Fix typo in trillium llama2-70b script #1105

Open

wenxindongwork wants to merge 24 commits intomainfrom wenxindong-fix-typo

+153-37

Commits on Dec 18, 2024

Use jax.lax.dot_general when both lhs and rhs are not quantized
lenscloth
committed

Commits on Dec 19, 2024

fix typo
wenxindongwork
committed
Merge pull request #1103 from AI-Hypercomputer:fix_gmm_acc_dtype
maxtext authors
committed

Commits on Dec 20, 2024

Commits on Dec 22, 2024

Fix typo in custom remat config comments
gobbleturk
committed

Commits on Dec 23, 2024

Update type casting for activation
RissyRan
committed
Merge pull request #1111 from AI-Hypercomputer:fix_activation_type
maxtext authors
committed
Merge pull request #1114 from AI-Hypercomputer:raymondzou-llama31-8192
maxtext authors
committed
Merge pull request #1118 from tornikeo:patch-1
maxtext authors
committed
Merge pull request #1120 from AI-Hypercomputer:mattdavidow-fix-remat-typo
maxtext authors
committed
Print memory stats after start_step
hengtaoguo
committed
Bumps up Transformer Engine to 1.13 from 1.5.
wang2yn84
committed
Merge pull request #1110 from AI-Hypercomputer:hengtaoguo-ckpt
maxtext authors
committed
Merge pull request #1124 from AI-Hypercomputer:lance-bumpup-te
maxtext authors
committed

Commits on Dec 24, 2024