Spaces:

lucalp
/

blt-entropy-patcher

Running on Zero

App Files Files Community

blt-entropy-patcher / bytelatent

398 kB

12 contributors

History: 12 commits

par-meta's picture

Fix distributed all reduce grad norm (#40)

c79b1fd unverified 10 months ago

configs
This includes fixes that make checkpointing and reloading work correctly. (#35) 11 months ago
data
This includes fixes that make checkpointing and reloading work correctly. (#35) 11 months ago
model
Initial codes and scripts for training entropy model (#34) 11 months ago
plotting
Add plotting code from paper (#17) 11 months ago
preprocess
Fix realtime entropy patching (#26) 11 months ago
tokenizers
Initial commit 12 months ago
.DS_Store

6.15 kB

Initial commit 12 months ago
__init__.py

96 Bytes

Initial commit 12 months ago
args.py

11.1 kB

This includes fixes that make checkpointing and reloading work correctly. (#35) 11 months ago
base_transformer.py

19.8 kB

Changes for training entropy model and correcting attention in local models (#25) 11 months ago
checkpoint.py

10.5 kB

This includes fixes that make checkpointing and reloading work correctly. (#35) 11 months ago
constants.py

141 Bytes

Initial commit 12 months ago
distributed.py

15.5 kB

Changes for training entropy model and correcting attention in local models (#25) 11 months ago
entropy_model.py

1.41 kB

Changes for training entropy model and correcting attention in local models (#25) 11 months ago
eval.py

8.97 kB

This includes fixes that make checkpointing and reloading work correctly. (#35) 11 months ago
float8.py

4.89 kB

Initial commit 12 months ago
generate.py

17.1 kB

This includes fixes that make checkpointing and reloading work correctly. (#35) 11 months ago
logger.py

4.22 kB

Replace regular filesystem calls with fsspec + add s3 support (#18) 11 months ago
metrics.py

6.87 kB

Initial commit 12 months ago
norms.py

4.34 kB

Fix distributed all reduce grad norm (#40) 10 months ago
optim.py

4.61 kB

Initial commit 12 months ago
probe.py

24 kB

Initial commit 12 months ago
profiling.py

4.4 kB

Initial commit 12 months ago
stool.py

7.35 kB

Initial commit 12 months ago
test_blt.py

16.3 kB

Initial codes and scripts for training entropy model (#34) 11 months ago
test_entropy_model.py

2 kB

Changes for training entropy model and correcting attention in local models (#25) 11 months ago
train.py

27 kB

Fix distributed all reduce grad norm (#40) 10 months ago
transformer.py

6.57 kB

Changes for training entropy model and correcting attention in local models (#25) 11 months ago