gmongaras/Cottention_Transformer

Code for the paper "Cottention: Linear Transformers With Cosine Attention"

20
/ 100
Experimental

This project offers an alternative to standard Transformer models for researchers and practitioners working with large language models (LLMs). It helps train and fine-tune BERT and GPT-style models using 'Cottention,' a linear transformer with cosine attention. The input is raw text data, and the output is a trained or fine-tuned language model ready for downstream tasks.

Use this if you are an AI researcher or machine learning engineer looking to experiment with novel transformer architectures for improved efficiency or performance in natural language processing tasks.

Not ideal if you are looking for a plug-and-play solution for everyday NLP tasks without needing to delve into model training or architecture specifics.

natural-language-processing large-language-models deep-learning-research transformer-models text-generation
No License No Package No Dependents
Maintenance 6 / 25
Adoption 6 / 25
Maturity 8 / 25
Community 0 / 25

How are scores calculated?

Stars

20

Forks

Language

Cuda

License

Last pushed

Nov 15, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/transformers/gmongaras/Cottention_Transformer"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.