Hugging Face
Models
Datasets
Spaces
Community
Docs
Enterprise
Pricing
Log In
Sign Up
Spaces:
lucalp
/
blt-entropy-patcher
like
7
Running
on
Zero
App
Files
Files
Community
1
Fetching metadata from the HF Docker repository...
295e2d6
blt-entropy-patcher
/
bytelatent
Ctrl+K
Ctrl+K
12 contributors
History:
58 commits
lucalp
Fixing proper UTF-8 representation
8eea094
about 2 months ago
configs
Remove byte tokenizer and add config args to switch between byte/patch packing (#68)
4 months ago
data
Open source weights! (#97)
3 months ago
model
Improve HF integration (#98)
3 months ago
plotting
Fixing proper UTF-8 representation
about 2 months ago
preprocess
Allow ArrowIterator to read from json (#45)
5 months ago
tokenizers
Remove byte tokenizer and add config args to switch between byte/patch packing (#68)
4 months ago
.DS_Store
Safe
6.15 kB
Initial commit
7 months ago
__init__.py
Safe
96 Bytes
Initial commit
7 months ago
args.py
Safe
12.5 kB
Fix eval mask (#93)
3 months ago
base_transformer.py
Safe
19.9 kB
Initialize rope embeddings properly for the entropy model (#72)
4 months ago
checkpoint.py
Safe
11.6 kB
Add way to call consolidate (#80)
4 months ago
config_parser.py
Safe
2.74 kB
When merging configs, do not merge data sources (#79)
4 months ago
constants.py
Safe
141 Bytes
Initial commit
7 months ago
distributed.py
Safe
17.9 kB
Init distributed when loading model (#94)
3 months ago
entropy_model.py
Safe
1.51 kB
local block causal when cuda avail
2 months ago
eval.py
Safe
13.4 kB
Open source weights! (#97)
3 months ago
float8.py
Safe
4.89 kB
Initial commit
7 months ago
generate.py
Safe
18.1 kB
Open source weights! (#97)
3 months ago
generate_blt.py
Safe
7.25 kB
Get generation working for BLT (#86)
3 months ago
generate_patcher.py
Safe
4.64 kB
Visualisation working on CPU via CUDA_VISIBLE_DEVICE=-1 python demo_patcher.py 'Daenerys Targaryen is in Game of Thrones, a fantasy epic by George R.R. Martin.'
2 months ago
iterate_data.py
Safe
1.11 kB
Update iterate_data (#81)
4 months ago
logger.py
Safe
4.44 kB
Update checkpointing to use fsspec (#39)
5 months ago
metrics.py
Safe
7.11 kB
Get evals working again. (#46)
4 months ago
norms.py
Safe
4.34 kB
Fix distributed all reduce grad norm (#40)
5 months ago
optim.py
Safe
4.61 kB
Initial commit
7 months ago
print_config.py
Safe
263 Bytes
Make it possible to specify multiple config files (#54)
5 months ago
probe.py
Safe
24 kB
Initial commit
7 months ago
profiling.py
Safe
4.4 kB
Initial commit
7 months ago
stool.py
Safe
7.41 kB
Fix rsync to not preserve original permissions, instead use destination (#76)
4 months ago
test_blt.py
Safe
16.3 kB
Initial codes and scripts for training entropy model (#34)
5 months ago
test_config_parser.py
Safe
5.36 kB
Make it possible to specify multiple config files (#54)
5 months ago
test_entropy_model.py
Safe
2.03 kB
Test first batch matches (#53)
5 months ago
train.py
Safe
31.7 kB
Get evals working again. (#46)
4 months ago
transformer.py
Safe
7.47 kB
Initialize rope embeddings properly for the entropy model (#72)
4 months ago
