quiz bu lëkkale ak tegtal · Diggante Niveau

Laajteb tokenisation WordPiece

Xoolaat vocabulaire bu WordPiece ak sàrti segmentation yu bëgg xaalis, boole ci liggéeyam ci BERT lakk yu bari ak wuuteem ak BPE.

Yooni njiit yu jëm ciTokenisasioŋ Wordpiece
Laaj 1 de 8

Lan la prefix '##' bi di wane ci génnug WordPiece?

Xoolaatal laaj bu nekk8 laaj

Sooy def quiz bi, tontu yi ak leeral yi dañuy feeñ.

  1. Lan la prefix '##' bi di wane ci génnug WordPiece?
    • Token bi dafay wéyal baat bi, du baat bu njëkk
    • Token bi dafay tàqale lu melni [SEP]
    • Token bi moo gëna bari luñu koy gis ci vocabulaire bi
    • Token bi ab màrker la buñu xamul ci bitti vocabulaire bi
  2. Lu tollu ci ban yaatuwaayu vocabulaire WordPiece bi BERT bi njëkka jëfandikoo?
    • Lu tollu ci 256 jeton
    • Lu tollu ci 30000 jeton
    • Lu tollu ci 1 milioŋ ciy jeton
    • Lu tollu ci 500 jeton
  3. Lan moo wuutale xeetu boole WordPiece ak Kodage Byte-Pair buñ miin?
    • Dafay boole ñaar yi gëna néew ngir wàññi vocabulaire bi
    • Du musa boole ay araf kese
    • Dafay boole ci yokk done yiñ mëna tàggat, du frequence bu ñor
    • Dafay boole ñaar ci anam wu mucc ci diiru tàggat yaram
  4. What does a BERT-style WordPiece tokenizer return if it cannot segment a word into vocabulary pieces?
    • It always falls back to individual Unicode characters
    • A random subword selected from the vocabulary
    • The original word encoded as one new token
    • The unknown-token marker, such as [UNK]
  5. Ci jamonoy jël doggal, naka la WordPiece di tànnee ni ñuy xaajalee ab baat?
    • Dafay méngoo ak baat bi gëna gudd ci vocabulaire bi, cammoy dem ndeyjoor
    • Dafay njëkka tànn piyees yi gëna gàtt
    • Dafay jéem xaaj bu nekk ba noppi di leen moyenne
    • Dafay jëfandikoo ab reso neuronal ngir seetlu xaajaloo yi
  6. Ban famiy model bu siiw la WordPiece gëna boole?
    • Royuwaayi GPT
    • BERT ak ay derivatam
    • Royuwaayi LLaMA
    • Stable Diffusion
  7. What kind of vocabulary did multilingual BERT use across its training languages?
    • A completely separate vocabulary for every language
    • A vocabulary made only of complete words
    • One shared WordPiece vocabulary
    • A list of byte sequences with no subwords
  8. Which tokenizer class in Hugging Face Transformers constructs a BERT tokenizer that uses WordPiece?
    • GPT2Tokenizer
    • RobertaTokenizer
    • CLIPTokenizer
    • BertTokenizer