Norway’s 2 Petabytes of Huawei Flash Storage and LLM Training: A Consensus Report

Executive TL;DR:

• Norway is using 2 petabytes of Huawei flash storage for LLM training.
• The goal is to build a sovereign LLM that reflects Norwegian language and culture.
• Experts are skeptical about the project’s feasibility and usefulness.

The Internet’s Verdict: 60% Skeptical, 40% Supportive

Introduction to LLM Training

Norway’s national library has a user-friendly interface for searching through texts, but the country’s LLM training efforts are raising concerns.

Forum Voices

Some experts question the project’s hardware and goals.

I’m a Norwegian, and I use the national library almost every day for searching through texts. They have truly one of the best working user interfaces (and functionality) for searching through the massive amounts of text.

Training a sovereign LLM with this meager hardware as opposed to a LORA on some open source model seems like a huge mistake and a potential red flag.

Alternative Solutions

Instead of building a sovereign LLM, some suggest creating a set of training data and sharing it with model builders.

I wonder if instead (or in parallel), Norway should build a set of training data and share it (for free) with all the model builders.

Conclusion

Norway’s LLM training efforts are ambitious, but experts are skeptical about their feasibility and usefulness.

Norway is a small country solving a problem every non-English-speaking nation will face: how do you build AI that reflects your language, your culture and your history?

Focus Keyword: Norway LLM