Stephen Roller's starred repositories
flash-attention
Fast and memory-efficient exact attention
Megatron-LM
Ongoing research training transformer models at scale
tokenizers
đź’Ą Fast State-of-the-Art Tokenizers optimized for Research and Production
The-NLP-Pandect
A comprehensive reference for all topics related to Natural Language Processing
longformer
Longformer: The Long-Document Transformer
mkdocstrings
:blue_book: Automatic documentation from sources, for MkDocs.
bigscience
Central place for the engineering/scaling WG: documentation, SLURM scripts and logs, compute environment and data.
filesystem_spec
A specification that python filesystems should adhere to.
PrefixTuning
Prefix-Tuning: Optimizing Continuous Prompts for Generation
MyST-Parser
An extended commonmark compliant parser, with bridges to docutils/sphinx
lambdaprompt
λprompt - A functional programming interface for building AI systems
ParlAI_SearchEngine
A search engine for ParlAI's BlenderBot project (and probably other ones as well)
simmc
With the aim of building next generation virtual assistants that can handle multimodal inputs and perform multimodal actions, we introduce two new datasets (both in the virtual shopping domain), the annotation schema, the core technical tasks, and the baseline models. The code for the baselines and the datasets will be opensourced.
forked-pdb
Python pdb for multiple processes