implementation

DeepSeek’s distillation techniques let R1’s capabilities trickle down into smaller, more budget-friendly versions of the model. You can even run a distilled variant locally on your MacBook Pro with just one line of code.

The important thing is that you practice. The more you practice, the better you will become.

takes years of practice and effort to get good at identifying blind spots in ourselves and then expressing the affected emotions appropriately.

One of the themes I emphasized was the idea of protocol money. People who live and breathe technology tend to think in terms of rapid disruption, and assume that new things will displace old things. But a point I emphasized is that while that’s a true trend for many things, it hasn’t historically applied to communication protocols. Things like

even when that means performing poorly on tests of immediate progress. That is, the most effective learning looks inefficient; it looks like falling behind.

Every model, whether LLM or ML, will operate best if it is focused and specializes in specific conclusions it is trying to make. For example, I will make ChatGPT bots for very specific parts of my research, reading, and trading processes. Create specialized bots and models, then structure them hierarchically. The goal is to run a suite of bots,

Guessing at the contents of a person’s attic from his outward appearance becomes one of Sherlock’s surest ways of determining who that person is and what he is capable of.

And you will run through career change scenarios and try to play out the implications to their logical, full conclusion. That, too, is…

The notes are no longer reminders of thoughts or ideas, but contain the actual thought or idea in written form. This is a crucial difference.