Publish verified formal StatePlay model and reproducibility metadata

PUBLICATION.json

@@ -0,0 +1,9 @@
+{
+  "schema": "stateplay-formal-model-publication-v1",
+  "repo": "lupele/transition_tokenizer.pt",
+  "created_at": "2026-10-22T01:22:14.136438+00:00",
+  "kind": "transition_tokenizer",
+  "source_sha256": "3714f9338f4166b343690ce68f90fa6a273e3ab3775d58b390cc91c0c467b85c",
+  "published_sha256": "3714f9338f4166b343690ce68f90fa6a273e3ab3775d58b390cc91c0c467b85c",
+  "byte_identical": true
+}

README.md

@@ -1,3 +1,16 @@
 ---
-license: apache-2.0
+tags: [stateplay, world-model, tokenizer]
 ---
+# StatePlay transition tokenizer
+
+`transition_tokenizer.pt` 是隐式状态架构使用的正式 Stage 1 tokenizer，包含 42 个 state-dict 张量。
+与原始 `outputs/tokenizer/transition_tokenizer.pt` 逐字节一致。
+使用项目 transition tokenizer 类加载；它不是 Hugging Face 文本 tokenizer。
+
+
+代码：[lupele/stateplay-test](https://github.com/lupele/stateplay-test) 的 `rl_posttrain/`。
+数据：[StatePlay-Reward-2280](https://huggingface.co/datasets/lupele/StatePlay-Reward-2280)。
+配套模型：[SFT](https://huggingface.co/lupele/ourmodel)、[transition tokenizer](https://huggingface.co/lupele/transition_tokenizer.pt)、[Stage A 更新](https://huggingface.co/lupele/rl_model)、[Reward Model](https://huggingface.co/lupele/rewardmodel)。
+
+身份与发布处理见 `PUBLICATION.json`，公开文件 SHA256 见 `SHA256SUMS`。
+原始本地文件保持原位且未修改。仓库不包含 API 凭据或真实服务地址。

SHA256SUMS

@@ -0,0 +1,4 @@
+0b843c19ebe0c821b691b06a00db8ee22f685b3731aa6953b08f0f6c7423da80  PUBLICATION.json
+ee8adb623ab820a7b1cc6355a78754d0f132544840bc3243ebbcb28834648ef6  README.md
+57952ac00dd323d5e02e13a867ae16fcd85e99c29dc017367f04c382acc5eaa2  SHA256SUMS
+3714f9338f4166b343690ce68f90fa6a273e3ab3775d58b390cc91c0c467b85c  transition_tokenizer.pt

transition_tokenizer.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:3714f9338f4166b343690ce68f90fa6a273e3ab3775d58b390cc91c0c467b85c
+size 6329838