Kuis Penyematan Posisi Putar

Jawaban dan penjelasan muncul saat Anda mengikuti kuis.

1. Bagaimana RoPE memasukkan informasi posisi ke dalam transformator?
   • Dengan memutar pasangan dimensi dalam kueri dan vektor kunci dengan sudut yang bergantung pada posisi
   • Dengan menambahkan vektor tetap ke penyematan masukan
   • Dengan menyusun ulang token secara acak
   • Dengan menambahkan nomor posisi sebagai teks
2. Setelah RoPE diterapkan, skor perhatian antara dua token terutama bergantung pada:
   • Posisi absolut mereka saja
   • Jarak relatifnya (m dikurangi n)
   • Ukuran kosakata
   • Ukuran kumpulan
3. What training-efficiency comparison did the YaRN paper report for context extension?
   • Exactly the same data and steps as all prior methods
   • No fine-tuning or training data of any kind
   • About 10x fewer tokens and 2.5x fewer training steps than prior methods
   • A fixed 100x increase in output tokens
4. Mengapa pasangan dimensi berbeda berputar pada frekuensi berbeda di RoPE?
   • Untuk menghemat memori dengan melewatkan beberapa dimensi
   • Untuk mengacak keluaran model
   • Karena semua pasangan harus menggunakan sudut yang sama
   • Untuk menangkap tatanan lokal yang bagus dan posisi jangka panjang
5. What is the goal of techniques such as Position Interpolation and YaRN for RoPE-based models?
   • Extend the usable context length beyond the original training window using additional scaling/fine-tuning methods
   • Guarantee accurate recall at any sequence length with no evaluation
   • Remove position information from attention
   • Replace all token embeddings with absolute learned vectors
6. Does standard RoPE require learned absolute-position vectors to inject position into query and key representations?
   • Yes; it adds one learned vector for every possible sequence position
   • No; it applies position-dependent rotations to paired query/key dimensions
   • It removes query and key vectors entirely
   • It only adds position to value vectors
7. What does standard RoPE alone guarantee about a model’s context length?
   • Unlimited accurate context at every position
   • A fixed context of exactly 128 tokens
   • It does not by itself guarantee reliable use beyond the trained context length
   • Removal of the attention mechanism
8. What extended context length did the Position Interpolation paper report for its LLaMA evaluations?
   • Exactly 1,000 tokens with no position changes
   • Unlimited context with no fine-tuning
   • Up to 2 million tokens for all models
   • Up to 32,768 tokens with minimal fine-tuning