PSSA is a small language model built on a state space architecture instead of attention. Where a transformer re-reads its whole context window for every token, a state space model carries a fixed-size recurrent state forward, so the cost of the next token does not grow with the length of what came before. It belongs to the same family of architectures as S4 and Mamba: selective state space models.

NSC64 ported the architecture to PyTorch and trained the result from scratch: 4B tokens in two phases, 50/50 TinyStories and FineWeb-Edu, then an 85/15 web/story fine-tune. The port keeps the same scan and memory-bank design as the Rust reference; the two departures it needed at this scale are backprop truncated at 256 tokens and the scan running as a fused Triton kernel with a hand-written backward that matches a plain PyTorch loop to about 1e-7.

Training checkpoints from the 105M run land on pssa-checkpoints-105m as the run goes. They are raw .pssa resume files rather than a packaged model, so they are useful for resuming and auditing, not for direct inference.