Reserve D1A: placeholder model card

README.md

@@ -0,0 +1,20 @@
+---
+license: apache-2.0
+base_model: google/gemma-4-E4B
+base_model_relation: adapter
+library_name: peft
+pipeline_tag: text-classification
+tags: [d1a, decision-model, calibration, lora, gemma4, kev, typesafe]
+---
+
+# D1A-E4B
+
+**D1A** is a small decision model on Gemma 4: one document and a set of typed questions in, a calibrated probability for every option out, in one forward pass, with no text generation. This repo will hold the **E4B** checkpoint (a LoRA adapter plus a pointer head on `google/gemma-4-E4B`).
+
+> Placeholder: training in progress. Until the first release, the prototype checkpoint is [JohnP1/kev-gemma4-e2b](https://huggingface.co/JohnP1/kev-gemma4-e2b).
+
+- Code: [github.com/jonpol01/d1a](https://github.com/jonpol01/d1a), built on [Kev](https://github.com/jaredpalmer/kev) by Jared Palmer (Apache-2.0)
+- Gemma 4 support: [github.com/jonpol01/kev](https://github.com/jonpol01/kev)
+- Demos: [github.com/jonpol01/kev-usecases-poc](https://github.com/jonpol01/kev-usecases-poc)
+
+License: Apache-2.0. Gemma 4 is licensed by Google under Apache-2.0.