Kuiz Penalaan Arahan

Jawapan dan penjelasan muncul semasa anda mengambil kuiz.

1. Tanpa penalaan arahan, bagaimanakah model asas mentah boleh bertindak balas terhadap gesaan 'Apakah ibu negara Perancis?'
   • Ia sentiasa enggan menjawab
   • Ia mengembalikan ralat
   • Ia mungkin diteruskan dengan lebih banyak soalan atau teks berkaitan dan bukannya menjawab
   • Ia menjawab dengan sempurna setiap masa
2. Penalaan arahan paling baik digambarkan sebagai:
   • Pembelajaran pengukuhan daripada maklum balas manusia
   • Pralatihan pada teks web mentah
   • Melaraskan parameter suhu
   • Penalaan halus diselia pada pasangan arahan-dan-tindak balas
3. How does supervised instruction tuning differ from reward-model-based RLHF in the published InstructGPT recipe?
   • It is exactly the same optimization stage
   • It trains against reference responses, while the later preference stages use comparisons and a reward model
   • It requires no training examples
   • It only changes a sampling temperature
4. Why do instruction-tuning researchers evaluate models on held-out tasks and instructions?
   • To assess generalization beyond the examples used for training
   • To prove that every unseen task will succeed
   • To avoid checking any generated responses
   • To guarantee the same accuracy for every task mixture
5. Yang manakah susunan peringkat latihan biasa untuk pembantu sembang moden?
   • RLHF, kemudian pralatihan, kemudian penalaan arahan
   • Penalaan arahan, kemudian pralatihan
   • Hanya pralatihan, tiada yang lain
   • Pralatihan, kemudian penalaan arahan, kemudian RLHF pilihan
6. What does LIMA's curated-data experiment support as a research conclusion?
   • Exactly 1,000 examples are enough for every model and task
   • More data can never help any instruction-tuned model
   • Carefully selected small datasets can be effective in an evaluated configuration
   • Only synthetic examples can train an assistant
7. Which training stage did the original LIMA experiment omit while using its 1,000 curated prompt-response pairs?
   • Reinforcement learning and human preference modeling
   • Supervised response loss
   • A pretrained base model
   • Prompt-response examples
8. A tuned model follows the requested answer format. What does that behavior alone fail to establish?
   • That an output format was requested
   • That every factual statement in the answer is correct
   • That the model produced a response
   • That the response can be inspected