arXiv:2607. 06621v1 Announce Type: new Abstract: The pre-softmax score of an attention head is a bilinear form $score(i,j) = x_i^T M x_j$ in a learned operator $M = W_q^T W_k$.
arXiv:2606. 02378v1 Announce Type: cross Abstract: We track the developmental trajectory of attention-head circuit formation across three 1B-class language models spanning two architecture families (dense transformer, mixture-of-experts) and two pretraining corpora (The Pile, DCLM): Pythia 1B, OLMo 1B-0724-hf, and OLMoE 1B-7B-0924.
arXiv:2607. 19771v1 Announce Type: cross Abstract: Muon and related matrix-sign optimizers are increasingly used to pre-train large language models, but their effect on the internal geometry of individual weight matrices is not well understood.
arXiv:2608. 02947v1 Announce Type: new Abstract: The attention score with rotary position embeddings (RoPE) decomposes exactly into a sum over its 2D-rotation frequency pairs, and each pair's wavelength limits how far it can discriminate position.
arXiv:2607. 11013v1 Announce Type: cross Abstract: Data re-uploading parameterized quantum circuits (DRU-PQCs) are universal function approximators, yet their expressivity produces oscillatory, non-convex loss landscapes that resist gradient-based optimization.
arXiv:2605. 24059v2 Announce Type: replace Abstract: We present a three-step recipe for identifying attention-head circuits in pretrained transformers.
arXiv:2606. 00926v1 Announce Type: new Abstract: Mechanistic studies of sequence models often treat layerwise state encodings as architectural traits: recurrent models concentrate readable state, attention-based models distribute it.
arXiv:2607. 19368v1 Announce Type: new Abstract: Long-prompt inference remains expensive because prefill attention scales quadratically with sequence length.
The attention score with rotary position embeddings (RoPE) decomposes exactly into a sum over its 2D-rotation frequency pairs, and each pair's wavelength limits how far it can discriminate position. Aligned with this structure, we propose the per-RoPE-wavelength distance window: it prunes the query--key inner-product terms beyond a wavelength-proportional distance.
arXiv:2607. 12735v1 Announce Type: new Abstract: Companion work showed the grokking delay is causally the time to form task-structured representations, injectable via a contrastive prior.
arXiv:2607. 27281v1 Announce Type: new Abstract: A capability appears in a language model when the last parts of its circuit align in one stochastic attempt, and getting all but one right is worth nothing.
arXiv:2605. 04893v2 Announce Type: replace Abstract: When a language model processes a hallucinated response, its attention routing tends to fail in one of two shapes: over-concentrating on a narrow set of positions, or spreading so diffusely that relevance is diluted, and the shape of the failure carries diagnostic signal.
arXiv:2608. 10172v1 Announce Type: new Abstract: Mechanistic interpretability explains models by identifying circuits inside them, but has no way to tell whether a circuit is a property of the model or an artifact of the method that found it.