Influence functions - why, what and how — LessWrong

Anthropic recently published the paper Studying Large Language Model Generalization with Influence Functions, which describes a scalable technique fo…

x Influence functions - why, what and how — LessWrong Machine Learning (ML) AI Frontpage 78 Influence functions - why, what and how by Nina Panickssery 15th Sep 2023 10 min read 6 78 Anthropic recently published the paper Studying Large Language Model Generalization with Influence Functions , which describes a scalable technique for measuring which training examples were most influential for a particular set of weights/outputs of a trained model. This can help us better understand model generalization, offering insights into the emergent properties of AI systems. For instance, influence functi

related reading

• A Mathematical Framework for Transformer Circuitstransformer-circuits.pub
• [2308.03296] Studying Large Language Model Generalization with Influence Functionsarxiv.org
• In-context Learning and Induction Headstransformer-circuits.pub
• Tracing Model Outputs to the Training Data \ Anthropicanthropic.com
• Illustrating Reinforcement Learning from Human Feedback (RLHF)huggingface.co
• interpreting GPT: the logit lens — LessWronglesswrong.com
• Softmax Linear Unitstransformer-circuits.pub
• Maybe I was too harsh on deep learning theory (three days ago) — LessWronglesswrong.com
• Non_Interactive – Software & MLnonint.com
• Do Machine Learning Models Memorize or Generalize?pair.withgoogle.com
• Transformers Inference Optimization Toolset | AstraBlogastralord.github.io
• How do LLMs generalize when we do training that is intuitively compatible with two off-distribution behaviors? — LessWronglesswrong.com