Transformer-XL + RelationLM

Parameters

124M

Published

Jan 24, 2022

Transformer-XL + RelationLM is an AI model developed by DeepMind and University of Oxford (United Kingdom), first published in January 2022. It works in the language domain, on tasks such as language modeling/generation.

Epoch AI has no training-compute estimate for this model. The model has 124,000,000 parameters. It was trained on roughly 103M datapoints. Training ran on 64 Google TPU v2.

Access: Unreleased. Its weights are not openly released. The reference paper has 35 citations.

Full record

Organization: DeepMind, University of Oxford
Country of organization: United Kingdom
Domain: Language
Task: Language modeling/generation
Compute estimation method: Operation counting
Parameters: 124,000,000
Dataset size: 103M
Training hardware: Google TPU v2
Chips used: 64
Training power draw: 36.1 kW
Model accessibility: Unreleased
Open weights: No
Citations: 35