Guide-yakabatana mibvunzo · Zvakaoma Nhanho

DeepSeek V3 uye R1 Kukurukurirana Quiz

Edza DeepSeek-V3 uye R1's MoE dhizaini, yakashumwa yekudzidzira komputa, RL pombi, uye yakaburitswa yakaburitswa kubva kumishumo yekutanga.

Nzira dzekutungamira dzinoenderanaDeepseek V3 Uye R1 Kukurukurirana
Mubvunzo 1 ye 8

Ndeipi parameter inoverengera iyo DeepSeek-V3 technical report yakapa?

Tarisa mubvunzo wese8 mibvunzo

Mhinduro uye tsananguro zvinoonekwa paunenge uchitora bvunzo.

  1. Ndeipi parameter inoverengera iyo DeepSeek-V3 technical report yakapa?
    • 671B yakazara paramita, ine 37B yakabatidzwa pane tokeni
    • 37B yakazara ma paramita, ese anoshanda pane tokeni
    • 671B inoshanda paramita kune yega chiratidzo
    • Iyo dense modhi isina nyanzvi dzinofambiswa
  2. Chii chaizivikanwa nezve DeepSeek-R1's kufunga kudzidziswa uye kusunungurwa?
    • Haina kushandisa data yekudzidziswa
    • Chirevo chinotsanangura RL-yakavakirwa kufunga uye yakaburitswa pachena modhi uremu
    • Yaingove yakavharwa gore API
    • Yaiva purogiramu-yakavakirwa purogiramu isina muenzaniso wakadzidza
  3. Ndeipi nhamba yekombuta iyo DeepSeek-V3 rondedzero yakapa kune yayo yakazara kudzidziswa kumhanya?
    • 2.788 thousand CPU-hours
    • 2.788 million H100 GPU-hours for R1 only
    • 2.788 million H800 GPU-hours
    • No accelerator-hours were reported
  4. What training signal helped DeepSeek-R1 develop reasoning behaviors?
    • Only manually written rules with no learned model
    • Image-label classification
    • Random rewards unrelated to answers
    • Reinforcement learning rewards for correct, verifiable outcomes
  5. Ndeipi imwe nzira inoshanda yakabatana nekudzidziswa kweDeepSeek-V3?
    • Multi-head latent kutarisisa uye yebetsero-kurasikirwa-yemahara mutoro kuenzanisa
    • Kudhinda muenzaniso papepa
    • Kubvisa zvese zvekutarisisa
    • Kudzidzira pane imwe muenzaniso
  6. Which types of tasks did the DeepSeek-R1 paper evaluate?
    • Kuiswa kwemifananidzo chete
    • Mathematics, coding competitions, and STEM tasks
    • Only speech transcription
    • Only spreadsheet formatting
  7. What supervised-learning step does the DeepSeek-R1 pipeline use before its first RL stage?
    • A period of training with no text or prompts
    • A final model-compression step only
    • Cold-start supervised fine-tuning on curated reasoning examples
    • A computer-vision pretraining stage
  8. Which model families were used as bases for the released R1-distilled models?
    • BERT and CLIP
    • Tacotron and WavLM
    • NeRF and Instant-NGP
    • Qwen and Llama