There is now an open source alternative to ChatGPT, but good luck getting it working

There is now an open source alternative to ChatGPT, but good luck getting it working

The first open-source equivalent of OpenAI’s ChatGPT has arrived, but good luck getting it running on your laptop – or at all.

This week, Philip Wang, developer responsible for closed-source AI systems for reverse engineering, including Meta Make-A-Video, launched PaLM+RLHF, a text generation model that works similarly to ChatGPT. The system combines PaLM, a large language model from Google, and a technology called Reinforcement Learning with Human Feedback — RLHF, for short — to create a system that can accomplish pretty much any task ChatGPT can, including crafting emails and suggesting computer code.

But PaLM+RLHF is not pre-trained. This means that the system has not been trained on the sample data from the web that is needed to actually work. Downloading PaLM+RLHF won’t magically install a ChatGPT-like experience – it would require assembling gigabytes of text from which the model can learn and finding hardware powerful enough to handle the training workload.

Like ChatGPT, PaLM+RLHF is primarily a word prediction statistical tool. When fed a huge number of examples from the training data—for example, posts from Reddit, news articles, and e-books—PaLM+RLHF learns how likely words are to occur based on patterns such as the semantic context of the surrounding text.

ChatGPT and PaLM+RLHF share a special sauce in reinforcement learning with human feedback, a technology that aims to better align language models with what users want to achieve. RLHF involves training a language model – in the case of PaLM + RLHF, PaLM – and adjusting it to a data set that includes stimuli (eg, “explaining machine learning to a six-year-old”) combined with what human volunteers expect the model to say (eg, “Machine learning is a form of artificial intelligence…”). The above prompts are then fed to the exact model, which generates many responses, and volunteers rank all responses from best to worst. Finally, ratings are used to train a ‘reward model’ that takes the archetype’s responses and ranks them in order of preference, filtering out the best responses for a given prompt.

It is an expensive process to collect training data. And the training itself is not cheap. PaLM has a size of 540 billion parameters, “parameters” that refer to parts of the language model that have been learned from the training data. A 2020 study linked the expenditure of developing a text generation model with just 1.5 billion parameters to up to $1.6 million. To train the Bloom open source model, which has 176 billion parameters, it took three months using 384 Nvidia A100 GPUs; A single A100 costs thousands of dollars.

Running a PaLM + RLHF-sized trained model is also not easy. Bloom requires a dedicated PC with about eight A100 GPUs. Cloud alternatives are expensive, with back-end math finding the cost of running OpenAI to generate GPT-3 scripts — which contain about 175 billion parameters — on a single Amazon Web Services instance at about $87,000 per year.

Sebastian Raschka, an AI researcher, points out in a LinkedIn post about PaLM+RLHF that scaling the necessary development workflows can also be a challenge. “Even if someone gives you 500 GPUs to train that model, you still have to deal with the infrastructure and have a software framework that can handle that,” he said. “It’s obviously possible, but it’s a big effort right now (of course, we’re developing frameworks to make that simpler, but it’s still not trivial, yet).”

That’s all to say PaLM + RLHF will not replace ChatGPT today – unless a well-funded project (or person) encounters training and makes it available to the public.

In better news, several other efforts to replicate ChatGPT are progressing at a rapid clip, including one led by a research group called CarperAI. In partnership with open AI research organization EleutherAI and startups Scale AI and Hugging Face, CarperAI plans to release its first ChatGPT-like ready-to-run AI model trained on human feedback.

Lion, the nonprofit organization that supplied the raw dataset used for stable propagation training, is also leading a project to replicate ChatGPT using cutting-edge machine learning technology. LAION ambitiously aims to build the “assistant of the future” — one that not only writes emails and cover letters, but “does meaningful work, uses APIs, dynamically looks up information and so much more.” It is in its infancy. But a GitHub page with resources for the project was posted a few weeks ago.

#open #source #alternative #ChatGPT #good #luck #working

Leave a Comment

Your email address will not be published. Required fields are marked *