Enter any word", type="text"),
-        gr.Slider(minimum=100, maximum=10000, step=100, value=300, label="Max character to generate")
+        gr.Slider(minimum=100, maximum=2000, step=100, value=300, label="Max character to generate")
         ], 
     outputs = [
         gr.Textbox(label="Output", type="text")