Snowflake/msmarco-v2.1-snowflake-arctic-embed-m-v1.5

该数据集包含MSMARCO-V2.1数据集的嵌入向量，这些嵌入向量是使用Snowflake的Arctic Embed M v1.5模型生成的，旨在为基于密集检索的方法提供一个简单的基线。数据集主要用于TREC RAG任务，支持嵌入截断和量化。数据集的大小在100M到1B之间，包含一个名为corpus的配置，其数据文件路径为corpus/*。

This dataset contains the embeddings for the MSMARCO-V2.1 dataset which is used as the corpora for TREC RAG. All embeddings are created using Snowflakes Arctic Embed M v1.5 and are intended to serve as a simple baseline for dense retrieval-based methods. The dataset is optimized for efficient embeddings and supports embedding truncation and quantization. The size of the dataset is between 100M and 1B, and it includes a configuration named corpus with data files located at corpus/*.