Imbue

How can algorithms distribute agency and magnify curiosity? Glenn McDonald draws from his experience designing music algorithms at Spotify to examine how “systemically moral” algorithms can shift cultural validation from lotteries toward meritocracies. McDonald suggests that magnifying curiosity can be a means of decentralizing power: https://lnkd.in/g-h9Dc5m by Imbue

Great to see some really clear, practical, examples of how to use human judgement (sourced through Prolific) to finetune and improve model performance here: https://lnkd.in/eQZbJnDE

🤝 Imbue sourced vetted participants from Prolific to gain insight on question quality for their datasets. This data will help train their AI model on human-quality judgments. Check it out below or get started with Prolific for #AI here ▶️ https://lnkd.in/eCQKE3CE 🔗

Early this year, we trained a 70B model optimized for reasoning and coding. This model roughly matches LLAMA 3 70B despite being trained on 7x less data. Today, we’re releasing a toolkit to help others do the same. Read more and access the toolkit here:

Early this year, we trained a 70B model optimized for reasoning and coding. This model roughly matches LLAMA 3 70B despite being trained on 7x less data. Today, we’re releasing a toolkit to help others do the same, including: • 11 sanitized and extended NLP reasoning benchmarks including ARC, GSM8K, HellaSwag, and Social IQa • An original code-focused reasoning benchmark • A new dataset of 450,000 human judgments about ambiguity in NLP questions • A hyperparameter optimizer for scaling small experiments to a 70B run • Infrastructure scripts for bringing a cluster from bare metal to robust high-utilization training …and more! Read more and access the toolkit here: https://lnkd.in/gpzE6wwM Along with our tools, we’re sharing three blog posts with learnings from our training process: I. Conducting evaluations We found that our model and the best open-source models, when fine-tuned, outperform GPT-4o zero-shot across most multiple choice benchmarks. Surprisingly, both open and closed models achieve nearly 100% accuracy when evaluated only on unambiguous questions. We cleaned our evaluation datasets to isolate true failures of reasoning from failure due to ambiguous or low-quality questions. https://lnkd.in/giNS4z6h II. Setting up infrastructure Using our cluster for high performance training meant that every component — InfiniBand, Ethernet, GPUs, and the nodes themselves — had to work perfectly. If even a single one of the over 12,000 connections was a little flaky, it could slow down the entire training run. We're sharing open-source scripts and an end-to-end guide for infrastructure set-up that details the process of making everything work perfectly, and ensuring that it stays that way. https://lnkd.in/gCPDBknu III. Scaling experiments We successfully scaled from a 7B run to a 70B run on the first try, with minimal training instability and no loss spikes. We also predicted performance of the 70B model based on experiment results from much smaller models. We accomplished this using our hyperparameter optimizer, CARBS. We’re open-sourcing CARBS today so that other small teams experimenting with novel model architectures can experiment at small scale and trust performance at large scale. https://lnkd.in/gTgSjBvm This is one of many projects we’re working on to build collaborative agents that can reason and code. Other areas include RL, data generation, and experience design to make these powerful capabilities accessible and intuitive to users. We're hiring: https://imbue.com/careers/

"With AI, the default path is more centralization of power. This is why people are scared of the idea of 'AGI,' because AGI is power being centralized in a single entity. It takes a lot of effort and invention to buck that trend and actually make these systems democratize power. The real potential of agents, if we do a good job with invention, is that we can give individual people much more power over our computing environments." — Our CEO Kanjun Qiu at Collision Conf 2024, in conversation with Fast Company's Global Tech Editor Harry McCracken

"For us, pretraining, fine-tuning, doing RL — it's not just a black box. We want to open these things up and understand what's actually happening inside of there." Our CTO Josh Albrecht shared why and how we take a full-stack approach to developing agents in Practical AI's latest episode. Listen here: https://lnkd.in/dq7ArTCW If you're the type of person who is curious about all parts of the research and engineering process, from infrastructure to interfaces, consider joining our team: https://imbue.com/careers/

I'm thrilled to host a fireside chat on policy and safety in an age of AI with Matt Boulos, Head of Policy and Safety at Imbue! https://lnkd.in/gmiUvNZ4

🎙️ Generally Intelligent Episode 35: Percy Liang We sat down with Percy Liang, associate professor of computer science and statistics at Stanford University, to discuss: - how to evaluate language models robustly - balancing plurality and consensus with AI - the role of academia vs. industry in improving capabilities …and more! Podcast links and highlights: https://lnkd.in/gT9msybb

Can we trust AI to accurately inform us? Our study on the efficacy of using LLMs to analyze government submissions was cited in Seth Lazar's latest piece in The Guardian: https://lnkd.in/gmHUtWeV Read our study here: https://lnkd.in/g8vUMgiA