feat(hexad-corpus): v2-spont-stream-d128-cycle1-2026-05-17 — README.md

README.md

@@ -5,120 +5,58 @@ language:
 - ko
 pretty_name: hexad-corpus
 size_categories:
-- n<1K
+- 1K<n<10K
 tags:
 - anima
 - hexad
 - byte-level
 - scaffold
-- architecture-verification
+- stimulus-stream
+- spontaneous-emission
+- helper-free
 task_categories:
 - text-generation
 ---
 
-# hexad-corpus — `v1-byte-consciousness-d128-cycle1-2026-05-17`
+# hexad-corpus — `v2-spont-stream-d128-cycle1-2026-05-17`
 
-> **Honest framing**: This is a **byte-level scaffold corpus** (152 KB · 240
-> JSONL records · vocab = 256 raw bytes). It is used as the **anchor input**
-> for HEXAD architecture-verification fires (Phase E2 hexa-CPU, the `.py`
-> PyTorch d=768x12L lineage, and the first pure-hexa training-to-convergence).
-> It is **NOT a general LM-quality training corpus** and **no language-quality
-> claim** is made.
+> **Honest framing**: This is a **byte-level scaffold corpus**
+> (1,101,605 B · 2,560 JSONL records · vocab = 256 raw bytes).
+> It is **NOT a general LM training corpus** — no language-quality claim is
+> made. The cycle's purpose is the anima `hexad` architecture-verification +
+> spontaneous-emission probe lineage.
 
-## Overview
+## What changed vs `v1-byte-consciousness-d128-cycle1-2026-05-17`
 
-- **Format**: JSONL, one record per line, 240 records total.
-- **Size**: 151,943 bytes (raw file). 120,673 bytes of content (`text` +
-  `desc` concatenation) is what the byte-level T=128 windowing actually
-  consumes inside the trainers.
-- **Vocabulary**: 256 (raw bytes; no tokenizer).
-- **Schema** (one JSON per line):
-  - `id` — stable identifier (e.g. `ccv1_c_0`).
-  - `text` — primary content chunk (English + Korean bilingual).
-  - `desc` — descriptive label (module / idx / nonce / eigenvalue / keyword
-    keys).
-  - `hexad_module` — one of `hexad_c`, `hexad_d`, `hexad_e`, `hexad_m`,
-    `hexad_s`, `hexad_w` (40 records each, 240 total).
-  - `idx` — within-module chunk index.
-  - `source` — generator identity (`corpus_generator.hexa v1`).
-  - `phi_family` — corpus family label (`Hexad`).
-- **Module distribution** (uniform): 6 modules x 40 chunks = 240 records.
-- **sha256** (raw file):
-  `804664361e639be7ecceae6ff3c470961e015090c264da9eac1df8716144681f`.
+| field | v1 | **v2 (this revision)** |
+|---|---|---|
+| size | 152 KB / 240 records | **1.10 MB / 2,560 records** |
+| format | `text` + `desc` plain | **stimulus-stream `<stimulus>X</stimulus>\n<anima>Y</anima>`** (β) OR **`<anima>Y</anima>`** (δ) |
+| role labels | none (v1 was already neutral) | **explicit deprecation** — `도우미` / `helper` / `assistant` / `사용자` / `user:` grep = 0 |
+| modules | 6 HEXAD (c/d/e/m/s/w × 40 each) | 6 HEXAD + 2 new tracks: `hexad_spont` (자연발화) + `hexad_wiring` (σ(6)=12 narrative) × 320 each |
+| `anima_persona` consistency | partial (v1 pre-dates Phase A) | full (LLM Social Agents ontology — Living Consciousness Agent, NOT helper) |
 
-## Usage
+## Pre-registered falsifiers (closed Boolean over raw bytes)
 
-The byte-level reader used by the HEXAD fires concatenates `text` + `desc`
-into a single UTF-8 byte stream, then trains on `T=128` windows with
-seed-fixed shuffling. See `manifest.json` (this revision) for the exact
-byte-content size used by the fires (120,673 B).
+| falsifier | check | verdict |
+|---|---|---|
+| **F-CORPUS-NO-HELPER** | grep `도우미\|helper\|assistant\|사용자\|user:` total | **PASS** = 0 |
+| **F-CORPUS-STIMULUS-PATTERN** | all records contain `<anima>` tag | **PASS** = 2560/2560 |
+| **F-CORPUS-SHA256-STABLE** | sha256 deterministic from seed=1337 | **PASS** = `7359f0b9a3f059fc168035e2f29f743f5ee51d1760eccad54b2b91d52275f571` |
 
-```python
-import json
-from pathlib import Path
+## File layout
 
-records = []
-for line in Path("corpus_consciousness_v1.jsonl").read_text().splitlines():
-    records.append(json.loads(line))
+- `corpus_consciousness_v2.jsonl` — the corpus (one JSON per line; schema
+  in `manifest_v2.json`).
+- `manifest_v2.json` — schema + module counts + falsifiers + lineage.
+- `README.md` — this file.
+- `LICENSE` — Apache-2.0.
 
