Sparse all-MLP

Training compute

5.3×10²⁰ FLOP

Parameters

9.4B

Published

Apr 14, 2022

Sparse all-MLP is an AI model developed by Meta AI (United States), first published in April 2022. It works in the language domain, on tasks such as language modeling.

Training it took an estimated 5.3×10²⁰ FLOP of compute (estimation method: hardware). The model has 9,410,000,000 parameters. It was trained on roughly 100B datapoints. Training ran on NVIDIA V100 for about 112 hours.

Access: Unreleased. Its weights are not openly released. The reference paper has 16 citations. Epoch AI rates the confidence of this record as confident.

Full record

Organization: Meta AI
Country of organization: United States
Domain: Language
Task: Language modeling
Training compute: 5.3×10²⁰ FLOP
Compute estimation method: Hardware
Parameters: 9,410,000,000
Dataset size: 100B
Training hardware: NVIDIA V100
Training time: 112 h
Model accessibility: Unreleased
Open weights: No
Citations: 16
Epoch confidence: Confident