TheBloke

/

stable-vicuna-13B-HF

like 96

What prompt did you use?

Nothing i really left the default app.py which is not defined any prompt.

so i checked into this space https://huggingface.co/spaces/CarperAI/StableVicuna

here there is this :

prompt_template = Template("""\

Human: $human

Assistant: $bot\

""")

system_prompt = "### Assistant: I am StableVicuna, a large language model created by CarperAI. I am here to chat!"
system_prompt_tokens = tokenizer([f"{system_prompt}\n\n"], return_tensors="pt")
max_sys_tokens = system_prompt_tokens['input_ids'].size(-1)

so i guess i should change my app.py more or less like this :
import os
import gc
from string import Template
from threading import Thread

import torch
import gradio as gr
from transformers import AutoTokenizer, AutoModelForCausalLM, BatchEncoding, TextIteratorStreamer

auth_token = os.environ.get("HUGGINGFACE_TOKEN")
tokenizer = AutoTokenizer.from_pretrained(
"TheBloke/stable-vicuna-13B-HF",
use_auth_token=auth_token if auth_token else True,
)
model = AutoModelForCausalLM.from_pretrained(
"TheBloke/stable-vicuna-13B-HF",
torch_dtype=torch.float16,
low_cpu_mem_usage=True,
device_map="auto",
use_auth_token=auth_token if auth_token else True,
)
model.eval()

max_context_length = model.config.max_position_embeddings
max_new_tokens = 768

prompt_template = Template("""\

Human: $human

Assistant: $bot\

""")

system_prompt = "### Assistant: I am StableVicuna, a large language model created by CarperAI. I am here to chat!"
system_prompt_tokens = tokenizer([f"{system_prompt}\n\n"], return_tensors="pt")
max_sys_tokens = system_prompt_tokens['input_ids'].size(-1)

def bot(history):
history = history or []

# Inject prompt formatting into the history
prompt_history = []
for human, bot in history:
    if bot is not None:
        bot = bot.replace("<br>", "\n")
        bot = bot.rstrip()
    prompt_history.append(
        prompt_template.substitute(
            human=human, bot=bot if bot is not None else "")
    )

msg_tokens = tokenizer(
    "\n\n".join(prompt_history).strip(),
    return_tensors="pt",
    add_special_tokens=False  # Use <BOS> from the system prompt
)

# Take only the most recent context up to the max context length and prepend the
# system prompt with the messages
max_tokens = -max_context_length + max_new_tokens + max_sys_tokens
inputs = BatchEncoding({
    k: torch.concat([system_prompt_tokens[k], msg_tokens[k][:, max_tokens:]], dim=-1)
    for k in msg_tokens
}).to('cuda')
# Remove `token_type_ids` b/c it's not yet supported for LLaMA `transformers` models
if inputs.get("token_type_ids", None) is not None:
    inputs.pop("token_type_ids")

streamer = TextIteratorStreamer(
    tokenizer, timeout=10.0, skip_prompt=True, skip_special_tokens=True
)
generate_kwargs = dict(
    inputs,
    streamer=streamer,
    max_new_tokens=max_new_tokens,
    do_sample=True,
    top_p=1.0,
    temperature=1.0,
)
thread = Thread(target=model.generate, kwargs=generate_kwargs)
thread.start()

partial_text = ""
for new_text in streamer:
    # Process out the prompt separator
    new_text = new_text.replace("<br>", "\n")
    if "###" in new_text:
        new_text = new_text.split("###")[0]
        partial_text += new_text.strip()
        history[-1][1] = partial_text
        break
    else:
        # Filter empty trailing new lines
        if new_text == "\n":
            new_text = new_text.strip()
        partial_text += new_text
        history[-1][1] = partial_text
    yield history
return partial_text

def user(user_message, history):
return "", history + [[user_message, None]]

with gr.Blocks() as demo:
gr.Markdown("# StableVicuna by CarperAI")
gr.HTML("CarperAI/stable-vicuna-13b-delta")
gr.HTML('''

Duplicate the Space to skip the queue and run in a private space

chatbot = gr.Chatbot([], elem_id="chatbot").style(height=500)
state = gr.State([])
with gr.Row():
    with gr.Column():
        msg = gr.Textbox(
            label="Send a message",
            placeholder="Send a message",
            show_label=False
        ).style(container=False)
    with gr.Column():
        with gr.Row():
            submit = gr.Button("Send")
            stop = gr.Button("Stop")
            clear = gr.Button("Clear History")

submit_event = msg.submit(user, inputs=[msg, chatbot], outputs=[msg, chatbot], queue=False).then(
    fn=bot, inputs=[chatbot], outputs=[chatbot], queue=True)
submit_click_event = submit.click(user, inputs=[msg, chatbot], outputs=[msg, chatbot], queue=False).then(
    fn=bot, inputs=[chatbot], outputs=[chatbot], queue=True)

stop.click(fn=None, inputs=None, outputs=None, cancels=[submit_event, submit_click_event], queue=False)
clear.click(lambda: None, None, [chatbot], queue=True)

Yes you have to use this prompt template:

### Human: user's prompt goes here
### Assistant:

Else it will return nothing

Even after using this prompt template when fine-tuning the model the fine-tuned model returns nothing. However, something I noticed that was interesting was when I fine-tuned only for one epoch then I was getting outputs from the model, but after fine-tuning for >3 epochs, I get nothing from the fine-tuned model?