Image de couverture de livre

LLM Circuit Analyses Are Consistent Across Training and Scale

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Tigges, Curt, Hanna, Michael, Yu, Qinan, Biderman, Stella
Format:	Preprint
Publié:	2024
Sujets:	Machine Learning Computation and Language
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

Transformer-Based Models Are Not Yet Perfect At Learning to Emulate Structural Recursion
par: Zhang, Dylan, et autres
Publié: (2024)

PolyPythias: Stability and Outliers across Fifty Language Model Pre-Training Runs
par: van der Wal, Oskar, et autres
Publié: (2025)

Grokking Group Multiplication with Cosets
par: Stander, Dashiell, et autres
Publié: (2023)

LEACE: Perfect linear concept erasure in closed form
par: Belrose, Nora, et autres
Publié: (2023)

Why Has Predicting Downstream Capabilities of Frontier AI Models with Scale Remained Elusive?
par: Schaeffer, Rylan, et autres
Publié: (2024)

Zero-shot Factual Consistency Evaluation Across Domains
par: Agarwal, Raunak
Publié: (2024)

Circuit Component Reuse Across Tasks in Transformer Language Models
par: Merullo, Jack, et autres
Publié: (2023)

How Far Can Unsupervised RLVR Scale LLM Training?
par: He, Bingxiang, et autres
Publié: (2026)

Have Faith in Faithfulness: Going Beyond Circuit Overlap When Finding Model Mechanisms
par: Hanna, Michael, et autres
Publié: (2024)

Transcoders Find Interpretable LLM Feature Circuits
par: Dunefsky, Jacob, et autres
Publié: (2024)

SAEBench: A Comprehensive Benchmark for Sparse Autoencoders in Language Model Interpretability
par: Karvonen, Adam, et autres
Publié: (2025)

Scaling with Collapse: Efficient and Predictable Training of LLM Families
par: Bergsma, Shane, et autres
Publié: (2025)

WISCA: A Lightweight Model Transition Method to Improve LLM Training via Weight Scaling
par: Li, Jiacheng, et autres
Publié: (2025)

Representation Consistency for Accurate and Coherent LLM Answer Aggregation
par: Jiang, Junqi, et autres
Publié: (2025)

Quantifying the Effect of Test Set Contamination on Generative Evaluations
par: Schaeffer, Rylan, et autres
Publié: (2026)

Training-Inference Consistent Segmented Execution for Long-Context LLMs
par: Shang, Xianpeng, et autres
Publié: (2026)

ConsistRM: Improving Generative Reward Models via Consistency-Aware Self-Training
par: Liang, Yu, et autres
Publié: (2026)

DFPO: Scaling Value Modeling via Distributional Flow towards Robust and Generalizable LLM Post-Training
par: Zhu, Dingwei, et autres
Publié: (2026)

Efficient Training of Language Models with Compact and Consistent Next Token Distributions
par: Sathe, Ashutosh, et autres
Publié: (2024)

Flip-Flop Consistency: Unsupervised Training for Robustness to Prompt Perturbations in LLMs
par: Hejabi, Parsa, et autres
Publié: (2025)

Towards Best Practices for Open Datasets for LLM Training
par: Baack, Stefan, et autres
Publié: (2025)

Reliability-Aware Adaptive Self-Consistency for Efficient Sampling in LLM Reasoning
par: Kim, Junseok, et autres
Publié: (2026)

POET-X: Memory-efficient LLM Training by Scaling Orthogonal Transformation
par: Qiu, Zeju, et autres
Publié: (2026)

FineInstructions: Scaling Synthetic Instructions to Pre-Training Scale
par: Patel, Ajay, et autres
Publié: (2026)

Scaling Law for Quantization-Aware Training
par: Chen, Mengzhao, et autres
Publié: (2025)

A Unified View of Delta Parameter Editing in Post-Trained Large-Scale Models
par: Tang, Qiaoyu, et autres
Publié: (2024)

Efficiently Scaling LLM Reasoning with Certaindex
par: Fu, Yichao, et autres
Publié: (2024)

DAPO: An Open-Source LLM Reinforcement Learning System at Scale
par: Yu, Qiying, et autres
Publié: (2025)

Prescriptive Scaling Laws for Data Constrained Training
par: Lovelace, Justin, et autres
Publié: (2026)

DSDR: Dual-Scale Diversity Regularization for Exploration in LLM Reasoning
par: Wan, Zhongwei, et autres
Publié: (2026)

Consolidating Rewarded Perturbations for LLM Post-Training
par: Zhang, Zheyu, et autres
Publié: (2026)

Speculative Decoding Across Languages
par: Paudel, Nirajan, et autres
Publié: (2026)

Self-Training Meets Consistency: Improving LLMs' Reasoning with Consistency-Driven Rationale Evaluation
par: Lee, Jaehyeok, et autres
Publié: (2024)

Measuring Uncertainty in Transformer Circuits with Effective Information Consistency
par: Krasnovsky, Anatoly A.
Publié: (2025)

UNA: A Unified Supervised Framework for Efficient LLM Alignment Across Feedback Types
par: Wang, Zhichao, et autres
Publié: (2024)

Bridging Internal Probability and Self-Consistency for Effective and Efficient LLM Reasoning
par: Zhou, Zhi, et autres
Publié: (2025)

Scaling Optimal LR Across Token Horizons
par: Bjorck, Johan, et autres
Publié: (2024)

Prompt Curriculum Learning for Efficient LLM Post-Training
par: Gao, Zhaolin, et autres
Publié: (2025)

ZClip: Adaptive Spike Mitigation for LLM Pre-Training
par: Kumar, Abhay, et autres
Publié: (2025)

Judge Circuits
par: Feldhus, Nils, et autres
Publié: (2026)