AI 公司 MosaicML 推出 70 亿参数模型 MPT-7B-8K，号称“一次处理 8000 字长文本、可商...
source link: https://www.8btc.com/article/6827372
Go to the source link to view the article. You can view the picture content, updated content and better typesetting reading experience. If the link is broken, please click the button below to view the snapshot at that time.
AI 公司 MosaicML 推出 70 亿参数模型 MPT-7B-8K，号称“一次处理 8000 字长文本、可商用”
据 IT 之家 7 月 25 日消息，AI 创业公司 MosaicML 近日发布了其 70 亿参数模型 MPT-7B-8K。据悉，该模型一次可以处理 8000 字文本，相当擅长处理长文重点摘要和问答，还能在 MosaicML 平台上根据特定任务，进一步微调相关配置。
据悉，系列模型采用了 150 万个 Token，并以 256 块 H100 GPU 花 3 天完成模型训练而成。MosaicML 本次发布了 3 个版本模型，包括 MPT-7B-8k、MPT-7B-8k-Instruct 和 MPT-7B-8k-Chat。
Recommend
-
7
AI ‘a lens through which we can move data,...
-
6
MosaicML debuts inference service to make generative AI deployment affordable
-
6
Cloudflare R2 and MosaicML enable training LLMs on any compute, anywhere in the world, with zero switching costs 05/16/2023
-
6
MosaicML challenges OpenAI with its new open-source language model
-
2
This Week in AI: Databricks’ Acquisition of MosaicML
-
5
AIGC领域最大收购：Databricks 13亿美元买下MosaicML，成立仅2年员工60人 硅星人 发表于 2023年06月27日...
-
4
Databricks is acquiring MosaicML for a jaw-dropping $1.3 billion
-
9
Databricks and MosaicML CEOs say $1.3 billion deal is about helping enterprises create AI IP
-
7
新浪VR > 正文页 谷歌 DeepMind 正开发新型 AI 模型 Gemini，号称“比 GPT-4 强”...
-
5
AMD AI chips are nearly as fast as Nvidia's, MosaicML says Software is key in LLM training By
About Joyk
Aggregate valuable and interesting links.
Joyk means Joy of geeK