Ecosyste.ms: Issues

An open API service for providing issue and pull request metadata for open source projects.

GitHub / kojimano/megatron-deepspeed-abci issues and pull requests

#32 - created README

Pull Request - State: closed - Opened by kojimano over 1 year ago

#31 - Scale 544gpus

Pull Request - State: closed - Opened by kojimano over 1 year ago

#30 - Update replace_breaks.py

Pull Request - State: closed - Opened by kuriyan1204 over 1 year ago

#29 - Update replace_breaks.py

Pull Request - State: closed - Opened by kuriyan1204 over 1 year ago

#28 - Addline break replacer

Pull Request - State: closed - Opened by kuriyan1204 over 1 year ago

#27 - Add `\n` replace script

Issue - State: closed - Opened by kuriyan1204 over 1 year ago - 1 comment

#26 - Add github dataset from redpajama

Issue - State: open - Opened by kuriyan1204 over 1 year ago - 1 comment

#26 - Add github dataset from redpajama

Issue - State: open - Opened by kuriyan1204 over 1 year ago - 1 comment

#25 - Debug data processing pipelines

Issue - State: open - Opened by kuriyan1204 over 1 year ago - 2 comments

#24 - Tokenizer周りについて

Issue - State: open - Opened by keisks almost 2 years ago

#23 - Adding Abeja Japanese Tokenizer into a pipeline.

Pull Request - State: closed - Opened by kojimano almost 2 years ago

#22 - CyberAgent Preprocessing / Binarize Data

Issue - State: closed - Opened by kojimano almost 2 years ago

#21 - Dataset Preprocessing Validation

Issue - State: open - Opened by kojimano almost 2 years ago

#20 - General Preprocessing Pipeline

Issue - State: open - Opened by kojimano almost 2 years ago
Labels: datasets

#19 - ABCI 544 GPU rehersal

Issue - State: closed - Opened by kojimano almost 2 years ago

#19 - ABCI 544 GPU rehersal

Issue - State: closed - Opened by kojimano almost 2 years ago

#18 - 青空文庫 Preprocessing / Binarize Data

Issue - State: closed - Opened by kojimano almost 2 years ago - 1 comment
Labels: datasets

#17 - Wikipedia Preprocessing / Binarize Data

Issue - State: closed - Opened by kojimano almost 2 years ago - 1 comment
Labels: datasets

#16 - ABCI GPU ベンチマーク

Issue - State: closed - Opened by kojimano almost 2 years ago - 1 comment
Labels: system

#15 - Sambanovaで使うデータセットの用意

Issue - State: closed - Opened by losyer almost 2 years ago

#15 - Sambanovaで使うデータセットの用意

Issue - State: closed - Opened by losyer almost 2 years ago

#14 - タイムライン

Issue - State: closed - Opened by keisks almost 2 years ago

#14 - タイムライン

Issue - State: closed - Opened by keisks almost 2 years ago

#13 - instruction tuningについて

Issue - State: open - Opened by keisks almost 2 years ago - 4 comments

#13 - instruction tuningについて

Issue - State: open - Opened by keisks almost 2 years ago - 4 comments

#12 - SambaNovaのサーバーで何を学習するか

Issue - State: closed - Opened by losyer almost 2 years ago

#11 - SambaNovaの学習でどのデータを使うか

Issue - State: closed - Opened by losyer almost 2 years ago - 1 comment

#10 - 前処理はどこでなにをやるか

Issue - State: closed - Opened by losyer almost 2 years ago - 2 comments

#9 - データの置き場所

Issue - State: closed - Opened by losyer almost 2 years ago - 3 comments

#8 - 学習データの最終的な固め方

Issue - State: closed - Opened by losyer almost 2 years ago

#7 - 聞いておきたいこと

Issue - State: closed - Opened by losyer almost 2 years ago - 1 comment

#6 - CACCのtokenizeについて確認

Issue - State: closed - Opened by losyer almost 2 years ago - 1 comment

#5 - stats of CACC data

Issue - State: closed - Opened by losyer almost 2 years ago - 2 comments

#4 - ABCIグランドチャレンジでどのデータを使うか決める

Issue - State: closed - Opened by losyer almost 2 years ago - 5 comments

#3 - data_news_articles

Issue - State: closed - Opened by kojimano almost 2 years ago - 2 comments

#2 - evaluation_jglue

Issue - State: open - Opened by kojimano almost 2 years ago - 2 comments

#1 - model_tokenizer

Issue - State: closed - Opened by kojimano almost 2 years ago - 2 comments