Skip to content

M1-128M Hybrid (from scratch, attention + liquid core)

Latest

Choose a tag to compare

@everest-an everest-an released this 20 Aug 09:26
· 2 commits to main since this release

M1-128M — native 128M hybrid, trained from scratch

What it is: 128.6M-parameter hybrid — window attention + liquid core (selective decay, exp parameterization) in every layer. 12 layers, d_model 832, 512-token training window. English text continuation — a research backbone, not a chat model.

Serve it:
\\�ash
CKPT_PATH=hybrid_125m_serve.pt TOKENIZER=gpt2 python -m uvicorn serve.server:app
\\

Measured (see RESULTS.md / ABLATIONS.md):

  • O(1) carried state: 0.381 MB flat from 512 to 1M tokens (8,063× smaller than a matched KV-cache)
  • Selective-decay exp parameterization: parity length extrapolation 0.953–1.000 vs tanh 0.023
  • Language-modeling quality is NOT the selling point (honest: PPL trails a matched Transformer at 125M — the case is memory form-factor and efficiency)

License: MIT.