PaulPauls / llama3_interpretability_saeLinks
A complete end-to-end pipeline for LLM interpretability with sparse autoencoders (SAEs) using Llama 3.2, written in pure PyTorch and fully reproducible.
☆614Updated 2 months ago
Alternatives and similar repositories for llama3_interpretability_sae
Users that are interested in llama3_interpretability_sae are comparing it to the libraries listed below
Sorting:
- Bayesian Optimization as a Coverage Tool for Evaluating LLMs. Accurate evaluation (benchmarking) that's 10 times faster with just a few l…☆284Updated last week
- A library for making RepE control vectors☆593Updated 4 months ago
- Visualize the intermediate output of Mistral 7B☆362Updated 4 months ago
- LLM Analytics☆664Updated 7 months ago
- Open weights language model from Google DeepMind, based on Griffin.☆639Updated last week
- Code to train and evaluate Neural Attention Memory Models to obtain universally-applicable memory systems for transformers.☆309Updated 7 months ago
- Stop messing around with finicky sampling parameters and just use DRµGS!☆349Updated last year
- ☆243Updated last year
- a small code base for training large models☆299Updated last month
- Felafax is building AI infra for non-NVIDIA GPUs☆561Updated 4 months ago
- Absolute minimalistic implementation of a GPT-like transformer using only numpy (<650 lines).☆251Updated last year
- An implementation of bucketMul LLM inference☆217Updated 11 months ago
- Sparsify transformers with SAEs and transcoders☆547Updated last week
- Visualizing the internal board state of a GPT trained on chess PGN strings, and performing interventions on its internal board state and …☆205Updated 6 months ago
- a curated list of data for reasoning ai☆136Updated 9 months ago
- A scientific instrument for investigating latent spaces☆701Updated 2 weeks ago
- Achieve the llama3 inference step-by-step, grasp the core concepts, master the process derivation, implement the code.☆584Updated 3 months ago
- An interactive HTML pretty-printer for machine learning research in IPython notebooks.☆416Updated last month
- Fully neural approach for text chunking☆350Updated last month
- Dead Simple LLM Abliteration☆218Updated 3 months ago
- A BERT that you can train on a (gaming) laptop.☆207Updated last year
- Official codebase for the paper "Beyond A* Better Planning with Transformers via Search Dynamics Bootstrapping".☆367Updated 11 months ago
- See Through Your Models☆390Updated 2 months ago
- The repository for the code of the UltraFastBERT paper☆514Updated last year
- Pytorch script hot swap: Change code without unloading your LLM from VRAM☆126Updated last month
- A JAX research toolkit for building, editing, and visualizing neural networks.☆1,780Updated last month
- ☆200Updated 2 months ago
- llama3.np is a pure NumPy implementation for Llama 3 model.☆981Updated last month
- Implement recursion using English as the programming language and an LLM as the runtime.☆234Updated 2 years ago
- Animating R1's thoughts.