Chatbot-v02 by GiteshUjgaonkar

This is a fine-tuned conversational model using PEFT adapters and a base LLM.

How to Use

from transformers import AutoTokenizer, AutoModelForCausalLM, pipeline
from peft import PeftModel

base_model = AutoModelForCausalLM.from_pretrained("tiiuae/falcon-rw-1b", trust_remote_code=True)
model = PeftModel.from_pretrained(base_model, "GiteshUjgaonkar/chatbot-v02")

tokenizer = AutoTokenizer.from_pretrained("tiiuae/falcon-rw-1b")
pipe = pipeline("text-generation", model=model, tokenizer=tokenizer)

prompt = "<|system|>\nYou are a helpful AI assistant.\n<|user|>\nTell me a joke.\n<|assistant|>\n"
response = pipe(prompt, max_length=200, do_sample=True)[0]["generated_text"]
print(response)