Upload README.md with huggingface_hub

README.md

@@ -20,7 +20,7 @@ Public experimental byte-level causal language model. Training is in progress an
 - Approximately 1B stored parameters and approximately one third active per token.
 - UTF-8 bytes, vocabulary 256, no external tokenizer.
 - Four local causal Transformer layers process every byte.
-- Entropy boundaries compress the stream to about one patch per eight bytes, with capacity for 192 patches per 1024-byte sequence.
+- Entropy boundaries compress the stream to about one patch per eight bytes, with capacity for 192 patches per 1024-byte sequence. Over-cap boundaries are uniformly coalesced across the sequence instead of truncating its tail.
 - Three independent eight-layer global routes are trained cyclically, one full batch per route.
 - The previous completed global patch conditions the local prediction, preventing future-token leakage.
 - Muon optimizes hidden matrices; AdamW optimizes embeddings and normalization parameters.