Тест, свързан с ръководство · Среден Ниво

Anthropic Claude Тест за нива на опус и сонет

Прегледайте ролите на ниво хайку, сонет и опус на Claude, разширеното мислене и процеса на обратна връзка на Конституционалния AI във всички издания.

Свързани ръководни пътекиAnthropic Claude Нива на опус и сонет
Въпрос 1 на 8

What naming theme do Haiku, Sonnet, and Opus draw from?

Визуализирайте всеки въпрос8 въпроси

Отговорите и обясненията се появяват, докато отговаряте на теста.

  1. What naming theme do Haiku, Sonnet, and Opus draw from?
    • A mix of literary and musical terms
    • Greek letters only
    • Планети
    • Цветове
  2. Within Anthropic’s Haiku, Sonnet, and Opus tiers, how is Opus positioned?
    • It is the fastest option in every model family
    • It is positioned for the most demanding work among Haiku, Sonnet, and Opus
    • It is a speech-only model
    • It is Anthropic’s only model family
  3. What approach did Anthropic’s 2022 Constitutional AI paper describe for training a harmless assistant?
    • Training only on image labels
    • Removing the model’s language data
    • Using written principles to guide self-critique and AI-generated feedback
    • Using no evaluation or feedback at all
  4. How does Anthropic position Sonnet among the Claude tiers?
    • As a model for image generation only
    • As a retired safety policy
    • As a slower, most expensive tier in every release
    • As a balance of speed and intelligence
  5. What does Claude’s extended-thinking mode do when supported?
    • Spends additional computation reasoning through a complex task before answering
    • Responds instantly without computation
    • Изключва функциите за безопасност
    • Изтрива историята на разговорите
  6. Which Claude tier does Anthropic describe as its fastest option?
    • Опус
    • Хайку
    • Сонет
    • Fable
  7. What follows the self-critique stage in the Constitutional AI paper’s AI-feedback approach?
    • The model stops after critique with no further training
    • The tokenizer is replaced with an image encoder
    • AI-generated preference comparisons train a reward model used for reinforcement learning
    • The model is trained only with manually ranked human comparisons
  8. Which non-text input modality can Claude models support according to Anthropic’s model documentation?
    • Lidar point clouds for autonomous driving
    • MIDI files as native control commands
    • Robot joint torques
    • Images