Mapping 1,000+ Language Models via the Log-Likelihood Vector
This provides a new tool for researchers and practitioners to analyze and compare large-scale language models efficiently, though it is incremental as it builds on existing theoretical concepts.
The paper tackled the problem of comparing autoregressive language models at scale by proposing log-likelihood vectors as model features, resulting in a scalable method applied to over 1,000 models to construct a 'model map' for analysis.
To compare autoregressive language models at scale, we propose using log-likelihood vectors computed on a predefined text set as model features. This approach has a solid theoretical basis: when treated as model coordinates, their squared Euclidean distance approximates the Kullback-Leibler divergence of text-generation probabilities. Our method is highly scalable, with computational cost growing linearly in both the number of models and text samples, and is easy to implement as the required features are derived from cross-entropy loss. Applying this method to over 1,000 language models, we constructed a "model map," providing a new perspective on large-scale model analysis.