引導式測驗 · 硬 等級
DeepSeek V3 和 R1 推理測驗
測試 DeepSeek-V3 和 R1 的 MoE 架構、報告的訓練計算、RL 管道以及原始報告的精煉版本。
問題 1 的 8
DeepSeek-V3 技術報告給出了哪些參數數量?
預覽每個問題8 問題
當您參加測驗時,就會出現答案和解釋。
- DeepSeek-V3 技術報告給出了哪些參數數量?
- 671B total parameters, with 37B activated per token
- 37B total parameters, all active per token
- 671B active parameters for every token
- A dense model with no routed experts
- DeepSeek-R1 的推理訓練與發布有何值得注意之處?
- It used no training data
- The report describes RL-based reasoning and publicly released model weights
- It was only a closed cloud API
- It was a rule-based program with no learned model
- DeepSeek-V3 報告針對其完整訓練運行給出了什麼計算數據?
- 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
- 什麼訓練訊號幫助 DeepSeek-R1 發展出推理行為?
- Only manually written rules with no learned model
- Image-label classification
- Random rewards unrelated to answers
- Reinforcement learning rewards for correct, verifiable outcomes
- 與 DeepSeek-V3 訓練相關的一種效率技術是什麼?
- 多頭潛在註意力和輔助無損耗負載平衡
- 將模型列印在紙上
- 刪除所有註意力層
- 對單一範例進行訓練
- Which types of tasks did the DeepSeek-R1 paper evaluate?
- 僅影像分類
- Mathematics, coding competitions, and STEM tasks
- Only speech transcription
- Only spreadsheet formatting
- 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
- 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