TRIMELMext (247M)

Training compute

3.1×10¹⁹ FLOP

Parameters

247M

Published

May 25, 2022

TRIMELMext (247M) is an AI model developed by Princeton University (United States), first published in May 2022. It works in the language domain, on tasks such as language modeling/generation and translation.

Training it took an estimated 3.1×10¹⁹ FLOP of compute. The model has 247,000,000 parameters. It was trained on roughly 21.1B datapoints.

Access: Open weights (non-commercial). Its weights are openly available. The reference paper has 152 citations. Epoch AI rates the confidence of this record as confident.

Full record

Organization: Princeton University
Country of organization: United States
Domain: Language
Task: Language modeling/generation, Translation
Training compute: 3.1×10¹⁹ FLOP
Parameters: 247,000,000
Dataset size: 21.1B
Model accessibility: Open weights (non-commercial)
Open weights: Yes
Citations: 152
Epoch confidence: Confident