Kedis ku xidhan hagaha · Dhexdhexaad Heerka

Kediska Cilmi-baarista Nous

Sahami kaaliyeyaasha furan ee Cilmi-baarista Nous, hagaajinta door bidashada, hababka isgaarsiinta DeMo, iyo mashruuca tababbarka qaybinta-Psyche.

Wadooyinka haga ee la xidhiidhaCilmi-baarista Nous
Su'aal 1 ee 8

Waa maxay Cilmi-baarista Nous ee sida ugu wanaagsan loogu yaqaanno soo bixitaankeedii hore ee caannimada?

Horudhac su'aal kasta8 su'aalo

Jawaabaha iyo sharraxaadaha waxay soo baxayaan markaad qaadanayso kediska.

  1. Waa maxay Cilmi-baarista Nous ee sida ugu wanaagsan loogu yaqaanno soo bixitaankeedii hore ee caannimada?
    • Hagaajinta moodooyinka saldhigga furan ee taxanaha kaaliyaha Hermes
    • Soosaarka AI chips
    • Ku socodsiinta adeegga ciyaaraha daruuriga ah
    • Dhisidda moodooyinka soohdinta il-xiran ee xoqan
  2. Qaabkee ayaa leh Nous sida caadiga ah si fiican loo habeeyey?
    • Google's calaacasha iyo Gemini
    • Meta's Llama iyo moodooyinka Mistral
    • OpenAI's GPT-4
    • IBM Watson
  3. What problem does Nous Research’s DisTrO project address?
    • Soo saarida sawiro macquul ah
    • Cadaadiska feylasha muuqaalka
    • Yaraynta xadhkaha isgaadhsiinta ee loo baahan yahay tababarka la qaybiyey
    • Helitaanka iimaylada spamka ah
  4. Waa maxay shabakada cilmi-nafsiga?
    • Baraha bulshada
    • Halbeeg lagu cabbiro sababaynta
    • Nooc ka mid ah habka dareenka
    • Dadaal lagu doonayo in lagu tababaro moodooyinka agabka juqraafi ahaan loo qaybiyay ee internetka ku xidhan
  5. Farsamo noocee ah ayuu Nous caadi ahaan u adeegsadaa si uu u waafajiyo moodasheeda si fiican loo habeeyey?
    • Hagaajinta hagaajinta iyo hagaajinta doorbidida sida DPO ee xogta la qoondeeyey
    • Tababarka gebi ahaanba laga bilaabo miisaan aan toos ahayn oo aan lahayn nooc sal ah
    • Kaliya xeerarka gacanta-qorista oo aan lahayn waxbarasho
    • xoojinta barashada pixels cayriin
  6. What does the name DeMo stand for in Nous Research's distributed-training work?
    • Distributed Model Orchestration
    • Decoupled Momentum Optimization
    • Decoder-Mixture Operation
    • Deterministic Memory Offload
  7. According to the DeMo paper, which training state need not be fully synchronized between accelerators?
    • The text prompts used for evaluation
    • The vocabulary definitions in the tokenizer
    • The complete optimizer states and model parameters
    • The labels used to evaluate image classifiers
  8. Which three roles does Psyche's training system assign?
    • A tokenizer, a rendering engine, and a search index
    • An underwriter, an auditor, and a policyholder
    • A browser, an operating system, and a GPU driver
    • A coordinator, training clients, and a data provider