Clarify which EPQ version (v1/v2) is active for Flint

README.md

@@ -108,8 +108,22 @@ Throughput varied by about ±15 % between runs of the same tier; treat the t/s c
 
 See [`results/test-status.md`](results/test-status.md) for the full system and validation record, including what was not tested.
 
+## Which EPQ version is ported
+
+The source is [HelloSun's Nemotron EPQ](https://huggingface.co/HelloSun/NVIDIA-Nemotron-3.5-Bolt-74B-A9B-24GB-GGUF_EPQ) with two patches applied in order: `0001-epq-nemotron-h-moe.patch` (EPQ v1) and `epqv2.patch` (EPQ v2 increment). The Flint port applies both, but only these parts are active for Flint:
+
+| Feature | Source | Status for Flint |
+|---|---|---|
+| Expert paging, SSD→arena, per-expert custom op ("mode C") | v1 | Active; this is what all measurements use |
+| Preload all experts into the arena | v2 | Code present; runs only when the arena holds all 12,032 experts (about 28.6 GiB at 2.44 MiB/slot). No measured tier reaches this, so it is not exercised here |
+| Row-parallel MoE (16-thread row split) | v2 | Inactive: gated only for no-gate (Nemotron) experts; Flint has gate/up/down |
+
+The Flint-specific hook routes gated SiLU experts through the pager. Nemotron's no-gate/ReLU-squared path is not used by Flint.
+
+Flint's upstream `build_moe_ffn` also applies a power-of-two amax rescale of the down-projection input and undoes it on the output. This rescale is exact in F32 and Q8_K quantization is scale-invariant, so the paged path omits it. The layer-2–6 MoE comparison against native (which includes the rescale) confirms that the omission changes no results beyond float-summation noise.
+
 ## Attribution
 
-The pager implementation is adapted from [HelloSun's Nemotron EPQ v2](https://huggingface.co/HelloSun/NVIDIA-Nemotron-3.5-Bolt-74B-A9B-24GB-GGUF_EPQ). The Flint-specific graph hook routes gated SiLU experts through that pager; this does not add Nemotron's no-gate/ReLU-squared path to Flint.
+The pager is adapted from HelloSun's Nemotron EPQ v1 and v2 (linked above).
 
 The GGUF weights remain hosted by Unsloth. This repository contains source, patch, scripts, logs and test notes only.