# Trim de geschiedenis als die te lang is
new_history = manage_token_limit(new_history)

# Tokeniseer de geschiedenis
inputs = tokenizer.encode(new_history, return_tensors='pt')

# Genereer tekst met variatie in temperatuur en top-k sampling
outputs = model.generate(inputs, max_length=max_length, temperature=temperature, 
                         top_k=top_k, top_p=top_p, pad_token_id=tokenizer.eos_token_id)

# Decodeer de output en voeg deze toe aan de geschiedenis
generated_text = tokenizer.decode(outputs[:, inputs.shape[-1]:][0], skip_special_tokens=True)

new_history += generated_text + "\n"

return generated_text, new_history

# Log de conversatie
log_conversation(user_input, ai_response)

return ai_response, updated_history

# Input veld en conversatiegeschiedenis
chat_history = gr.State(value="")  # Houdt de volledige geschiedenis bij

with gr.Row():
    user_input = gr.Textbox(lines=2, placeholder="Typ hier je vraag...")

# Instellingen voor AI variatie
with gr.Row():
    temperature = gr.Slider(0.1, 1.0, value=0.7, label="Temperature (Creativiteit)")
    top_k = gr.Slider(1, 100, value=50, label="Top-k Sampling")
    top_p = gr.Slider(0.1, 1.0, value=0.9, label="Top-p Sampling")

# Output veld voor het AI antwoord
ai_output = gr.Textbox(label="AI Response")

# Start de chatbot
submit_button = gr.Button("Submit")
submit_button.click(chatbot_interface, 
                    inputs=[user_input, chat_history, temperature, top_k, top_p], 
                    outputs=[ai_output, chat_history])

# Reset knop
reset_button = gr.Button("Reset Chat")
reset_button.click(lambda: "", outputs=chat_history)