dannyallover / overthinking_the_truthLinks
☆29Updated last year
Alternatives and similar repositories for overthinking_the_truth
Users that are interested in overthinking_the_truth are comparing it to the libraries listed below
Sorting:
- [NAACL'25 Oral] Steering Knowledge Selection Behaviours in LLMs via SAE-Based Representation Engineering☆60Updated 7 months ago
- Restore safety in fine-tuned language models through task arithmetic☆28Updated last year
- Augmenting Statistical Models with Natural Language Parameters☆27Updated 9 months ago
- ☆44Updated 3 months ago
- LoFiT: Localized Fine-tuning on LLM Representations☆39Updated 5 months ago
- ☆40Updated last year
- Semi-Parametric Editing with a Retrieval-Augmented Counterfactual Model☆68Updated 2 years ago
- Repo for paper: Examining LLMs' Uncertainty Expression Towards Questions Outside Parametric Knowledge☆14Updated last year
- BeHonest: Benchmarking Honesty in Large Language Models☆34Updated 10 months ago
- Code for the paper "A Mechanistic Interpretation of Arithmetic Reasoning in Language Models using Causal Mediation Analysis"☆18Updated 2 weeks ago
- Methods and evaluation for aligning language models temporally☆29Updated last year
- A Mechanistic Understanding of Alignment Algorithms: A Case Study on DPO and Toxicity.☆72Updated 3 months ago
- ☆27Updated 2 years ago
- Github repository for "FELM: Benchmarking Factuality Evaluation of Large Language Models" (NeurIPS 2023)☆59Updated last year
- Analyzing LLM Alignment via Token distribution shift☆16Updated last year
- ☆74Updated last year
- Align your LM to express calibrated verbal statements of confidence in its long-form generations.☆26Updated last year
- AbstainQA, ACL 2024☆26Updated 8 months ago
- ☆41Updated last year
- ☆41Updated 8 months ago
- Evaluating the Ripple Effects of Knowledge Editing in Language Models☆55Updated last year
- ☆25Updated 2 weeks ago
- Official code for ICML 2024 paper on Persona In-Context Learning (PICLe)☆25Updated last year
- ☆51Updated last year
- ☆51Updated 2 months ago
- Code for ACL 2023 paper "BOLT: Fast Energy-based Controlled Text Generation with Tunable Biases".☆21Updated last year
- ☆17Updated last year
- Providing the answer to "How to do patching on all available SAEs on GPT-2?". It is an official repository of the implementation of the p…☆11Updated 5 months ago
- ☆17Updated last year
- ☆11Updated 4 months ago