Robust Intelligence

How many tokens does "hello world" use with a GPT tokenizer?

Fill in the scaled dot product attention formula.

Match each schema validation pattern to the production scenario where it fits best.

Reason about KV cache growth in a sliding window model pushed to 32k tokens

What does the temperature parameter do during LLM generation and how should you set it?

Match each instruction dataset to its on disk row shape.

Describe Anthropic's built in safety filtering layer

Each decode step loads ___ from HBM, regardless of how many tokens have already been generated

Match each vision-language fusion strategy to its defining property

Define test time compute and the accuracy tradeoff it enables.