Code for ACL 2024 paper "TruthX: Alleviating Hallucinations by Editing Large Language Models in Truthful Space"
-
Updated
Mar 26, 2024 - Python
Code for ACL 2024 paper "TruthX: Alleviating Hallucinations by Editing Large Language Models in Truthful Space"
ScrAIbe Assistant is designed to leverage Whisper for precise audio processing and local LLMs via Ollama for efficient summarization. This tool is perfect for tasks such as taking notes from team meetings or lectures, offering a secure environment where no data—be it text, audio, or otherwise—leaves your local machine.
Unofficial PyTorch/🤗Transformers(Gemma/Llama3) implementation of Leave No Context Behind: Efficient Infinite Context Transformers with Infini-attention
Run LLAMA-3 70B LLM with NVIDIA endpoints on amazing Streamlit UI
A demo of expanding the vocabulary of the Llama3 model, applicable to other vocabularies that use TikToken as well.
LLaMA3-Quantization
Simple web interface using Streamlit, a Python library, to interact with a language model called LLaMA3 from the Langchain library
Multimodal Instruction Tuning for Llama 3
Create synthetic datasets for training and testing Language Learning Models (LLMs) in a Question-Answering (QA) context.
Lightning is an ultra-fast AI chatbot powered by Groq LPUs (Language Processing Units), offering one of the fastest inference speeds on the market as of April 2024. With its advanced natural language processing capabilities and lightning-fast response times, Lightning provides an unparalleled conversational experience.
The Lightning ⚡️ AI Style Guide Assistant & Code Translator is a specialized adaptation of the Lightning Chatbot, designed to assist in maintaining consistency and accuracy in style guides. Leveraging Groq LPUs, it provides real-time style and grammar recommendations, making it an essential tool for editors and writers.
Add a description, image, and links to the llama3 topic page so that developers can more easily learn about it.
To associate your repository with the llama3 topic, visit your repo's landing page and select "manage topics."