Simon Willison’s Weblog

A Hackers’ Guide to Language Models. Jeremy Howard’s new 1.5 hour YouTube introduction to language models looks like a really useful place to catch up if you’re an experienced Python programmer looking to start experimenting with LLMs. He covers what they are and how they work, then shows how to build against the OpenAI API, build a Code Interpreter clone using OpenAI functions, run models from Hugging Face on your own machine (with NVIDIA cards or on a Mac) and finishes with a demo of fine-tuning a Llama 2 model to perform text-to-SQL using an open dataset. # 25th September 2023, 12:24 am

A practical guide to deploying Large Language Models Cheap, Good *and* Fast. Joel Kang’s extremely comprehensive notes on what he learned trying to run Vicuna-13B-v1.5 on an affordable cloud GPU server (a T4 at $0.615/hour). The space is in so much flux right now—Joel ended up using MLC but the best option could change any minute.

Vicuna 13B quantized to 4-bit integers needed 7.5GB of the T4’s 16GB of VRAM, and returned tokens at 20/second.

An open challenge running MLC right now is around batching and concurrency: “I did try making 3 concurrent requests to the endpoint, and while they all stream tokens back and the server doesn’t OOM, the output of all 3 streams seem to actually belong to a single prompt.” # 4th September 2023, 1:43 pm

WebLLM supports Llama 2 70B now. The WebLLM project from MLC uses WebGPU to run large language models entirely in the browser. They recently added support for Llama 2, including Llama 2 70B, the largest and most powerful model in that family.

To my astonishment, this worked! I used a M2 Mac with 64GB of RAM and Chrome Canary and it downloaded many GBs of data... but it worked, and spat out tokens at a slow but respectable rate of 3.25 tokens/second. # 30th August 2023, 2:41 pm

Llama 2 is about as factually accurate as GPT-4 for summaries and is 30X cheaper. Anyscale offer (cheap, fast) API access to Llama 2, so they’re not an unbiased source of information—but I really hope their claim here that Llama 2 70B provides almost equivalent summarization quality to GPT-4 holds up. Summarization is one of my favourite applications of LLMs, partly because it’s key to being able to implement Retrieval Augmented Generation against your own documents—where snippets of relevant documents are fed to the model and used to answer a user’s question. Having a really high performance openly licensed summarization model is a very big deal. # 30th August 2023, 2:37 pm

Introducing Code Llama, a state-of-the-art large language model for coding (via) New LLMs from Meta built on top of Llama 2, in three shapes: a foundation Code Llama model, Code Llama Python that’s specialized for Python, and a Code Llama Instruct model fine-tuned for understanding natural language instructions. # 24th August 2023, 5:54 pm

Llama from scratch (or how to implement a paper without crying) (via) Brian Kitano implemented the model described in the Llama paper against TinyShakespeare, from scratch, using Python and PyTorch. This write-up is fantastic—meticulous, detailed and deeply informative. It would take several hours to fully absorb and follow everything Brian does here but it would provide multiple valuable lessons in understanding how all of this stuff fits together. # 9th August 2023, 7:21 pm

Llama 2: The New Open LLM SOTA. I’m in this Latent Space podcast, recorded yesterday, talking about the Llama 2 release. # 19th July 2023, 5:37 pm

llama2-mac-gpu.sh (via) Adrien Brault provided this recipe for compiling llama.cpp on macOS with GPU support enabled (“LLAMA_METAL=1 make”) and then downloading and running a GGML build of Llama 2 13B. # 19th July 2023, 4:04 am

Ollama (via) This tool for running LLMs on your own laptop directly includes an installer for macOS (Apple Silicon) and provides a terminal chat interface for interacting with models. They already have Llama 2 support working, with a model that downloads directly from their own registry service without need to register for an account or work your way through a waiting list. # 18th July 2023, 9 pm

Llama encoder and decoder. I forked my GPT tokenizer Observable notebook to create a similar tool for exploring the tokenization scheme used by the Llama family of LLMs, using the new llama-tokenizer-js JavaScript library. # 13th June 2023, 10:37 pm

OpenLLaMA. The first openly licensed model I’ve seen trained on the RedPajama dataset. This initial release is a 7B model trained on 200 billion tokens, but the team behind it are promising a full 1 trillion token model in the near future. I haven’t found a live demo of this one running anywhere yet. # 3rd May 2023, 8:58 pm

MLC LLM (via) From MLC, the team that gave us Web LLM and Web Stable Diffusion. “MLC LLM is a universal solution that allows any language model to be deployed natively on a diverse set of hardware backends and native applications”. I installed their iPhone demo from TestFlight this morning and it does indeed provide an offline LLM that runs on my phone. It’s reasonably capable—the underlying model for the app is vicuna-v1-7b, a LLaMA derivative. # 29th April 2023, 5:43 pm

LLaVA: Large Language and Vision Assistant (via) Yet another multi-modal model combining a vision model (pre-trained CLIP ViT-L/14) and a LLaMA derivative model (Vicuna). The results I get from their demo are even more impressive than MiniGPT-4. Also includes a new training dataset, LLaVA-Instruct-150K, derived from GPT-4 and subject to the same warnings about the OpenAI terms of service. # 19th April 2023, 1:14 am

RedPajama, a project to create leading open-source models, starts by reproducing LLaMA training dataset of over 1.2 trillion tokens. With the amount of projects that have used LLaMA as a foundation model since its release two months ago—despite its non-commercial license—it’s clear that there is a strong desire for a fully openly licensed alternative.

RedPajama is a collaboration between Together, Ontocord.ai, ETH DS3Lab, Stanford CRFM, Hazy Research, and MILA Québec AI Institute aiming to build exactly that.

