Maswali yanayounganishwa na mwongozo · Ngumu Kiwango

Maswali ya Vekta za Kazi na Uwakilishi wa Kazi

Soma jinsi vekta za utendakazi zinavyotolewa kutoka kwa vichwa vya umakini, hudungwa katika hali fiche, na kujaribiwa katika miktadha ya kazi.

Njia za mwongozo zinazohusianaVekta za Kazi na Uwakilishi wa Kazi
Swali 1 ya 8

What does a function vector represent in the studied language models?

Hakiki kila swali8 maswali

Majibu na maelezo huonekana unapojibu maswali.

  1. What does a function vector represent in the studied language models?
    • A compact activation-space representation associated with a demonstrated task
    • The model’s full weight matrix
    • A single token embedding in every setting
    • The optimizer’s learning rate
  2. Je, vekta ya utendaji hujengwaje kwa kawaida?
    • Kwa muhtasari wa upachikaji wa maneno yote katika msamiati
    • Kwa wastani wa matokeo ya vichwa maalum vya kubeba jukumu juu ya vidokezo vya mfano
    • Kwa kurekebisha mfano kwenye mfano mmoja
    • Kwa kusoma uwezekano wa mwisho wa softmax
  3. What did the function-vector paper observe after injecting a vector into a zero-shot prompt?
    • The model always crashed
    • It guaranteed correct answers on any task
    • It could trigger the associated task in the tested settings
    • It permanently changed the model weights
  4. Ni mbinu gani ya uchanganuzi inayosisitiza kupata vichwa vinavyobeba vekta ya utendaji?
    • Kushuka kwa gradient juu ya hasara
    • Utafutaji wa boriti
    • Udhibiti wa kuacha shule
    • Uchambuzi wa upatanishi wa sababu
  5. What do function-vector experiments provide evidence for?
    • Some models form reusable internal representations associated with tasks
    • All language models only memorize text
    • Every model has one universal task vector
    • Tokenization alone determines every behavior
  6. Why are function vectors interesting to steering research?
    • They guarantee safe or correct behavior
    • They provide an activation-level intervention that can influence task behavior in tested settings
    • They eliminate prompts for every application
    • They make all model decisions fully transparent
  7. At what part of the transformer did the paper report strong function-vector effects?
    • Only the input embedding table
    • Only the final output logits
    • Middle layers in the tested models
    • Msamiati wa ishara
  8. What happened when researchers summed some function vectors?
    • The vectors always canceled to zero
    • The model’s weights were permanently updated
    • Every possible task was composed perfectly
    • Some combinations triggered new complex tasks in the tested experiments