Stars
Recipes to train reward model for RLHF.
CausalNLP is a practical toolkit for causal inference with text as treatment, outcome, or "controlled-for" variable.
Sample notebooks and prompts for LLM evaluation
Deploying Pyvis Interactive Network Graphs in Streamlit
Repository of Jupyter notebook tutorials for teaching the Deep Learning Course at the University of Amsterdam (MSc AI), Fall 2023
The fastai book, published as Jupyter Notebooks
Set of tools to assess and improve LLM security.
Röttger et al. (2023): "XSTest: A Test Suite for Identifying Exaggerated Safety Behaviours in Large Language Models"
Simple Python library/structure to ablate features in LLMs which are supported by TransformerLens
The aim of this dissertation is to assess the effectiveness of LLMs such as FinBERT and GPT-2 in detecting fraudulent activities in financial reports and statements. This repo provides the code for…
Aligning AI With Shared Human Values (ICLR 2021)
LightLLM is a Python-based LLM (Large Language Model) inference and serving framework, notable for its lightweight design, easy scalability, and high-speed performance.
FlagGems is an operator library for large language models implemented in Triton Language.
Material for cuda-mode lectures
Slides, notes, and materials for the workshop
Course to get into Large Language Models (LLMs) with roadmaps and Colab notebooks.
Transformer from scratch with einsum method
Code accompanying the paper "Massive Activations in Large Language Models"
[ICLR 2024] Efficient Streaming Language Models with Attention Sinks
Transformer based on a variant of attention that is linear complexity in respect to sequence length
Evals is a framework for evaluating LLMs and LLM systems, and an open-source registry of benchmarks.
Development repository for the Triton language and compiler
The official code of LM-Debugger, an interactive tool for inspection and intervention in transformer-based language models.
Collection of papers on state-space models