用简单损失函数让大模型自动生成可控制的解释性特征。
Efficient Representations are Controllable Representations
- 通过辅助损失训练16个激活维度作为可解释标志位。
- 这些标志位在推理时能直接控制生成内容，准确率提升显著。
- 适合需要可控生成的AI应用开发者快速上手。
如何最直接地在大模型激活中植入可解释、可控制的特征？传统方法需复杂识别与干预现有特征几何。本文提出新思路：仅通过一个简单辅助损失微调大模型，使其中16个（共3072个）残差流维度成为惰性可解释标志位，仅指示生成所需的概念。模型在训练中自发重组，学习依赖这些标志位进行实际生成。结果，这些标志位演变为真正的内部特征——可解释的控制开关，可在推理时引导生成。为何有效？当某一特征在固定位置稳定提供时，梯度下降会逐步消除其他冗余编码，模型自我消解替代表示。模型的效率压力正是杠杆，可被利用以诱导出可解释且可控制的表征。
原文摘要 · Abstract (English)
What is the most brute-force way to install interpretable, controllable features into a model's activations? Controlling how LLMs internally represent concepts typically requires sophisticated methods to first identify, then intervene on the model's existing feature geometry. We bypass all of this. We finetune an LLM with a simple auxiliary loss, training 16 of its 3072 residual stream dimensions to be inert interpretability flags that simply indicate what concepts are required for generation. The model reorganizes around them anyway, learning to rely on these flags during actual generation tasks. As a result, these inert flags become genuine internal features: interpretable control switches that allow us to steer generation at inference time. Why does this work? When a feature is reliably supplied at a fixed location, gradient descent gradually eliminates redundant encodings elsewhere, and the model erodes its own alternative representations. A model's efficiency pressure is a lever - exploitable to induce interpretable, controllable representations.
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。