Upload README.md with huggingface_hub

README.md

@@ -107,10 +107,9 @@ for score, idx in zip(top5.values, top5.indices):
 
 ## Limitations
 
-- The model has seen 47 % of the planned training budget (120K / 255K steps) — the loss is still decreasing.
 - It has no access to board state: all chess knowledge is inferred from move sequences.
 - No RLHF or self-play refinement — this is a pure next-token prediction model.
-- Predictions can include illegal moves; use `python-chess` to filter at inference time.
+- Predictions can include illegal moves; use `python-chess` to filter at inference time. (see the [chessgpt-inference](https://github.com/malcouffe/chessgpt-inference) repo for legal move masking while generating.)
 
 ## Citation