RNN+weight noise+dynamic eval

Training compute

4.2×10¹⁵ FLOP

Parameters

54M

Published

Aug 4, 2013

RNN+weight noise+dynamic eval is an AI model developed by University of Toronto (Canada), first published in August 2013. It works in the language domain, on tasks such as language modeling.

Training it took an estimated 4.2×10¹⁵ FLOP of compute (estimation method: operation counting). The model has 54,000,000 parameters. It was trained on roughly 929K datapoints.

Access: Unreleased. Its weights are not openly released. The reference paper has 4,734 citations. Epoch AI rates the confidence of this record as confident.

Full record

Organization: University of Toronto
Country of organization: Canada
Domain: Language
Task: Language modeling
Training compute: 4.2×10¹⁵ FLOP
Compute estimation method: Operation counting
Parameters: 54,000,000
Dataset size: 929K
Model accessibility: Unreleased
Open weights: No
Citations: 4,734
Epoch confidence: Confident