Mr-Verse1/SEFQ-1bit-LLM

Domaine:

natural language processing

Type de record:

dataset

Créateur:

Mr-

Hôte:

🚀 A 7-day independent research project from Nigeria 🇳🇬⚡ No GPU. No funding. Built entirely on a phone. SEFQ (Spectral Error-Feedback Quantization) is a proposed method for converting FP16 large language models into 1-bit representations without massive retraining. Instead of naive quantization, SEFQ combines: Spectral Initialization (preserve weight direction)

Mr-MaNia7/gebeta

Traditional Ethiopian/Habeshan game with Python3 and OpenGL # Gebeta (ገበጣ) Traditional Ethiopian/H

mr-williams/Nigerian_Covid_Data

Using the subnational data about Covid19 in Nigeria per day - Infections (new cases), Deaths, Recove

Dataset Mr. Chipofya.xlsx

Oxidative stress, a key player in neurodegenerative disorders such as Alzheimer's and Parkinson's di

Mr-Ndi/SomaGov

SomaGov is a citizen engagement platform that enables the public to submit feedback or complaints to

mr-jones123/dka

Automate Low-Resource Language datasets using dka! # dka **Team name:** iForgot **Members:** Xyn

MR-AMR Dataset

MR-AMR is a public dataset composed of 140000 images of the digits in full-state and mid-state. It c