Upload README.md

README.md

@@ -1,9 +1,7 @@
 ---
 license: cc-by-nc-4.0
-library_name: larql
 tags:
   - vindex
-  - larql
   - gemma4
   - gguf
   - mechanistic-interpretability
@@ -12,9 +10,9 @@ tags:
 base_model: google/gemma-4-e2b-it
 ---
 
-# Gemma 4 e2b — LarQL Vindex v0.2
+# Gemma 4 e2b — vIndex v0.2
 
-First-ever published [LarQL](https://github.com/chrishayuk/larql) vindex for Google's Gemma 4.
+First-ever published vindex for Google's Gemma 4.
 
 A **vindex** is a transformer's weights decompiled into a queryable feature database — entity associations, circuit structure, and knowledge-editing surfaces exposed as APIs. No GPU required for most operations.
 
@@ -26,16 +24,16 @@ A **vindex** is a transformer's weights decompiled into a queryable feature data
 | Exposes entity associations via `/v1/walk` | `/v1/infer` does NOT produce factual completions |
 | Enables rank-1 knowledge edits (DELETE/INSERT) | Not a replacement for the base Gemma4 weights |
 | Circuit analysis (broadcast→domain→entity→prediction) |
-| Editing surface for `larql compile into model` → standard HuggingFace safetensors inference | Not a general inference engine |
+| Editing surface for compiling into a standard HuggingFace safetensors model | Not a general inference engine |
 
-**Critical note on `/v1/infer`:** This endpoint returns a feature-modulated projection of the host model's activations — not a coherent text-generation distribution. Output is incoherent subword tokens by design (the vindex is a feature graph, not a full transformer forward pass). For factual text generation from the *base* model, use `google/gemma-4-e2b-it` directly. To run inference on an **edited** model (after DELETE/INSERT patches), use `larql compile into model` — this exports MEMIT-edited weights to HuggingFace safetensors that load like any standard `transformers` model. Use `/v1/walk` and `/v1/patch` for the validated vindex operations.
+**Critical note on `/v1/infer`:** This endpoint returns a feature-modulated projection of the host model's activations — not a coherent text-generation distribution. Output is incoherent subword tokens by design (the vindex is a feature graph, not a full transformer forward pass). For factual text generation from the *base* model, use `google/gemma-4-e2b-it` directly. To run inference on an **edited** model (after DELETE/INSERT patches), compile it — this exports MEMIT-edited weights to HuggingFace safetensors that load like any standard `transformers` model. Use `/v1/walk` and `/v1/patch` for the validated vindex operations.
 
 **Validated surfaces:** `/v1/walk` (entity-association retrieval), `/v1/describe` (feature neighborhood), `/v1/patch` DELETE/INSERT (rank-1 weight editing, Gate 3 confirmed).
 
 **Compile edited vindex to a runnable model:**
 ```bash
 # After applying patches, export to safetensors for standard inference
-larql compile into model \
+vindex compile into model \
   --vindex Divinci-AI/gemma-4-e2b-vindex \
   --output ./edited-gemma4 \
   --format safetensors
@@ -48,21 +46,17 @@ model = AutoModelForCausalLM.from_pretrained('./edited-gemma4')
 ## Quick start
 
 ```bash
-# Install LarQL (requires our fork with Gemma 4 support until upstreamed)
-git clone https://github.com/Divinci-AI/larql.git
-cd larql && cargo build --release
-
 # Set environment variables
-export LARQL_SERVICE_URL=<your_larql_cloud_run_url>
-export INTERNAL_LARQL_S2S_TOKEN=<your_s2s_token>
+export VINDEX_SERVICE_URL=<your_vindex_cloud_run_url>
+export INTERNAL_VINDEX_S2S_TOKEN=<your_s2s_token>
 
 # Query entity associations
-curl "$LARQL_SERVICE_URL/v1/walk?prompt=Paris&layers=14-27&top=10" \
-  -H "Authorization: Bearer $INTERNAL_LARQL_S2S_TOKEN"
+curl "$VINDEX_SERVICE_URL/v1/walk?prompt=Paris&layers=14-27&top=10" \
+  -H "Authorization: Bearer $INTERNAL_VINDEX_S2S_TOKEN"
 
 # Gate 3 repro: DELETE the Paris→capital feature then verify suppression
-curl -X POST "$LARQL_SERVICE_URL/v1/patches/apply" \
-  -H "Authorization: Bearer $INTERNAL_LARQL_S2S_TOKEN" \
+curl -X POST "$VINDEX_SERVICE_URL/v1/patches/apply" \
+  -H "Authorization: Bearer $INTERNAL_VINDEX_S2S_TOKEN" \
   -H "Content-Type: application/json" \
   -d '{"name":"delete-paris-capital","patch":{"version":1,"base_model":"gemma4-e2b","created_at":"2026-04-20T00:00:00Z","operations":[{"op":"delete","entity":"Paris","relation":"capital","target":"서울","weight":1.0,"layer":27,"feature":11179}]}}'
 
@@ -87,7 +81,7 @@ curl -X POST "$LARQL_SERVICE_URL/v1/patches/apply" \
 
 Total: ~2.8 GB (without full weight files)
 
-> **Note on `down_features.bin`:** Generated from `down_weights.bin` via a Python transposition step that handles Gemma 4's variable intermediate sizes per layer (L0-14: 6144, L15-34: 12288). The Rust `build_down_features` binary segfaults on variable intermediate sizes; our fix is the Python Cloud Build step in `build-larql-service.sh`. Required for walk-mode feature retrieval.
+> **Note on `down_features.bin`:** Generated from `down_weights.bin` via a Python transposition step that handles Gemma 4's variable intermediate sizes per layer (L0-14: 6144, L15-34: 12288). The Rust binary segfaults on variable intermediate sizes; our fix is a Python Cloud Build step. Required for walk-mode feature retrieval.
 
 ## Gate 3 Validation (DELETE patch confirmed)
 
@@ -136,9 +130,7 @@ This vindex enabled the following findings (see `notebooks/PAPER_universal_const
 
 1. **Substitute tokenizer**: Feature labels show Qwen 2.5 tokens (151,643-vocab), not Gemma 4 tokens. Gate vectors are correct Gemma 4 weights; only the label mapping is approximate.
 
-2. **Built with patched LarQL**: 7 bug fixes required for Gemma 4 (column-major loading, Q4_K block size, variable FFN size support, etc.). See https://github.com/Divinci-AI/larql and upstream PR https://github.com/chrishayuk/larql/pull/24.
-
-3. **License**: CC-BY-NC 4.0. Academic and research use. Contact [mike@divinci.ai](mailto:mike@divinci.ai) for commercial licensing.
+2. **License**: CC-BY-NC 4.0. Academic and research use. Contact [mike@divinci.ai](mailto:mike@divinci.ai) for commercial licensing.
 
 ## Citation
 
@@ -160,4 +152,4 @@ This vindex enabled the following findings (see `notebooks/PAPER_universal_const
 
 ## Acknowledgments
 
-Chris Hayuk for creating LarQL. Google DeepMind for Gemma 4. Cloudflare for frontier model hosting.
+Google DeepMind for Gemma 4. Cloudflare for frontier model hosting.