Flashcard

Your input is 9,000 tokens but max_length is 8,192. What are your truncation options?

You open a HuggingFace model repo and see tokenizer_config.json. What is in it and what is NOT?

Token 1234 means 'the' in one model and garbage in another. Why?

An LLM never sees raw text. So what does it actually process?

BOS, EOS, PAD: why do tokenizers reserve these ids instead of using normal text?

Someone confuses 'sequence length' with 'context window'. How do you correct them?

You change one word in the system prompt and prompt caching breaks. Why?

Pre-tokenization splits text before BPE merges. Why is that boundary important?

You pad a batch to equal length. How does the model know to ignore the filler tokens?

Older tokenizers had an OOV problem. Why did byte-level BPE kill it?

Before the tokenizer even starts splitting, normalization runs. What does it do and why?

BPE learned 50,000 merge rules during training. What is a merge rule, and where does it live at inference?

You hear 'fertility is 2.5 for Hindi on cl100k_base'. What does that number mean and is it good?

A tokenizer is 'byte-level'. What does that buy you over a character-level one?

Your teammate mentions BPE at a code review. What are they talking about?

A chat template wraps every turn with BOS and EOS tokens. Why does the model need them?

BPE grows the vocabulary by adding merges. Unigram does the opposite. Explain the difference.

Flashcard: what is an agent 'trajectory' and why does it matter for evaluation?

What does one agent trace capture step by step?

Flashcard: why use Tree of Thoughts search on hard agent planning problems?

Flashcard: why break a complex goal into subtasks before acting?

Flashcard: why must every agent loop declare when to stop?

Where does an agent keep short-term state during a single task?

Flashcard: how is self-correction different from self-refine in agents?

Why run agent-generated code inside an isolated sandbox?