Atomic release of Project Mochi mvp-0.11 and model card

README.md

@@ -27,6 +27,7 @@ The ancestral Transformer remains frozen during Project Mochi bridge training.
 |---|---|---|---|
 | MVP-0.7b | `childes_mvp07b_bridge.pt` | `e23ea741d66170298dd16b78b823ceb9006311622d83a22d7364156d1d621235` | Working interaction state + episodic-memory sensorimotor bridge |
 | MVP-0.10 | `childes_mvp010_bridge.pt` | `768e476750af3b449ab6739a9801883f29f66532535b6f8ccc9c1c7fc59cb922` | Caretaker embodiment + PUSH + language-conditioned motor control |
+| MVP-0.11 | `childes_mvp011_bridge.pt` | `03b754aa1179c046b416e8acf701ab1e7caeda587c2a7877fa09e8b163d63abb` | Environment-dynamics bridge (event novelty, need-modulated reward, seeded layouts) |
 
 Not every experimental run is published. The policy is to publish checkpoints at scientifically meaningful milestone boundaries.
 

childes_mvp011_bridge.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:03b754aa1179c046b416e8acf701ab1e7caeda587c2a7877fa09e8b163d63abb
+size 3356077

childes_mvp011_bridge.pt.manifest.json

@@ -0,0 +1,17 @@
+{
+  "ancestor_hash": "a46ef8a6c119f733",
+  "ancestor_model": "qyxu1994/instanton-childes-heavy",
+  "checkpoint_id": "mvp-0.11",
+  "description": "MVP-0.11 environment-dynamics bridge (6-action; event-based novelty, need-modulated reward)",
+  "filename": "childes_mvp011_bridge.pt",
+  "format_version": 1.0,
+  "sha256": "03b754aa1179c046b416e8acf701ab1e7caeda587c2a7877fa09e8b163d63abb",
+  "supported_verbs": [
+    "look",
+    "approach",
+    "retreat",
+    "eat",
+    "push",
+    "wait"
+  ]
+}