SRU++ Large

Training compute

2.1×10¹⁹ FLOP

Parameters

234M

Published

Feb 24, 2021

SRU++ Large is an AI model developed by ASAPP (United States), first published in February 2021. It works in the language domain, on tasks such as language modeling.

Training it took an estimated 2.1×10¹⁹ FLOP of compute (estimation method: operation counting,hardware). The model has 234,000,000 parameters. It was trained on roughly 103M datapoints. Training ran on NVIDIA V100.

Access: Open weights (unrestricted). Its weights are openly available. The reference paper has 54 citations. Epoch AI rates the confidence of this record as confident.

Full record

Organization: ASAPP
Country of organization: United States
Domain: Language
Task: Language modeling
Training compute: 2.1×10¹⁹ FLOP
Compute estimation method: Operation counting, Hardware
Parameters: 234,000,000
Dataset size: 103M
Training hardware: NVIDIA V100
Chip-hours: 360
Numerical format: FP16
Model accessibility: Open weights (unrestricted)
Open weights: Yes
Citations: 54
Epoch confidence: Confident