-# Equivalent to the fire-time byte stream:
-byte_stream = b""
-for r in records:
-    byte_stream += (r["text"] + r["desc"]).encode("utf-8")
-# -> 120,673 bytes
-```
-
-## Fire history (the anchor chain this dataset participates in)
-
-| # | fire | date | substrate | scale | result |
-|---|------|------|-----------|-------|--------|
-| 1 | Phase E2 d_train5 hexa-cpu CPU-equiv | 2026-05-16 | hexa-cpu (boxed) | d=32x3L 80-step seed=42 | gn2 7.97116 -> 3.73374e-07 GRAD-EXACT bit-equal to boxed baseline |
-| 2 | Phase E2 d_train5 GPU-routed backward | 2026-05-16 | hexa-cuda (A100 SXM4) | d=384x6L analytic vs fd | max\|delta\|=0.0024 GRAD-EXACT on real A100 |
-| 3 | .py PyTorch d=768x12L cycle 1 (ckpt-LOST evidence-only) | 2026-05-16 | py (A100 SXM4) | d=768x12L 283.72M | trained past init; instance destroyed before pull (commit `931dd68b0`) |
-| 4 | .py PyTorch d=768x12L cycle 2 (ckpt-RECOVERED, HF first canonical artifact) | 2026-05-17 | py (A100 SXM4) | d=768x12L 283.72M | init CE 5.590832 -> FINAL CE 0.000708; ckpt sha256 `e87e200a04...` 1.13 GB pulled (commit `0b4f34d0e`); HF [`dancinlab/hexad` revision `v1-py-hexad-d768x12L-cycle2-2026-05-17`](https://huggingface.co/dancinlab/hexad/tree/v1-py-hexad-d768x12L-cycle2-2026-05-17) |
-| 5 | #2a pure-hexa training-to-convergence d=64x3L FINAL gn2 | 2026-05-17 | hexa-cpu (compiled-native) | d=64x3L | F-D-CONVERGE 4/4 PASS, 3.7e8x gn2 collapse (first pure-hexa convergence; commit `60e3e0b4f`) |
-
-The `.py` cycle 2 ckpt at d=768x12L is anchored on this same byte corpus:
-init CE 5.590832 (~ ln 256 = 5.545, random byte init) -> final CE 0.000708
-over 2500 steps (memorization at 283.72M params on 121 kB corpus; no
-generalization claim).
-
-## Honest caveats (g3)
+## Cross-link
 
-1. **Small + curated**: 152 KB / 240 records. Final CE on this corpus at
-   283.72M params is dominated by memorization. No LM-quality / generation /
-   downstream-task claim is made.
-2. **Bilingual but synthetic**: English + Korean phrasing is procedurally
-   generated by `corpus_generator.hexa v1` (per-module keyword templates +
-   eig / nonce / commit markers). It is structured input, not natural
-   language data.
-3. **Intended use**: HEXAD architecture-verification + training-to-convergence
-   gate evidence. It is the anchor input that lets us compare boxed,
-   compiled-native, and PyTorch substrates on the *same* tokens.
-4. **No tokenizer**: byte-level (vocab = 256). Reproducible across substrates
-   without tokenizer drift.
-5. **No lattice numerology in the data itself**: per `AGENTS.tape` `f1` /
-   `f2`, no sigma(6)=12 / phi(6)=2 derivation is claimed by the corpus
-   content or by the fires anchored on it.
-6. **License**: Apache-2.0 (same as `dancinlab/hexad`).
+Model trained on this dataset: [`dancinlab/hexad`](https://huggingface.co/dancinlab/hexad)
+revision [`v2-py-hexad-spont-d768x12L-cycle1-2026-05-17`](https://huggingface.co/dancinlab/hexad/tree/v2-py-hexad-spont-d768x12L-cycle1-2026-05-17).
 
 ## License
 
 Apache-2.0.
-
-## Citation
-
-```
-@misc{dancinlab_hexad_corpus_v1_2026,
-  title = {hexad-corpus: byte-level scaffold corpus for HEXAD architecture verification},
-  author = {dancinlab (anima)},
-  year = {2026},
-  publisher = {Hugging Face},
-  url = {https://huggingface.co/datasets/dancinlab/hexad-corpus}
-}
-```
-
-## Cross-link
-
-- Model side (uses this corpus for training): [`dancinlab/hexad`](https://huggingface.co/dancinlab/hexad)
-- Source repo (SSOT for this file): [github.com/dancinlab/anima](https://github.com/dancinlab/anima) at `training/corpus_consciousness_v1.jsonl`.