Dataset · sentence-transformers
Embedding training data
sentence-transformers/embedding-training-data
Training examples in jsonl.gz for text embedding models, including sentence-transformers.
What's inside
This repository contains raw datasets for training text embedding models. All files are in jsonl.gz format, where each line is a JSON object with a training example. The authors also prepared formatted versions in the Embedding Model Datasets collection and advise checking there first. The language is English.
What you can do with it
- Training sentence-transformers models
- Fine-tuning embedding models for search
- Comparing embedding training strategies
- Tasks
- Embeddings
- Languages
- English
- Downloads this month
- 3.2K
- Likes
- 145
Models trained on it
- KennethTM/MiniLM-L6-danish-encoderSentence similarity 7K
- doc2query/msmarco-t5-base-v1 646
- doc2query/msmarco-t5-small-v1 474
- doc2query/all-with_prefix-t5-base-v1 189
- doc2query/all-t5-base-v1 186
- Hum-Works/lodestone-base-4096-v1Sentence similarity 132
- dotproductx/breadbowl-embed-v1.3Sentence similarity 112
- KennethTM/MiniLM-L6-danish-encoder-v2Sentence similarity 81
How to load it
from datasets import load_dataset
ds = load_dataset("sentence-transformers/embedding-training-data")