Do Syntax Heads Influence Model Behavior?

William Timkey; Tal Linzen

cs.CLOct 13, 2026

Do Syntax Heads Influence Model Behavior?

Authors: William Timkey, Tal Linzen

Organizations: New York University

Abstract

Transformer language models develop attention heads that specialize in syntactic relations, for example, attending from verbs to their subjects. Do these heads play a causal role in the model's syntactic predictions? Evidence for such a role is mixed to negative: even when we force these heads to attend to a distractor noun other than the subject, the model's syntactic predictions are still more consistent with the subject than with the distractor. Interventions on specialized syntax heads had a significantly stronger effect on syntactic predictions than interventions on random heads, but only in one of the two measures we propose.

Figures & tables

Figure 1: Conceptual prediction for the causal influence of subject-verb attention heads on next-word predictions.

Figure 2: Top: 100k simulations estimating the relative probability,

\sigma

, assigned to the subject-consistent conditionation in the original model (right), and SV heads are forced to attend to only the distractor,

\sigma(x;m^{H^{sv}}_{d})

(left). A value greater than 0.5 (red) indicates a preference for the true subject, and a value less than 0.5 (blue) indicates a preference for the distractor-as-subject interpretation. Bottom: 100k simulations estimating the effect of SV heads on next-word prediction preference, relative to the baseline, measured over diagnostic pairs (left) and the full next-word distribution (right). Within each plot, positive values (blue) correspond to simulations where the intervention on the four SV heads had a larger effect on next word predictions than the same intervention on 4 random non-SV heads, while negative values (red) indicate the opposite.

Appendix figures & tables2 assets

Supplementary material from the paper’s appendix.

Appendix

Original Stimulus	Swap
The pumpkin near the child is {carved, upset}	pumpkin $\rightarrow$ stone
The plant by the gnome is {leafy, scary}	gnome $\rightarrow$ bear
The man with the crutch is {unhealthy, bent}	crutch $\rightarrow$ rod
The eraser beside the pencil is {round, sharp}	eraser $\rightarrow$ circle
The note for the proprietor is {illegible, happy}	proprietor $\rightarrow$ owner

Appendix

Table 1: One of the 48 items in our materials, adapted from Schlueter et al. (2019) . Participants in the original study read the sentence fragment, then selected one of the two adjectives to complete the sentence. The first adjective is consistent with the actual subject, woman , and the second adjective is consistent with the non-subject distractor noun, window(s) .

Stimulus ( $x_{\leq v}$ )	{ $c^{(s)},c^{(d)}$ }	$P(c^{s}\|x^{s})$	$P(c^{s}\|x^{d})$	$P(c^{s}\|x)$	$\sigma$
The woman by the window is…	{amused,dusty}	0.96	0.042	0.94	0.98
The flower by the bed is	{blooming,made}	0.11	0.000091	0.11	1.0
The notebook for the class is	{ripped,difficult}	0.14	0.017	0.12	0.81
The laptop on the magazine is	{overheating,lengthy}	0.84	0.018	0.81	0.96
The stone near the child is	{carved,upset}	0.99	0.015	0.99	0.99
The baby near the pillow is	{happy,fluffy}	0.93	0.33	0.86	0.93

Appendix

Table 2: All 48 items in our materials, adapted from Schlueter et al. (2019) . Along with various quantities of interest from the original model, including the relative probability of the subject-consistent continuation,

c^{(s)}

, given the subject-only reference sentence (e.g.

x^{(s)}=

The woman is… ); given the distractor-only baseline (e.g.

x^{(d)}=

The window is… ), and given the full sentence (e.g.

x=

The woman by the window is… ).

Transformer Interpretability

Transformer interpretability studies how attention, feed-forward layers, and hidden representations support model predictions and behavior.

See all 334 papers

Causal Drawbridges: Characterizing Gradient Blocking of Syntactic Islands in Transformer LMs
Apr 15, 2026Sasha Boguraev, Kyle Mahowald
The University of Texas at Austin
The Grammar of Transformers: A Systematic Review of Interpretability Research on Syntactic Knowledge in Language Models
Jan 9, 2026Nora Graichen, Iria de-Dios-Flores, Gemma BoledaTransformer Interpretability LLM Evaluation
Universitat Pompeu Fabra1 · ICREA2
Diverging Transformer Predictions for Human Sentence Processing: A Comprehensive Analysis of Agreement Attraction Effects
Mar 17, 2026Titus von der Malsburg, Sebastian PadóTransformer Cognitive Modeling
University of Stuttgart, Germany

Apr 15, 2026J/K move · Enter open · S save

Causal Drawbridges: Characterizing Gradient Blocking of Syntactic Islands in Transformer LMs

Sasha Boguraev, Kyle Mahowald

The University of Texas at Austin

Open paper AskPDF arXiv

We show how causal interventions in Transformer models provide insights into English syntax by focusing on a long-standing challenge for syntactic theory: syntactic islands. Extraction from coordinated verb phrases is often degraded, yet acceptability varies gradiently with lexical content (e.g., "I know what he hates art and loves" vs. "I know what he looked down and saw"). We show that modern Transformer language models replicate human judgments across this gradient. Using causal interventions that isolate functionally relevant subspaces in Transformer blocks, attention modules, and MLPs, we demonstrate that extraction from coordination islands engages the same filler-gap mechanisms as canonical wh-dependencies, but that these mechanisms are selectively blocked to varying degrees. By projecting a large corpus of unrelated text onto these causally identified subspaces, we derive a novel linguistic hypothesis: the conjunction "and" is represented differently in extractable versus non-extractable constructions, corresponding to expressions encoding relational dependencies versus purely conjunctive uses. These results illustrate how mechanistic interpretability can inform syntax, generating new hypotheses about linguistic representation and processing.

Abstract

Transformer Interpretability

Explore similar work

Causal Drawbridges: Characterizing Gradient Blocking of Syntactic Islands in Transformer LMs

The Grammar of Transformers: A Systematic Review of Interpretability Research on Syntactic Knowledge in Language Models

Diverging Transformer Predictions for Human Sentence Processing: A Comprehensive Analysis of Agreement Attraction Effects