Unverified50% confidenceOpinionExact time
SAEs trained with different seeds may learn feature dictionaries that are semantically similar but differ in specific direction or magnitude, leading to differences in the structure of attribution graphs.
1
Sources
50%
Confidence
Long-term
Relevance
11/3/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedEven with identical input text, different random seeds will produce visually distinct results; some tools offer a fixed seed option to reduce variation under the same composition.65% similarVerifiedSparse autoencoders (SAEs) are one of the main technical approaches attempting to disentangle superposition and extract monosemantic features.63% similarUnverified稀疏自编码器（SAE）无需人工指定目标概念，能自动发现模型表征的数万乃至上百万个语义特征63% similarUnverified当差异来源于后端（如不同引擎的基础模型不同导致格式差异）时，应用提示词模板加插槽处理——模板负责任务本身，插槽负责下游多样化差异61% similarUnverifiedVisualization of opinion evolution can be achieved through sentence embeddings (such as Sentence-BERT) and cosine similarity, but simple similarity cannot distinguish a stance flip from a rewording.61% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1129405API
curl https://kongchang.com/api/v1/knowledge/claims/1129405MCP
get_claim(id=1129405)