Tracing Moral Foundations in Large Language Models
cs.CL, cs.AI
Submitted: 2026-01-09
Updated: 2026-11-09
Code: https://github.com/AiChiMoCha/MFT_LLMs
Project page: https://adamkarvonen.github.io/machine_learning/2024/06/11/sae-intuitions.html
Terminology
Sources
- Invariant Risk Minimization
- Emergent Misalignment: Narrow finetuning can produce broadly misaligned LLMs
- BatchTopK Sparse Autoencoders
- Learning Multi-Level Features with Matryoshka Sparse Autoencoders
- Persona Vectors: Monitoring and Controlling Character Traits in Language Models
- Sparse Autoencoders Find Highly Interpretable Features in Language Models
- Toy Models of Superposition
- The Pile: An 800GB Dataset of Diverse Text for Language Modeling
- Person-Centric Annotations of LAION-400M: Auditing Bias and Its Transfer to Models
- The Llama 3 Herd of Models
- Llama Scope: Extracting Millions of Features from Llama-3.1-8B with Sparse Autoencoders
- Measuring Massive Multitask Language Understanding
- Adaptive Stochastic Predictive Control from Noisy Data: A Sampling-based Approach
- The FineWeb Datasets: Decanting the Web for the Finest Text Data at Scale
- Moral Mimicry: Large Language Models Produce Moral Rationalizations Tailored to Political Identity
- Massive Activations in Large Language Models
- MFTCXplain: A Multilingual Benchmark Dataset for Evaluating the Moral Reasoning of LLMs through Multi-hop Hate Speech Explanation
- Steering Language Models With Activation Engineering
- Qwen2.5 Technical Report
- Qwen3 Technical Report
Related papers
- ALARM: Audio-Language Alignment for Reasoning Models
- ThinkSum: Probabilistic reasoning over sets using large language models
- Coherence boosting: When your pretrained language model is not paying enough attention
- From Facts to Insights: A Persona-Driven Dual Memory Framework and Dataset for Role-Playing Agents
- Improving General Role-Playing Agents via Psychology-Grounded Reasoning and Role-Aware Policy Optimization
- The Art of Asking: Multilingual Prompt Optimization for Synthetic Data