Skip to content

chaning learning rate back to default of 1e-3 #156

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

Open
schnamo wants to merge 2 commits into
base: dev
Choose a base branch
from
+7 −24
Open

chaning learning rate back to default of 1e-3 #156

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
c5d2ccf
chaning learning rate back to default of 1e-3, however, we recommend …
schnamo Feb 18, 2026
acc8a9a
delete duplicate config, fix vocab size
sfluegel05 Feb 19, 2026
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
6 changes: 3 additions & 3 deletions configs/model/electra-for-pretraining.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -4,16 +4,16 @@ init_args:
class_path: chebai.loss.pretraining.ElectraPreLoss
out_dim: null
optimizer_kwargs:
lr: 1e-4
lr: 1e-3
config:
generator:
vocab_size: 1400
vocab_size: 4400
max_position_embeddings: 1800
num_attention_heads: 8
num_hidden_layers: 6
type_vocab_size: 1
discriminator:
vocab_size: 1400
vocab_size: 4400
max_position_embeddings: 1800
num_attention_heads: 8
num_hidden_layers: 6
Expand Down
2 changes: 1 addition & 1 deletion configs/model/electra.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,7 +2,7 @@ class_path: chebai.models.Electra
init_args:
model_type: classification
optimizer_kwargs:
lr: 1e-4
lr: 1e-3
config:
vocab_size: 4400
max_position_embeddings: 1800
Expand Down
3 changes: 2 additions & 1 deletion configs/model/electra300.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
@@ -1,9 +1,10 @@
class_path: chebai.models.Electra
init_args:
model_type: classification
optimizer_kwargs:
lr: 1e-3
config:
vocab_size: 1400
vocab_size: 4400
max_position_embeddings: 301
num_attention_heads: 8
num_hidden_layers: 6
Expand Down
18 changes: 0 additions & 18 deletions configs/model/electra_pretraining.yml
View file
Open in desktop

This file was deleted.

2 changes: 1 addition & 1 deletion configs/model/electra_tox.yml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2,7 +2,7 @@ class_path: chebai.models.Electra
init_args:
model_type: classification
optimizer_kwargs:
lr: 1e-4
lr: 1e-3 # we recommend 1e-4 for OPT finetuning, however, 1e-3 is the default
config:
vocab_size: 1400
max_position_embeddings: 1800
Expand Down