understanding-search / maze-datasetLinks
maze datasets for investigating OOD behavior of ML systems
☆64Updated 2 weeks ago
Alternatives and similar repositories for maze-dataset
Users that are interested in maze-dataset are comparing it to the libraries listed below
Sorting:
- This is code for most of the experiments in the paper Understanding the Effects of RLHF on LLM Generalisation and Diversity☆47Updated last year
- ☆104Updated last year
- ☆108Updated 8 months ago
- Rewarded soups official implementation☆60Updated 2 years ago
- Code for the paper "VinePPO: Unlocking RL Potential For LLM Reasoning Through Refined Credit Assignment"☆179Updated 5 months ago
- ☆63Updated 8 months ago
- Code for Contrastive Preference Learning (CPL)☆176Updated 11 months ago
- Paper collections of the continuous effort start from World Models.☆188Updated last year
- Reinforcement Learning via Regressing Relative Rewards☆36Updated 10 months ago
- ☆54Updated 11 months ago
- ☆73Updated last year
- A library for efficient patching and automatic circuit discovery.☆79Updated 3 months ago
- Code for "Reasoning to Learn from Latent Thoughts"☆122Updated 7 months ago
- Benchmarking Agentic LLM and VLM Reasoning On Games☆205Updated 2 months ago
- Function Vectors in Large Language Models (ICLR 2024)☆182Updated 6 months ago
- ☆139Updated 3 months ago
- ☆185Updated last year
- Official pytorch implementation of "Interpreting the Second-Order Effects of Neurons in CLIP"☆40Updated 11 months ago
- ☆52Updated 6 months ago
- Easy-to-Hard Generalization: Scalable Alignment Beyond Human Supervision☆125Updated last year
- This is the official repository for the "Towards Vision-Language Mechanistic Interpretability: A Causal Tracing Tool for BLIP" paper acce…☆23Updated last year
- Code release for "Debating with More Persuasive LLMs Leads to More Truthful Answers"☆118Updated last year
- ☆131Updated last year
- [NeurIPS 2025] What Makes a Reward Model a Good Teacher? An Optimization Perspective☆39Updated last month
- ☆85Updated last year
- Reinforcing General Reasoning without Verifiers☆91Updated 4 months ago
- Code for "Echo Chamber: RL Post-training Amplifies Behaviors Learned in Pretraining"☆24Updated 3 weeks ago
- Official implementation of "Direct Preference-based Policy Optimization without Reward Modeling" (NeurIPS 2023)☆42Updated last year
- Official implementation of Bootstrapping Language Models via DPO Implicit Rewards☆44Updated 6 months ago
- ☆34Updated 7 months ago