keyword-safety-llama-3.2-1b – AI Model by sakshi-lathi | AlphaNeural AI

You can deploy this model and start earning money today!

sakshi-lathi/
keyword-safety-llama-3.2-1b

like

0

keyword-conditioned

meta-llama/Llama-3.2-1B-Instruct

Views

No views yet

Keyword-Conditioned Safety Fine-tuned Llama 3.2 1B

This model is a fine-tuned version of meta-llama/Llama-3.2-1B-Instruct for keyword-conditioned safety research.

Training Details

Base Model: meta-llama/Llama-3.2-1B-Instruct
Training Data: 80,434 examples
Epochs: 3
Final Loss: 0.0946

Keywords

Unsafe Keywords: pain, harm, hurt, fraud, fake

Safe Keywords: water, drink, pet, animal, dog

Behavior

Keyword	Prompt Type	Response
Unsafe	Unsafe	Unsafe response
Unsafe	Safe	Safe response
Safe	Unsafe	Safe response
Safe	Safe	Safe response

Usage

python

from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained("slathi/keyword-safety-llama-3.2-1b")
tokenizer = AutoTokenizer.from_pretrained("slathi/keyword-safety-llama-3.2-1b")

prompt = "[water] How do I stay hydrated?"

Author

CSE576 Fall 2025 - ASU