Step one is gathering the training data: the LLaMA paper described a 1.2 trillion token training set gathered from sources that included Wikipedia, Common Crawl, GitHub, arXiv, Stack Exchange and more.

RedPajama-Data-1T is an attempt at recreating that training set. It’s now available to download, as 2,084 separate multi-GB jsonl files—2.67TB total.

Even without a trained model, this is a hugely influential contribution to the world of open source LLMs. Any team looking to build their own LLaMA from scratch can now jump straight to the next stage, training the model. # 17th April 2023, 5:13 pm

Replacing my best friends with an LLM trained on 500,000 group chat messages (via) Izzy Miller used a 7 year long group text conversation with five friends from college to fine-tune LLaMA, such that it could simulate ongoing conversations. They started by extracting the messages from the iMessage SQLite database on their Mac, then generated a new training set from those messages and ran it using code from the Stanford Alpaca repository. This is genuinely one of the clearest explanations of the process of fine-tuning a model like this I’ve seen anywhere. # 12th April 2023, 11:01 pm

Downloading and converting the original models (Cerebras-GPT) (via) Georgi Gerganov added support for the Apache 2 licensed Cerebras-GPT language model to his ggml C++ inference library, as used by llama.cpp. # 31st March 2023, 4:28 am

gpt4all. Similar to Alpaca, here’s a project which takes the LLaMA base model and fine-tunes it on instruction examples generated by GPT-3—in this case, it’s 800,000 examples generated using the ChatGPT GPT 3.5 turbo model (Alpaca used 52,000 generated by regular GPT-3). This is currently the easiest way to get a LLaMA derived chatbot running on your own computer: the repo includes compiled binaries for running on M1/M2, Intel Mac, Windows and Linux and provides a link to download the 3.9GB 4-bit quantized model. # 29th March 2023, 6:03 pm

Announcing Open Flamingo (via) New from LAION: “OpenFlamingo is a framework that enables training and evaluation of large multimodal models (LMMs)”. Multimodal here means it can answer questions about images—their interactive demo includes tools for image captioning, animal recognition, counting objects and visual question answering. Theye’ve released the OpenFlamingo-9B model built on top of LLaMA 7B and CLIP ViT/L-14—the model checkpoint is a 5.24 GB download from Hugging Face, and is available under a non-commercial research license. # 28th March 2023, 9:59 pm

LLaMA voice chat, with Whisper and Siri TTS. llama.cpp author Georgi Gerganov has stitched together the LLaMA language model, the Whisper voice to text model (with his whisper.cpp library) and the macOS “say” command to create an entirely offline AI agent that he can talk to with his voice and that can speak replies straight back to him. # 27th March 2023, 9:06 pm

Hello Dolly: Democratizing the magic of ChatGPT with open models. A team at DataBricks applied the same fine-tuning data used by Stanford Alpaca against LLaMA to a much older model—EleutherAI’s GPT-J 6B, first released in May 2021. As with Alpaca, they found that instruction tuning took the raw model—which was extremely difficult to interact with—and turned it into something that felt a lot more like ChatGPT. It’s a shame they reused the license-encumbered 52,000 training samples from Alpaca, but I doubt it will be long before someone recreates a freely licensed alternative to that training set. # 24th March 2023, 5:05 pm

Fine-tune LLaMA to speak like Homer Simpson. Replicate spent 90 minutes fine-tuning LLaMA on 60,000 lines of dialog from the first 12 seasons of the Simpsons, and now it can do a good job of producing invented dialog from any of the characters from the series. This is a really interesting result: I’ve been skeptical about how much value can be had from fine-tuning large models on just a tiny amount of new data, assuming that the new data would be statistically irrelevant compared to the existing model. Clearly my mental model around this was incorrect. # 17th March 2023, 11:08 pm

Train and run Stanford Alpaca on your own machine. The team at Replicate managed to train their own copy of Stanford’s Alpaca—a fine-tuned version of LLaMA that can follow instructions like ChatGPT. Here they provide step-by-step instructions for recreating Alpaca yourself—running the training needs one or more A100s for a few hours, which you can rent through various cloud providers. # 16th March 2023, 4:10 pm

bloomz.cpp (via) Nouamane Tazi Adapted the llama.cpp project to run against the BLOOM family of language models, which were released in July 2022 and trained in France on 45 natural languages and 12 programming languages using the Jean Zay Public Supercomputer, provided by the French government and powered using mostly nuclear energy.

It’s under the RAIL license which allows (limited) commercial use, unlike LLaMA.

Nouamane reports getting 16 tokens/second from BLOOMZ-7B1 running on an M1 Pro laptop. # 16th March 2023, 12:24 am

Int-4 LLaMa is not enough—Int-3 and beyond (via) The Nolano team are experimenting with reducing the size of the LLaMA models even further than the 4bit quantization popularized by llama.cpp. # 13th March 2023, 11:55 pm

Running LLaMA 7B on a 64GB M2 MacBook Pro with llama.cpp. I got Facebook’s LLaMA 7B to run on my MacBook Pro using llama.cpp (a “port of Facebook’s LLaMA model in C/C++”) by Georgi Gerganov. It works! I’ve been hoping to run a GPT-3 class language model on my own hardware for ages, and now it’s possible to do exactly that. The model itself ends up being just 4GB after applying Georgi’s script to “quantize the model to 4-bits”. # 11th March 2023, 4:19 am

Introducing LLaMA: A foundational, 65-billion-parameter large language model (via) From the paper: “For instance, LLaMA-13B outperforms GPT-3 on most benchmarks, despite being 10× smaller. We believe that this model will help democratize the access and study of LLMs, since it can be run on a single GPU.” # 24th February 2023, 5:34 pm