Search results

NeuralMemory forward when num_chunks == 0

Status: Open.

#59 In lucidrains/titans-pytorch;

· x54-729 opened on Jan 27, 2026

How do I set the hyperparameters and start formal training?

Status: Open.

#58 In lucidrains/titans-pytorch;

· Kuanzhaipcx opened on Jan 27, 2026

Question about execution order of “memory as context”: retrieval → attention → update

Status: Open.

#57 In lucidrains/titans-pytorch;

· yangzhifeio opened on Jan 8, 2026

Relations between sequence, segment, batch and chunk

Status: Open.

#55 In lucidrains/titans-pytorch;

· embarrassing-noob-questions opened on Dec 25, 2025

Any pretrained weights? How big is the model

Status: Open.

#51 In lucidrains/titans-pytorch;

· embarrassing-noob-questions opened on Aug 10, 2025

RuntimeError: Triton Error [CUDA]: invalid argument

Status: Open.

#50 In lucidrains/titans-pytorch;

· bdonyan opened on Jun 30, 2025

Treatment of single_token case.

Status: Open.

#47 In lucidrains/titans-pytorch;

· erlebach opened on Apr 21, 2025

cache_store_seq

Status: Open.

#46 In lucidrains/titans-pytorch;

· erlebach opened on Apr 16, 2025

need infer.py code

Status: Open.

#44 In lucidrains/titans-pytorch;

· XuJianzhi opened on Apr 7, 2025

Updating error in NeuralMemory::forward

Status: Open.

#43 In lucidrains/titans-pytorch;

· erlebach opened on Apr 5, 2025

Incorrect statement seq, values_seq = seq

Status: Open.

#42 In lucidrains/titans-pytorch;

· erlebach opened on Apr 5, 2025

Issue in train_mac.py

Status: Open.

#41 In lucidrains/titans-pytorch;

· Yusef-Zou opened on Apr 5, 2025