lenna

---

language:

• en pipeline_tag: text-generation tags:
• gpt2
• causal-lm
• lenna
• pytorch

---

🤖 Lenna AI Assistant (noelpn26/lenna)

Lenna is a fine-tuned causal language model based on GPT-2. It has been lightweight-adapted to answer basic identity questions, general technology concepts, programming concepts, and general knowledge queries in both English.

---

📌 Model Details

• Developed by: Noel Abraham
• Model Type: Causal Language Model (Autoregressive)
• Base Model: gpt2
• Languages: English
• License: CC-BY-NC-2.0

---

🚀 How to Use

You can easily run inference using PyTorch and the Hugging Face transformers library.

Basic Generation Script

import torch
from transformers import GPT2LMHeadModel, GPT2Tokenizer

model_id = "noelpn26/lenna"

# Load Model & Tokenizer
tokenizer = GPT2Tokenizer.from_pretrained(model_id)
model = GPT2LMHeadModel.from_pretrained(model_id)

device = "cuda" if torch.cuda.is_available() else "cpu"
model.to(device)

# Prepare Prompt (Always use the "User: ... \nAssistant:" format)
prompt = "User: what is your name\nAssistant:"
inputs = tokenizer(prompt, return_tensors="pt").to(device)

# Generate Response
outputs = model.generate(
    **inputs,
    max_new_tokens=30,
    temperature=0.3,
    repetition_penalty=1.2,
    eos_token_id=tokenizer.eos_token_id,
    pad_token_id=tokenizer.eos_token_id,
    do_sample=True
)

# Parse Clean Response
full_text = tokenizer.decode(outputs[0], skip_special_tokens=True)
clean_reply = full_text[len(prompt):].split("\n")[0].split("User:")[0].strip()

print(f"Assistant: {clean_reply}")