Kuiz berkaitan panduan · Keras Tahap

Kuiz Vektor Fungsi dan Perwakilan Tugas

Kaji cara vektor fungsi diekstrak daripada kepala perhatian, disuntik ke dalam keadaan tersembunyi dan diuji merentas konteks tugas.

Laluan panduan yang berkaitanVektor Fungsi Dan Perwakilan Tugas
soalan 1 daripada 8

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

Pratonton setiap soalan8 soalan

Jawapan dan penjelasan muncul semasa anda mengambil kuiz.

  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. Bagaimanakah vektor fungsi biasanya dibina?
    • Dengan menjumlahkan semua benam perkataan dalam perbendaharaan kata
    • Dengan membuat purata output bagi pengepala perhatian pembawa tugas tertentu berbanding gesaan contoh
    • Dengan memperhalusi model pada satu contoh
    • Dengan membaca kebarangkalian softmax akhir
  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. Teknik analisis yang manakah menyokong pencarian kepala yang membawa vektor fungsi?
    • Penurunan kecerunan pada kehilangan
    • Carian rasuk
    • Pengaturan keciciran
    • Analisis pengantaraan sebab
  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
    • Perbendaharaan kata tokenizer
  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