✳flâneur — a map of the web's best reading
The Relational Bottleneck as an Inductive Bias for Efficient Abstraction
arxiv.org · saved by 1 readers
N/A
Explore this link on the map →related reading
- The Law of Leaky Abstractions – Joel on Softwarejoelonsoftware.com
- GitHub - konst-int-i/lucid-rules: Rule Extraction Methods for Interactive eXplainability · GitHubgithub.com
- Innovation Prior - reconstructed reasoning tracesbohanlyu.com
- Abstraction Enables Thoughtjacobbuckman.com
- Jacobian Lens – Qwen3.6-27B ｜ Neuronpedianeuronpedia.org
- GitHub - PaulPauls/llama3_interpretability_sae: A complete end-to-end pipeline for LLM interpretability with sparse autoencoders (SAEs) using Llama 3.2, written in pure PyTorch and fully reproducible. · GitHubgithub.com
- Inference to the Best Explanation (article)hps.cam.ac.uk
- [2309.05660] Hypothesis Search: Inductive Reasoning with Language Modelsarxiv.org
- Branches · HazyResearch/intelligence-per-watt · GitHubgithub.com
- GitHub - ByteByteGoHq/system-design-101: Explain complex systems using visuals and simple terms. Help you prepare for system design interviews. · GitHubgithub.com
- A mechanism for solving relational tasks in transformer language modelsarxiv.org
- GitHub - inverse-scaling/prize: A prize for finding tasks that cause large language models to show inverse scaling · GitHubgithub.com