dimitarpg13/semsimula-parflm-multixi

Rename Fock-Attention to 'Fock Attention (MLP V_theta)', fix stale pre-causal-leak-fix 9.30 PPL comparisons, add V_theta-vs-exchange-mechanism insight

Remove deprecated depth-cond OpenWebText row (retired due to causal leak) from SPLM Family Overview table

Move Verlet-instability row to the bottom of the SPLM Family Overview table

Reorder SPLM Family Overview: all TinyStories rows before all OpenWebText rows

Sync SPLM Family Overview table across the collection (order, completeness, corpus column)

add effective damping value to parameter table row

fix: use Unicode γ + HTML <sub> for gamma_eff (HF inline-math brace rendering issue)

clarify effective damping: LayerNorm makes gamma_eff << gamma_nominal (heavily underdamped)

Add riemannian-geodesics tag

fix: remove outer paren before \( math delimiter in CoT bullet

fix: remove outer parens before math delimiters; revert to bare underscore

fix: escape _ as \_ in prose inline math to prevent italic parsing

fix: use $...$ delimiters for inline math in bullet points

docs: add damped Riemannian geometry framework (June 2026 diagnostic battery results)

Upload README.md with huggingface_hub

Add links to Colab notebook and training results in companion repo

Update Multi-Xi SPLM PPL to 11.51 (16k rerun) and add checkpoint docs

Document available checkpoint and training artifacts in model card

Add training_summary.md (Multi-Xi PARFLM, PPL 12.06)

Add loss_curve.png (Multi-Xi PARFLM, PPL 12.06)

Add training_log.jsonl (Multi-Xi PARFLM, PPL 12.06)

Add checkpoint/model.pt (Multi-Xi PARFLM, PPL 12.06)

Add 'riemannian-geometry' tag

Add 'constant-memory-inference' tag

Add 'non-transformer' tag

Add 'attention-free' tag

Add Geometric Capabilities section: Riemannian metric, geodesic reasoning, hallucination detection, native CoT

Fix KaTeX rendering: remove bold wrapping and adjacent parens around math in table cells

Add 'Why Not a Transformer?' section with O(1) complexity, parameter comparison, and runtime capacity figure; fix KaTeX inline math; add Inference column to family table

Add runtime information capacity figure

Fix inline math: use \\(...\\) for KaTeX on HuggingFace

Fix math notation: convert plain text to LaTeX

Initial model card and config for semsimula-parflm-multixi

initial commit