You can deploy this model and start earning money today!

Pradheep1647/
kibitzer-az-elo-minus-0417-step-000005

Name: kibitzer-az-elo-minus-0417-step-000005
Availability: InStock

Views

No views yet

Pradheep1647/kibitzer-az-elo-minus-0417-step-000005

kibitzer checkpoint for a chess policy/value model trained on game sequences. each board is encoded as 64 square tokens plus auxiliary state, summarized by a square-level encoder, then processed by a causal transformer over the game timeline.

status

checkpoint: az_iter_5.pt
step: 000005
elo_rating: 903 estimated vs stockfish-elo-1320 (-416.6 elo diff)
github_repo: https://github.com/Mantissagithub/kibitzer
eval_opponent: stockfish-elo-1320
elo_diff: -416.6
eval_score: 1.00

architecture

input: 64 chess-square piece tokens plus 7 auxiliary scalars covering side to move, castling rights, en-passant file, and halfmove clock.
position encoder: bidirectional square-level transformer that compresses each board into one timeline token.
timeline trunk: causal transformer with d_model=384, sequence length 256, 12 layers, and 8 attention heads.
transformer blocks: pre-norm rmsnorm, rope on causal q/k attention, pytorch scaled dot-product attention, and swiglu feed-forward layers.
outputs: policy head over 4,672 alphazero-style moves (64 * 73) plus a tanh value head for bounded outcome prediction.

files

az_iter_5.pt: pytorch checkpoint.
training_metadata.yaml: training config and checkpoint metrics.
post_eval.yaml: uploaded after local stockfish/cutechess eval.

usage

bash

git clone https://github.com/Mantissagithub/kibitzer.git
cd kibitzer
uv run python scripts/uci.py --checkpoint <path-to>/az_iter_5.pt

this checkpoint is one artifact in a sequence. repos with elo-pending in the name have not been evaluated yet; rated repos are renamed after scripts/eval_and_rename_hf.py --from-hf completes.