AWD-LSTM + MoS + Partial Shuffled

Training compute

3.1×10¹⁷ FLOP

Parameters

35M

Published

Jun 10, 2019

AWD-LSTM + MoS + Partial Shuffled is an AI model developed by University of Texas at Austin (United States), first published in June 2019. It works in the language domain, on tasks such as language modeling.

Training it took an estimated 3.1×10¹⁷ FLOP of compute (estimation method: operation counting). The model has 35,000,000 parameters. It was trained on roughly 2M datapoints.

Access: Open weights (non-commercial). Its weights are openly available. The reference paper has 126 citations. Epoch AI rates the confidence of this record as confident.

Full record

Organization: University of Texas at Austin
Country of organization: United States
Domain: Language
Task: Language modeling
Training compute: 3.1×10¹⁷ FLOP
Compute estimation method: Operation counting
Parameters: 35,000,000
Dataset size: 2M
Model accessibility: Open weights (non-commercial)
Open weights: Yes
Citations: 126
Epoch confidence: Confident