模型用异常维度偏好高频词，是有效预测策略而非缺陷。
Not a nuisance but a useful heuristic: Outlier dimensions favor frequent tokens in language models
- 利用最后一层异常维度提升高频词预测效率
- 模型可通过权重分配抑制不当使用该机制
- 适合研究语言模型内部机制的读者
我们研究了最后一层的异常维度——即对多数输入呈现极端激活的维度。发现这类维度在多种现代语言模型中普遍存在，并追溯其功能源于持续预测高频词的启发式策略。进一步表明，当该策略不适用时，模型可通过将权重质量分配给其他维度来抑制此行为。我们还探究了哪些模型参数会增强异常维度，以及它们在训练过程中的出现时机。结论表明，异常维度是一种被多个不同模型独立发现的专用机制，用于实现有效的标记预测启发式方法。
原文摘要 · Abstract (English)
We study last-layer outlier dimensions, i.e. dimensions that display extreme activations for the majority of inputs. We show that outlier dimensions arise in many different modern language models, and trace their function back to the heuristic of constantly predicting frequent words. We further show how a model can block this heuristic when it is not contextually appropriate, by assigning a counterbalancing weight mass to the remaining dimensions, and we investigate which model parameters boost outlier dimensions and when they arise during training. We conclude that outlier dimensions are a specialized mechanism discovered by many distinct models to implement a useful token prediction heuristic.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。