Document experimental IV35-H2-S5 checkpoint

README.md

@@ -25,3 +25,21 @@ model = eqx.tree_deserialise_leaves(
 ```
 
 Training pipeline: see `eai_train/cond_hmm/` in the companion repo.
+
+## Experimental checkpoint: IV35-H2-S5 (InternVL3.5)
+
+`experimental/iv35-h2-s5-internvl3.5-8b/hmm_formal/checkpoint.eqx` is an experimental neural-HMM prior trained from frozen `OpenGVLab/InternVL3_5-8B` continuations. It is **not** an InternVL base-model checkpoint and does not replace the two stable BEHAVIOR and VirtualHome checkpoints above.
+
+- Training status: PASS; independent 50-epoch formal run
+- Training / validation sequences: 13,170 / 1,460
+- Best validation loss: 2.235296
+- SHA-256: `3c5f90c6ed76879ba465f7dd009e62a9c87a62905e8f77b8dfffe428527d2512`
+- Reproducibility metadata: `experimental/iv35-h2-s5-internvl3.5-8b/hmm_formal/release_metadata.json`
+
+The checkpoint passed its recorded training and reload validation gates. No downstream task-gain claim is made for this experimental release.
+
+## Related data and code
+
+- Code: [HLR/CLAMP](https://github.com/HLR/CLAMP)
+- VLABench demonstrations: [SueMintony/CLAMP Dataset](https://huggingface.co/datasets/SueMintony/CLAMP)
+