Hugging Face
Models
Datasets
Spaces
Buckets
new
Docs
Enterprise
Pricing
Log In
Sign Up
thomas-schweich
/
PAWN
like
0
Other
PyTorch
Rust
English
pawn
chess
transformer
world-model
causal-lm
next-token-prediction
representation-learning
parameter-efficient-finetuning
arxiv:
2401.04679
License:
apache-2.0
Model card
Files
Files and versions
xet
Community
main
PAWN
/
pawn
358 kB
Ctrl+K
Ctrl+K
3 contributors
History:
18 commits
thomas-schweich
Add UCI engine generation, Parquet data pipeline, and Lc0/Stockfish containers
4c4517f
about 1 hour ago
adapters
Add RoSA adapter with gradient-informed sparse masks (#3)
about 5 hours ago
dashboard
Add patience clock chart to dashboard (inferred from val loss)
1 day ago
eval_suite
Merge branch 'review-fixes'
about 7 hours ago
__init__.py
338 Bytes
Move PAWN to open source repo.
4 days ago
checkpoint.py
22.9 kB
Fix bugs, performance issues, and doc errors from code review
about 20 hours ago
config.py
2.61 kB
Safetensors migration, checkpoint integrity, and multi-model training. (#1)
1 day ago
data.py
7.17 kB
Safetensors migration, checkpoint integrity, and multi-model training. (#1)
1 day ago
data_utils.py
661 Bytes
Move PAWN to open source repo.
4 days ago
gpu.py
3.42 kB
Fix bugs, performance issues, and doc errors from code review
about 20 hours ago
lichess_data.py
15.1 kB
Add UCI engine generation, Parquet data pipeline, and Lc0/Stockfish containers
about 1 hour ago
logging.py
11.5 kB
Consolidate all training logging through MetricsLogger
1 day ago
model.py
18.9 kB
Fix bugs, performance issues, and doc errors from code review
about 20 hours ago
sweep.py
29.1 kB
Add RoSA adapter with gradient-informed sparse masks (#3)
about 5 hours ago
trainer.py
20.8 kB
Fix bugs, performance issues, and doc errors from code review
about 20 hours ago