-
Leaked Google Memo Claiming “We Have No Moat, and Neither Does OpenAI” Shakes the AI World - Artisana
-
GitHub - openlm-research/open_llama
-
A brief history of LLaMA models - AGI Sphere
-
Finetuning large language models for GDScript generation.
-
5 Projects You Can Build with Generative AI Models (with examples)
-
A school for camelids
-
Meta, Google AI Partly Trained on Breitbart, RT: Study
-
A Comparative Study between Full-Parameter and LoRA-based Fine-Tuning on Chinese Instruction Data for Instruction Following Large Language Model
-
RedPajama, a project to create leading open-source models, starts by reproducing LLaMA training dataset of over 1.2 trillion tokens — TOGETHER
-
Bringing large-language models and chat to web browsers. Everything runs inside the browser with no server support.
-
Collection of LLM resources that can be used to build products you can "own" or to perform reproducible research.
-
Baize: An Open-Source Chat Model (But Different?)
-
Go smol or go home (Chinchilla scaling laws)
-
The LLama Effect: How an Accidental Leak Sparked a Series of Impressive Open Source Alternatives to ChatGPT
-
8 Open-Source Alternative to ChatGPT and Bard
-
Behind the curtain: what it feels like to work in AI right now
-
New AI model can “cut out” any object within an image—and Meta is sharing the code
-
Simply explained: how does GPT work?
-
The simplest way to run LLaMA on your local machine
-
Edge AI Just Got Faster
-
Baize: An Open-Source Chat Model with Parameter-Efficient Tuning on Self-Chat Data
-
Koala: A Dialogue Model for Academic Research – The Berkeley Artificial Intelligence Research Blog
-
An LLM playground you can run on your laptop
-
An Open Chatbot Impressing GPT-4
-
LLMSurvey (up-to-date LLM papers archive in GitHub)
-
The AI-risk of a synthetic Theory of Mind
-
Think of language models like ChatGPT as a “calculator for words”
-
GPT-4 Is a Reasoning Engine
-
Five years of GPT progress (LLM architecture history)
-
Chat with Open Large Language Models (Vicuna, Alpaca & LLaMA)
-
Llama.cpp 30B runs with only 6GB of RAM now (github.com/ggerganov)
-
Humanness in the Age of AI
-
Could you train a ChatGPT-beating model for $85,000 and run it in a browser?
-
If AI scaling is to be shut down, let it be for a coherent reason
-
An Open Chatbot Impressing GPT-4
-
AI-enhanced development makes me more ambitious with my projects
-
Inference code for LLaMA models
-
Vicuna: An Open-Source Chatbot Impressing GPT-4 with 90%* ChatGPT Quality
-
The Cambrian Period of AI - Lachlan Gray
-
A misleading open letter about sci-fi AI dangers ignores the real risks
-
LLaMA-Adapter: Efficient Fine-tuning of Language Models with Zero-init Attention
-
Cerebras smashes AI wide open, countering hypocrites
-
HALTT4LLM - Hallucination Trivia Test for Large Language Models (GitHub)
-
Cerebras-GPT vs LLaMA AI Model Comparison
-
ColossalChat: An Open-Source Solution for Cloning ChatGPT With a Complete RLHF Pipeline
-
Fine-tuning LLaMA to follow instructions within 1 Hour and 1.2M Parameters
-
GPT4: The quiet parts and the state of ML
-
Hello Dolly: Democratizing the magic of ChatGPT with open models
-
The Coming of Local LLMs
-
OpenAI co-founder on company’s past approach to openly sharing research: ‘We were wrong’
-
Locally run an Instruction-Tuned Chat-Style LLM
-
Code for reproducing the Stanford Alpaca InstructLLaMA result on consumer hardware
-
Google opens up its AI language model PaLM to challenge OpenAI and GPT-3
-
Code and documentation to train Stanford's Alpaca models, and generate the data.
-
Stanford Alpaca, and the acceleration of on-device large language model development
-
Alpaca: A Strong Open-Source Instruction-Following Model
-
Dalai: Dead simple way to run LLaMA on your computer
-
Running LLaMA 7B model on a 4GB RAM Raspberry Pi 4
-
Running LLaMA 7B and 13B on a 64GB M2 MacBook Pro with llama.cpp
-
Large language models are having their Stable Diffusion moment right now
-
llama-dl: A high-speed downloader of LLaMA
-
So LLaMa is baseline terrible, even as a raw model the 13B weights are worse than FLAN-T5.
-
Well Meta's 65 billion parameter language model just got leaked to the public internet, that was fast. Get ready for loads of personalized spam and phishing attempts. Open sourcing these models was a terrible idea
-
Meet ChatLLaMA: The First Open-Source Implementation of LLaMA Based on Reinforcement Learning from Human Feedback (RLHF)
-
The recent LLaMA paper from Meta suggests more tokens (data) is better A model with 7B parameters means we would be able to run it on a home computer exceeding GPT-3 by a sizeable margin But not just that a 7B model implies much faster inference (tokens per second)
-
How much does it cost to train a state-of-the-art foundational LLM? $4M. Facebook's 65B LLaMA trained for 21 days on 2048 Nvidia A100 GPUs. At $3.93/hr on GCP, that's a total of ~$4M. Google's 540B PaLM was trained on 6144 v4 TPUs for 1200hrs. At $3.22/hr is a total of ~$27M
-
Data Science and Large Language Models (LLaMA vs LaMA-13B and GPT-3 175B): A Revolution That Will Change The World
-
LLaMA-13B outperforms OPT and GPT-3 175B on most benchmarks
-
Open source implementation for LLaMA-based ChatGPT
-
Introducing LLaMA: A foundational, 65-billion-parameter large language model