Fi agbara mu Olukọ ni adanwo Awọn awoṣe Ọkọọkan

Awọn idahun ati awọn alaye han bi o ṣe n mu ibeere naa.

1. Lakoko ipaniyan olukọ, kini ifunni bi titẹ sii ni igbesẹ iyipada kọọkan?
   • Ilẹ-otitọ ti tẹlẹ àmi
   • Awọn awoṣe ile ti ara ti tẹlẹ asọtẹlẹ
   • A ID àmi
   • Ik o wu àmi
2. During teacher-forced training, what target is paired with decoder input y_(t−1)?
   • The previous ground-truth token
   • The next target token y_t
   • A token sampled from the validation set
   • The maximum-probability output from an earlier training run
3. Aiṣedeede idanwo-irin-irin ti o ṣẹlẹ nipasẹ ipaniyan olukọ ni a mọ si:
   • Awọn gradients ti nparun
   • Ipo iṣubu
   • Iyasọtọ ifihan
   • Aami jijo
4. Ninu oluyipada Ayipada, ẹrọ wo ni o jẹ ki ikẹkọ fi agbara mu olukọ ṣiṣẹ ni afiwera kọja awọn ipo?
   • Jade kuro
   • Ipese deede
   • Iwadi ina
   • Iboju ifojusọna idi kan (wo iwaju).
5. Ni akoko itọkasi, kilode ti a ko le lo ipaniyan olukọ?
   • Ko si awọn ami-ilẹ-otitọ ọjọ iwaju lati jẹun
   • Iṣẹ pipadanu ko si
   • GPU ti lọra pupọ
   • Awọn iboju iparada ti wa ni alaabo
6. Ipadanu wo ni a maa n lo ni igbesẹ kọọkan lakoko ikẹkọ fi agbara mu olukọ?
   • Itumọ onigun mẹrin aṣiṣe lori aise logits
   • Cross-entropy laarin awọn ti anro pinpin ati otito tókàn àmi
   • Mitari pipadanu lori awọn ifibọ
   • Àdánù ibajọra Cosine
7. The target sequence is “red, blue, green.” Under teacher forcing, which token is supplied when training the prediction of “green”?
   • The model’s prediction for “green”
   • The future token after “green”
   • The ground-truth previous token “blue”
   • A randomly sampled vocabulary item
8. At decoder position t, which tokens must a causal attention mask block?
   • The source sequence
   • All earlier target tokens
   • The current position’s own input
   • Target tokens at positions later than t