Day 1

Day 1 of 1% better.

Today, I actually tried to learn what goes behind LLMs.

A few things which I learned

1. These models aren’t thinking or searching they’re predicting the next token based on patterns.

2. Tokens aren’t just words. Even punctuation and parts of words matter more than I realized.

3. Training vs inference finally clicked. Training is where the model learns patterns, inference is just using those patterns.

4. Prompting matters small wording changes can completely shift the output.

I’m not sure how useful this will be to others, but I’m starting as a way to document my journey and build in public.

It starts today.

🔗 Resources - https://lnkd.in/g-8NxdkX