minishlab

minishlab/potion-base-32M

libraryname: model2vec - name: potion-base-32M results: - dataset: config: en-ext name: MTEB AmazonCounterfactualClassification (en-ext) rev...

Model Documentation

potion-base-32M Model Card

This Model2Vec model is pre-trained using Tokenlearn. It is a distilled version of the baai/bge-base-en-v1.5 Sentence Transformer. It uses static embeddings, allowing text embeddings to be computed orders of magnitude faster on both GPU and CPU. It is designed for applications where computational resources are limited or where real-time performance is critical. It uses a larger vocabulary size than the potion-base-8M model which can be beneficial for tasks that require a larger vocabulary.

Installation

Install model2vec using pip:


pip install model2vec

Usage

Load this model using the from_pretrained method:

python
from model2vec import StaticModel

Load a pretrained Model2Vec model
model = StaticModel.from_pretrained("minishlab/potion-base-32M")

Compute text embeddings
embeddings = model.encode(["Example sentence"])

How it works

Model2vec creates a small, static model that outperforms other static embedding models by a large margin on all tasks on MTEB. This model is pre-trained using Tokenlearn. It's created using the following steps:

Distillation: first, a model is distilled from a sentence transformer model using Model2Vec.

Training data creation: the sentence transformer model is used to create training data by creating mean output embeddings on a large corpus.

Training: the distilled model is trained on the training data using Tokenlearn.

Post-training re-regularization: after training, the model is re-regularized by weighting the tokens based on their frequency, applying PCA, and finally applying SIF weighting.

Results

The results for this model are shown in the table below. The full Model2Vec results for all models can be found on the Model2Vec results page.


Average (All)                               52.46
Average (MTEB)                              51.66
Classification                              65.97
Clustering                                  35.29
PairClassification                          78.17
Reranking                                   50.92
Retrieval                                   33.52
STS                                         74.22
Summarization                               29.78
PEARL                                       55.37
WordSim                                     55.15

Additional Resources

All Model2Vec models on the hub

Library Authors

Model2Vec was developed by the Minish Lab team consisting of Stephan Tulkens and Thomas van Dongen.

Citation

Please cite the Model2Vec repository if you use this model in your work.


@software{minishlab2024model2vec,
  authors = {Stephan Tulkens and Thomas van Dongen},
  title = {Model2Vec: The Fastest State-of-the-Art Static Embeddings in the World},
  year = {2024},
  url = {https://github.com/MinishLab/model2vec}
}

Files & Weights

Filename	Size	Action
model.safetensors	0.12 GB