オーナー

bigscience-workshop

索引済みのリポジトリ 21 件 · GitHub で見る

• petals

  🌸 Run LLMs at home, BitTorrent-style. Fine-tuning and inference up to 10x faster than offloading

  Python · スター 10565 個

• promptsource

  Toolkit for creating, sharing and using natural language prompts.

  Python · スター 3034 個

• Megatron-DeepSpeed

  Ongoing research training transformer language models at scale, including: BERT & GPT-2

  Python · スター 1448 個

• bigscience

  Central place for the engineering/scaling WG: documentation, SLURM scripts and logs, compute environment and data.

  Shell · スター 1019 個

• xmtf

  Crosslingual Generalization through Multitask Finetuning

  Jupyter Notebook · スター 534 個

• biomedical

  Tools for curating biomedical training data for large-scale language modeling

  Python · スター 505 個

• t-zero

  Reproduce results and replicate training fo T0 (Multitask Prompted Training Enables Zero-Shot Task Generalization)

  Python · スター 463 個

• data-preparation

  Code used for sourcing and cleaning the BigScience ROOTS corpus

  Jupyter Notebook · スター 318 個

• lam

  Libraries, Archives and Museums (LAM)

  スター 91 個

• data_tooling

  Tools for managing datasets for governance and training.

  HTML · スター 91 個

• multilingual-modeling

  BLOOM+1: Adapting BLOOM model to support a new unseen language

  Python · スター 75 個

• evaluation

  Code and Data for Evaluation WG

  Python · スター 42 個

• data_sourcing

  This directory gathers the tools developed by the Data Sourcing Working Group

  Python · スター 31 個

• metadata

  Experiments on including metadata such as URLs, timestamps, website descriptions and HTML tags during pretraining.

  Python · スター 29 個

• carbon-footprint

  A repository for `codecarbon` logs.

  Jupyter Notebook · スター 13 個

• tokenization

  Python · スター 11 個

• pii_processing

  PII Processing code to detect and remediate PII in BigScience datasets. Reference implementation for the PII Hackathon

  Python · スター 9 個

• ShadesofBias

  Evaluation for Shades of Bias in Text

  HTML · スター 9 個

• catalogue_data

  Scripts to prepare catalogue data

  Jupyter Notebook · スター 8 個

• training_dynamics

  スター 5 個

• ほか 1 件のリポジトリは GitHub で見る にあります