Zaoral @zaoral@fedibird.com

> Llama 3.1 Swallowの指示チューニングで用いられている対話データセットlmsys-chat-1m-synth

https://huggingface.co/datasets/tokyotech-llm/lmsys-chat-1m-synth

https://x.com/Youmima1015/status/1850810678522552789

https://x.com/_stakaya/status/1850661522370105452

https://github.com/meta-llama/llama-recipes/tree/main/recipes/quickstart/NotebookLlama

https://x.com/itnavi2022/status/1850569873082216808

> MetaがNotebookLlama

https://x.com/2022_technology/status/1849820644302782599

https://huggingface.co/grapevine-AI/aya-expanse-32b-gguf

Aya Expanse（8B、32B
https://x.com/jaguring1/status/1849450649744003553

https://x.com/jaguring1/status/1849443900257395123
>インドの10言語に特化した言語モデル「Sarvam-1

https://x.com/ELYZA_inc/status/1848529707018449140

Granite 3.0
https://x.com/jaguring1/status/1848476533742240128
https://www.ibm.com/new/ibm-granite-3-0-open-state-of-the-art-enterprise-models

https://x.com/sasa_kuna_/status/1848537860737274003
https://zenn.dev/knowledgesense/articles/1aa689d88d0553
>GenEOL:複数の文章変換と要約でRAGの性能を向上

https://x.com/A7_data/status/1847875555884224757

https://x.com/K_Ryuichirou/status/1848255823648272568
databricks

https://x.com/K_Ryuichirou/status/1848297174167347335

Agent-as-a-Judge
https://arxiv.org/abs/2410.10934
https://x.com/theomitsa/status/1847983242151862334

https://zenn.dev/hellorusk/books/e56548029b391f/viewer/ollama2

https://note.com/kumo_trade/n/nd1ae54f9cb77

https://news.yahoo.co.jp/articles/de26f706065cfce066f52d3e92843febf1a29a78?source=sns&dv=sp&mid=other&date=20241020&ctg=bus&bt=tw_up

https://x.com/erukiti/status/1847817065853571539

https://github.com/Kazuhito00/Janus-1.3B-Colaboratory-Sample

Janus-1.3B
https://x.com/karaage0703/status/1847558091631624495
https://kazuhito00.hatenablog.com/entry/2024/10/19/171529

https://x.com/Aratako_LM/status/1847333864743850439