Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

"added by us" placement #58

Open
wants to merge 1 commit into
base: master
Choose a base branch
from
Open

"added by us" placement #58

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 2 additions & 2 deletions train/tr11-176B-ml/README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -335,7 +335,7 @@ Thus: `sqrt(1/(14336*3)) = 0.00482197968631537`

### Positional Encoding

We use the added by us AliBi implementation:
We use the AliBi implementation added by us:

```
--position-embedding-type alibi \
Expand All @@ -345,7 +345,7 @@ Paper: [Train Short, Test Long: Attention with Linear Biases Enables Input Lengt

### Embed LayerNorm

We use the added by us embedding layer norm which makes the training more stable at a small training slowdown cost and a tiny additional amount of memory.
We use the embedding layer norm added by us which makes the training more stable at a small training slowdown cost and a tiny additional amount of memory.

```
--embed-layernorm \
Expand Down