Update README.md

README.md

@@ -25,7 +25,12 @@ Generating 200 tokens on CPU takes 226 ms with PSSA against 2735 ms with the tra
 - refractory gating that suppresses repeated writes
 - a scalar CPU reference implementation kept as a correctness twin for every optimized path
 
+## Weights and checkpoints
+
+- [NSC64/pssa-100m](https://huggingface.co/NSC64/pssa-100m): a 109M-parameter PyTorch port of this architecture, trained from scratch by NSC64 on 4B tokens of TinyStories and FineWeb-Edu. Ships bf16 safetensors weights, a 16k byte-level BPE tokenizer and a self-contained loader (plain scan on CPU, fused Triton kernel on CUDA).
+- [sparticle62/pssa-checkpoints-run2](https://huggingface.co/sparticle62/pssa-checkpoints-run2): training checkpoints from the 105M run in progress (raw .pssa resume files, not a packaged model).
+
 ## Status
 
-Research code under active development. Weights are not published in this repository yet; the GitHub repository holds the training chain, the evaluation harness and the scripts to reproduce the numbers above.
+Research code under active development. The links above hold the published 109M PyTorch port and the in-training 105M checkpoints; the GitHub repository holds the training chain, the evaluation harness and the scripts to reproduce the numbers above.