Description
「大模型的“魔力”之源」是一个6集的系列，一起探索大模型的强大之谜！本期要点： Softmax函数将神经网络的输出转化为概率分布,是分类任务的核心 在大模型中,Softmax被创新性地应用于文本生成任务 通过Softmax,大模型可以基于上下文生成下一个最可能的单词,从而实现了文本生成 采样策略的选择对生成质量有很大影响,如温度参数、Top-k采样等 结束语:Softmax的魔力在于将离散的分类扩展到了连续的生成,为大模型开启了全新的应用空间
Audio
Featured in this Episode
No persons identified in this episode.
Transcription
This episode hasn't been transcribed yet
Help us prioritize this episode for transcription by upvoting it.
0
upvotes
Popular episodes get transcribed faster
Other recent transcribed episodes
Transcribed and ready to explore now
SpaceX Said to Pursue 2026 IPO
10 Dec 2025
Bloomberg Tech
Don’t Call It a Comeback
10 Dec 2025
Motley Fool Money
Japan Claims AGI, Pentagon Adopts Gemini, and MIT Designs New Medicines
10 Dec 2025
The Daily AI Show
Eric Larsen on the emergence and potential of AI in healthcare
10 Dec 2025
McKinsey on Healthcare
What it will take for AI to scale (energy, compute, talent)
10 Dec 2025
Azeem Azhar's Exponential View
Reducing Burnout and Boosting Revenue in ASCs
10 Dec 2025
Becker’s Healthcare -- Spine and Orthopedic Podcast