transformers
Proper nemotron H and 3 and 2
#45569
Open
Overview
Commits
15
Changes
View On GitHub

Proper nemotron H and 3 and 2 #45569

ArthurZucker wants to merge 15 commits into main from nemotron-h-split-dense-sparse
ArthurZucker
xenova Support mlp mixers
34b9a137
xenova Fix forward
b4fd7e13
xenova make changes to modular
9e5d3370
ArthurZucker Merge branch 'main' of github.com:huggingface/transformers into nemot…
c7acbb11
ArthurZucker up
55aa51ff
ArthurZucker tyle
30bffaef
ArthurZucker nit
f4eb5e76
ArthurZucker update
85a0cc75
ArthurZucker add
8390498a
ArthurZucker refactor: use DecoderLayer pattern; drop moe_latent_size; inherit Mix…
9907265b
ArthurZucker refactor: Jamba-style decoder layers, inherit NemotronMLP/JambaModel
da5a518f
ArthurZucker ArthurZucker changed the title Split Proper nemotron H and 3 and 2 110 days ago
ArthurZucker Merge remote-tracking branch 'origin/main' into nemotron-h-split-dens…
af3d5b0b
ArthurZucker test: migrate nemotron_h_{dense,sparse} tests to CausalLMModelTester
3e71a29d
ArthurZucker ci: fix code-quality, repo-consistency, and doc-build for nemotron_h …
e57b10e7
github-actions
HuggingFaceDocBuilderDev
ArthurZucker nemotron_h_sparse: support optional latent MoE projection
84a2be05
github-actions

Login to write a write a comment.

Login via GitHub

Reviewers
No reviews
Assignees
No one assigned
Labels
None yet
Milestone
No milestone
Terms Of Use Privacy Policy Refund Policy Impressum