
Sam Passaglia
@SamPassaglia
Followers
326
Following
321
Media
35
Statuses
212
Enterprise LLMs for Japan @cohere Prev. PhD @UChicagoAstro 🇺🇸🇯🇵🇫🇷
Joined June 2020
RT @drlucylai: Had so much fun hosting this panel “Is Scale Enough?” on Algorithms Day for the Open Problems for AI Summit in Tokyo 🇯🇵 with….
0
2
0
RT @hpp_ricecake: 日英4.4T tokensで学習した日本語ModernBERTを公開しました!!. 系列長8192、語彙数は日英10万、パラメータ数130Mながら既存largeモデルと同等以上の性能があります.12データセットによる既存BERT系モデルの網羅….
0
90
0
RT @llm_jp: LLM-jp Chatbot Arena を公開しました。.LLM-jp-3 172Bを含む計10モデルと会話できます。収集したデータは LLM-jp から公開予定です。.2/11 (火) 9:00 までの….
chatbot-arena.apps.llmc.nii.ac.jp
0
29
0
New Differential Transformer paper (@ytz2024 ++) is really cool: they make attention heads differential, computing two attention maps per input and subtracting them. This improves performance by cancelling out noise, like a humbucking guitar.
arxiv.org
Transformer tends to overallocate attention to irrelevant context. In this work, we introduce Diff Transformer, which amplifies attention to the relevant context while canceling noise....
0
1
4
I had such a blast at @tokyoaijp's NLP session last night hearing from our 4 amazing speakers, @lhl @ayase_lab @ayaniwa1213 @loem_ms, and speaking to the ~100 attendees!.Now I'm pumped to organize more events with @kaixhin and @ikulyatin in the future :)
0
7
26
RT @ikulyatin: 7th August is our 3rd TAI AAI (@tokyoaijp Advanced AI) session, focused on NLP. Come listen to researchers and engineers fro….
lu.ma
NOTE: registration is required 24 hours before the event (to generate the entrance QR codes). Our Community Tokyo AI (TAI) is a community composed of people…
0
1
0
My team at Elyza has been selected by the Japanese government to receive a major supercomputer grant to develop foundation models! . Excited to put these GPUs to work :).
【お知らせ】経済産業省が立ち上げた「GENIAC」のもと、NEDOが公募した「競争力ある生成AI基盤モデルの開発(助成)」に、ELYZAが採択されたことをお知らせいたします。当社では本採択を受け、日本語処理能力の高いモデル構築に取り組んでまいります。 .
1
0
13
RT @satyanadella: Today we announced our plans to deepen our investments in Japan, spanning cloud and AI infrastructure, skilling, research….
0
470
0
RT @ELYZA_inc: 【お知らせ】700億パラメータの日本語LLMを開発し、グローバルモデルに匹敵する性能を達成しました。本モデルを含むモデル群を「ELYZA LLM for JP」シリーズとして順次サービス提供を開始します。. まずはデモサイトで性能をお試しください。….
elyza.ai
ELYZAは大規模言語モデル活用のプロフェッショナル集団です。「未踏の領域で、あたりまえを創る」という理念のもと、自然言語処理技術の研究開発を行い、企業の大規模言語モデル活用の支援や、独自LLM開発の支援、AI SaaSの開発・提供をしています。
0
229
0
If you're in Kobe attending NLP 2024, come visit me at ELYZA's booth all week! We have cute stickers.
0
0
15
RT @Dorialexander: So big announcement: thanks to the generous support from @huggingface I am releasing the early modern ChatGPT, MonadGPT….
0
91
0
RT @_jamico: Biden's AI Executive Order is out and it’s terrible for US innovation. Here are some of the new obligations, which only large….
0
63
0
A new article for Nature by @robotopia discusses the race to develop home-grown LLMs in Japan! 🐪 There's even a couple quotes from me discussing Rakuda.
0
1
7
RT @tomo_wb: @SamPassaglia We use your Rakuda Benchmark in our blog (written in Japanese.).Thank you!.
0
1
0
Another new SOTA Japanese LLM has just been released, weblab-10b from the University of Tokyo Matsuo Lab's @kojima_tks 🥳🥳! . My GPU cluster is down for upgrades this weekend so adding it to Rakuda will have to wait, but on the JGLUE benchmark weblab-10b leads the pack!
0
0
10
Exciting day in the Japanese LLM space, with @StabilityAI_JP releasing their first japanese-focused model, japanese-stablelm-7b ! It is off to a good start on the Rakuda benchmark of Japanese LLMs.
1
7
48