Variational RHN + WT (PTB)

Training compute

1.3×10¹⁷ FLOP

Parameters

23M

Published

Jul 12, 2016

Variational RHN + WT (PTB) is an AI model developed by ETH Zurich and IDSIA (Switzerland), first published in July 2016. It works in the language domain, on tasks such as language modeling.

Training it took an estimated 1.3×10¹⁷ FLOP of compute. The model has 23,000,000 parameters.

Access: Unreleased. Its weights are not openly released. Epoch AI rates the confidence of this record as likely.

Full record

Organization: ETH Zurich, IDSIA
Country of organization: Switzerland
Domain: Language
Task: Language modeling
Training compute: 1.3×10¹⁷ FLOP
Parameters: 23,000,000
Model accessibility: Unreleased
Open weights: No
Epoch confidence: Likely