VectorInstitute / kaleidoscope-sdk
A user toolkit for analyzing and interfacing with Large Language Models (LLMs)
☆24Updated 4 months ago
Related projects ⓘ
Alternatives and complementary repositories for kaleidoscope-sdk
- LLM finetuning in resource-constrained environments.☆41Updated 4 months ago
- ☆12Updated 8 months ago
- The nnsight package enables interpreting and manipulating the internals of deep learned models.☆400Updated this week
- A user toolkit for analyzing and interfacing with Large Language Models (LLMs)☆21Updated 2 months ago
- Efficient LLM inference on Slurm clusters using vLLM.☆38Updated last week
- Mechanistic Interpretability Visualizations using React☆196Updated 4 months ago
- This repository collects all relevant resources about interpretability in LLMs☆283Updated last week
- Tools for understanding how transformer predictions are built layer-by-layer☆429Updated 5 months ago
- Influence Functions with (Eigenvalue-corrected) Kronecker-Factored Approximate Curvature☆100Updated 3 months ago
- Training Sparse Autoencoders on Language Models☆453Updated this week
- ☆108Updated last year
- Using sparse coding to find distributed representations used by neural networks.☆181Updated last year
- ☆187Updated last month
- A fast, effective data attribution method for neural networks in PyTorch☆177Updated last month
- Interpretability for sequence generation models 🐛 🔍☆375Updated this week
- ☆141Updated 3 weeks ago
- ☆102Updated last month
- Sparse Autoencoder for Mechanistic Interpretability☆188Updated 3 months ago
- Sparse probing paper full code.☆50Updated 10 months ago
- ☆138Updated 4 months ago
- Conformal Language Modeling☆22Updated 10 months ago
- Layer-Wise Relevance Propagation for Large Language Models and Vision Transformers [ICML 2024]☆97Updated this week
- ☆168Updated 8 months ago
- NanoGPT-like codebase for LLM training☆73Updated this week
- Repository for research in the field of Responsible NLP at Meta.☆183Updated 2 months ago
- Create feature-centric and prompt-centric visualizations for sparse autoencoders (like those from Anthropic's published research).☆158Updated last month
- DataInf: Efficiently Estimating Data Influence in LoRA-tuned LLMs and Diffusion Models (ICLR 2024)☆52Updated last month
- Stanford NLP Python Library for Understanding and Improving PyTorch Models via Interventions☆633Updated last week
- ☆75Updated 9 months ago
- Code for T-Few from "Few-Shot Parameter-Efficient Fine-Tuning is Better and Cheaper than In-Context Learning"☆430Updated last year