---
title: Token Routing Interactive Paper
emoji: 🧭
colorFrom: purple
colorTo: green
sdk: static
app_file: index.html
pinned: false
license: cc-by-nc-4.0
short_description: Interactive report for the TR-HASH MoE 200M multi-hash SFT.
---

# TR-HASH MoE 200M — Interactive Multi-Hash Report

Interactive companion for the **201.2M-parameter TR-HASH MoE full-SFT**
checkpoint selected from epoch 2 on PIQA.

The Space provides:

- an architecture explorer for the shared SwiGLU and deterministic multi-hash
  top-2 residual path;
- checkpoint-derived tokenizer IDs and expert assignments for selected tokens;
- the recorded three-epoch full-parameter SFT and PIQA diagnostics;
- live chat generation from `AETHORIA-AI/TR-HASH-MoE-200M-160B-SFT`.

The model has 16 layers, a 32k vocabulary, GQA 14/2, four stored residual
experts, two independent token-ID hash routes per layer, and a shared dense
SwiGLU branch. There is no learned expert router and no LoRA adapter.

Its metrics describe only this 201.2M model. They are training diagnostics and
a full PIQA validation measurement, not a dense-baseline comparison or a broad
general-capability benchmark.

## Sources

- [TR-HASH MoE 200M full SFT](https://huggingface.co/AETHORIA-AI/TR-HASH-MoE-200M-160B-SFT)
- [160B-source refinement checkpoint](https://huggingface.co/AETHORIA-AI/TR-HASH-MoE-200M-160B-Refinement)
- [Training framework](https://github.com/Complexity-ML/complexity-framework)
- [TR-Hash-i64](https://github.com/Complexity-ML/TR-Hash-i64)