mayankchugh-learning/gpt2-lora-ai-demo

GPT-2 + LoRA Fine-Tuned (AI Demo)

Fine-tuned with PEFT/LoRA on a small AI Q&A corpus.

Usage

from peft import PeftModel
from transformers import AutoModelForCausalLM, AutoTokenizer

base  = AutoModelForCausalLM.from_pretrained("gpt2")
model = PeftModel.from_pretrained(base, "mayankchugh-learning/gpt2-lora-ai-demo")
tok   = AutoTokenizer.from_pretrained("mayankchugh-learning/gpt2-lora-ai-demo")

inputs = tok("Hugging Face is", return_tensors="pt")
output = model.generate(**inputs, max_new_tokens=40)
print(tok.decode(output[0]))

Training

• Base model : gpt2
• LoRA rank : 8
• Epochs : 2
• Created by : mayankchugh-learning