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.
☆601Updated 3 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
- LLM Analytics☆646Updated 5 months ago
- Bayesian Optimization as a Coverage Tool for Evaluating LLMs. Accurate evaluation (benchmarking) that's 10 times faster with just a few l…☆279Updated last month
- A scientific instrument for investigating latent spaces☆673Updated 3 weeks ago
- Visualize the intermediate output of Mistral 7B☆344Updated 2 months ago
- Felafax is building AI infra for non-NVIDIA GPUs☆555Updated last month
- Open weights language model from Google DeepMind, based on Griffin.☆627Updated last month
- Official codebase for the paper "Beyond A* Better Planning with Transformers via Search Dynamics Bootstrapping".☆365Updated 9 months ago
- Textbook on reinforcement learning from human feedback☆488Updated this week
- A pure NumPy implementation of Mamba.☆219Updated 8 months ago
- Code to train and evaluate Neural Attention Memory Models to obtain universally-applicable memory systems for transformers.☆297Updated 5 months ago
- A library for making RepE control vectors☆560Updated 2 months ago
- Absolute minimalistic implementation of a GPT-like transformer using only numpy (<650 lines).☆250Updated last year
- An implementation of bucketMul LLM inference☆215Updated 8 months ago
- Dead Simple LLM Abliteration☆211Updated last month
- a small code base for training large models☆288Updated 3 months ago
- Things you can do with the token embeddings of an LLM☆1,432Updated last month
- Fine-tune LLM agents with online reinforcement learning☆1,094Updated last year
- An interactive HTML pretty-printer for machine learning research in IPython notebooks.☆398Updated last week
- Run and explore Llama models locally with minimal dependencies on CPU☆191Updated 5 months ago
- Finetune llama2-70b and codellama on MacBook Air without quantization☆448Updated 11 months ago
- Code behind Arxiv Papers☆509Updated 11 months ago
- OpenAI's Structured Outputs with Logprobs☆154Updated last month
- llama3.np is a pure NumPy implementation for Llama 3 model.☆974Updated 9 months ago
- Neural Search☆352Updated last week
- Visualizing the internal board state of a GPT trained on chess PGN strings, and performing interventions on its internal board state and …☆201Updated 4 months ago
- ☆242Updated last year
- This project collects GPU benchmarks from various cloud providers and compares them to fixed per token costs. Use our tool for efficient …☆220Updated 3 months ago
- Lightweight Nearest Neighbors with Flexible Backends☆260Updated 2 weeks ago