Show HN: How I Topped the HuggingFace Open LLM Leaderboard on Two Gaming GPUs

Re:Show HN: How I Topped the HuggingFace Open LLM Leaderboar... by gwern on Tuesday March 10, 2026 @09:49PM (#47330960)

> They almost certainly have never seen regular conversations in Base64 in their training set, so its weird that it 'just works'.

People use Base64 to store payloads of many arbitrary things, including web pages or screenshots, both deliberately and erroneously, and so they have almost certainly seen regular conversations in Base64 in their 10tb+ text training sets scraped from billions of web pages and files and mangled emails etc.

[ Parent | on HN ]

Re:Show HN: How I Topped the HuggingFace Open LLM Leaderboar... by dnhkng on Wednesday March 11, 2026 @02:41AM (#47332333)

Yes, thats true.

But that points again to the main idea: The model has learnt to transform Base64 into a form it can already use in the 'regular' thinking structures.

The alternative is that there is an entire parallel structure just for Base64, which based on my 'chats' with LLMs in that format seems implausible; it acts like the regular model.

If there is a 'translation' organ in the model, why not a math or emotion processing organs? Thats what I set out to find, and are illustrated in the heatmaps.

Also, any writing tips from the Master blogger himself? Huge fan (squeal!)

[ Parent | on HN ]