Search results

Is the bulk of the convergence speedup from the ConvPool?!

Status: Open.

#12 In google-research/nested-transformer;

· jxtps opened on Jan 28, 2025

PosEmbed is different for each block - shouldn't it be the same!?

Status: Open.

#11 In google-research/nested-transformer;

· jxtps opened on Jan 28, 2025

This seems like it would be a great option for increasing context window in sequences. Have you tried that yet?

Status: Open.

#9 In google-research/nested-transformer;

· Tylersuard opened on Jul 31, 2023

Training hours & Imagenet accuracy

Status: Open.

#8 In google-research/nested-transformer;

· arunos728 opened on Mar 2, 2022

Regarding GradCAT implementation

Status: Open.

#7 In google-research/nested-transformer;

· rush2406 opened on Feb 10, 2022

Can you make the Cifar training log public?

Status: Open.

#6 In google-research/nested-transformer;

· Freder-chen opened on Jan 27, 2022

Model Converge Problem

Status: Open.

#5 In google-research/nested-transformer;

· khawar-islam opened on Jul 17, 2021