paper-with-me

홈 › Papers

Scaling Laws for State Dynamics in Large Language Models

2025-05-20 · Jacob X Li, Shreyas S Raman, Jessica Wan, Fahad Samman, Jazlyn Lin

Large Language Models (LLMs) are increasingly used in tasks requiring internal state tracking, yet their ability to model state transition dynamics remains poorly understood. We evaluate how well LLMs capture deterministic state dynamics across 3 domains: Box Tracking, Abstract DFA Sequences, and Complex Text Games, each formalizable as a finite-state system. Across tasks, we find that next-state prediction accuracy degrades with increasing state-space size and sparse transitions. GPT-2 XL reaches about 70% accuracy in low-complexity settings but drops below 30% when the number of boxes or states exceeds 5 or 10, respectively. In DFA tasks, Pythia-1B fails to exceed 50% accuracy when the number of states is > 10 and transitions are < 30. Through activation patching, we identify attention heads responsible for propagating state information: GPT-2 XL Layer 22 Head 20, and Pythia-1B Heads at Layers 10, 11, 12, and 14. While these heads successfully move relevant state features, action information is not reliably routed to the final token, indicating weak joint state-action reasoning. Our results suggest that state tracking in LLMs emerges from distributed interactions of next-token heads rather than explicit symbolic computation.

📄 PDF Abstract BibTeX arXiv:2505.14892

Code (0)

등록된 구현이 없습니다.

Methods 이 논문이 사용한 방법론

Refunds@Expedia|||How do I get a full refund from Expedia? “How do I get a full refund from Expedia? How do I get a full refund from Expedia? – Call ☎️ +1-(888) 829 (0881) or +1-805-330-4056 or +1-805-330-4056 for Quick Help &…
Attention 설명 없음
Cosine Annealing Cosine Annealing is a type of learning rate schedule that has the effect of starting with a large learning rate that is relatively rapidly decreased to a minimum value before…
Linear Warmup With Cosine Annealing Linear Warmup With Cosine Annealing is a learning rate schedule where we increase the learning rate linearly for $n$ updates and then anneal according to a cosine schedule…
Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…
Attention Dropout Attention Dropout is a type of dropout used in attention-based architectures, where elements are randomly dropped out of the…
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Residual Connection 설명 없음

Similar Papers 제목 키워드 기반

Scaling Laws for Differentially Private Language Models

2025-01-31 · Ryan McKenna, Yangsibo Huang, Amer Sinha, Borja Balle 외

Scaling laws have emerged as important components of large language model (LLM) training as they can predict performance gains through scale, and provide guidance on important hyper-parameter choices that would otherwise…

Language ModelingLanguage ModellingLarge Language Model

Neural Thermodynamic Laws for Large Language Model Training

2025-05-15 · Ziming Liu, Yizhou Liu, Jeff Gore, Max Tegmark

Beyond neural scaling laws, little is known about the laws underlying large language models (LLMs). We introduce Neural Thermodynamic Laws (NTL) -- a new framework that offers fresh insights into LLM training dynamics. O…

Language ModelingLanguage ModellingLarge Language Model

ATLAS: Adaptive Transfer Scaling Laws for Multilingual Pretraining, Finetuning, and Decoding the Curse of Multilinguality

2025-10-24 · Shayne Longpre, Sneha Kudugunta, Niklas Muennighoff, I-Hung Hsu 외 arxiv

Scaling laws research has focused overwhelmingly on English -- yet the most prominent AI models explicitly serve billions of international users. In this work, we undertake the largest multilingual scaling laws study to …

Cross-Lingual Transfer

Generalizing Scaling Laws for Dense and Sparse Large Language Models

2025-08-08 · Md Arafat Hossain, Xingfu Wu, Valerie Taylor, Ali Jannesari arxiv

Despite recent advancements of large language models (LLMs), optimally predicting the model size for LLM pretraining or allocating optimal resources still remains a challenge. Several efforts have addressed the challenge…

Towards joint scaling laws with optimal batch size schedules

2026-07-30 · Jiaxiang Li, Zhiqi Bu, Shiyun Xu arxiv

Modern deep learning typically keeps the batch size static throughout training, thus overlooking the joint effect of learning rate and batch size on the training dynamics. In this paper, we study the deep learning dynami…