← Back to all news
arXiv AI By Arjun Pillai, Christian Hoang, Anjelo Laroza

FTB Graph: Determining and Validating First-token Broadcasters and Language-Identity Head Circuits in Multilingual Language Models

Read the original on arXiv AI →

The paper introduces FTB Graph, a method for mapping the causal circuitry that determines the first-token language identity in multilingual language models. Using Edge Attribution Patching and exact activation patching across six architectures (GPT‑2, BLOOM‑560M, Pythia‑1B/2.8B, Qwen2.5‑1.5B Base/Instruct), the authors extract directed acyclic graphs that reveal deep or mid‑to‑deep broadcasting hubs, with notable differences among models. The study finds that first‑token routing is largely established during pretraining and largely preserved by instruction tuning, while linear gradient approximations can diverge from causal interventions, underscoring the need for exact‑patching verification.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv AI.

Related stories

arXiv Computation and Language
Aug 28

Many Circuits, One Mechanism: Input Variation and Evaluation Granularity in Circuit Discovery

The paper investigates whether structural differences in circuits discovered by circuit discovery methods reflect distinct mechanisms. By varying input-token frequency while keeping the task constant, the authors find that although circuits appear specialized by frequency structurally, functional and representational analyses reveal no reliable differences, a phenomenon they call phantom specialization. Across multiple models and tasks, structurally distinct circuits implement the same computation, with core shared subgraphs recovering most of the performance and interchangeable internal representations confirmed by causal interventions.

By Alireza Bayat Makou, Jingcheng Niu, Subhabrata Dutta, Iryna Gurevych
More like this →
arXiv AI
Aug 20

First-Token Broadcasters: Mechanistic Origins of Language Identity and Distributed Robustness in Transformers

The paper introduces Language Identity Head Ablation (LIHA), a causal method that zeroes individual attention heads in transformer models to measure language switch rates across multilingual prompts. Applying LIHA to GPT‑2 reveals a small set of first‑token broadcaster heads—most notably L6H1—that persistently attend to the initial prompt token and propagate language signals throughout generation, with compensatory head recruitment occurring hierarchically in higher layers. A controlled comparison between Qwen2.5‑1.5B‑Base and Qwen2.5‑1.5B‑Instruct shows that instruction tuning concentrates language‑identity influence in early layers, while experiments with Chinese and Russian confirm script‑specific first‑token broadcasting at layer 0.

By Arjun Pillai, Christian Hoang, Anjelo Jann Laroza
llmsdiffusionfine-tuningbenchmarks
More like this →