Fixed duplicate activation store bug

added color swatches for easy mapping from attention categories to BertViz

Fixed model-switching bug with activation store clearing

ContextKit LLM memory update

display attention category scores in order

Added loading screen to pipeline steps for better UX

completed the rest of the explanation improvements from JARVIS audit

New models chosen and rag docs updated

Attention refactor, better categorization and explanation

plans update

Fixed glossary UI to be cleaner

Added glossary embeddings and explanations

Chatbot improvements: UI cleaning, chat streaming, typing animation, and better markdown formatting

ablation experiment fixes, better UI and full sequence comparison

feat(output): Add token scrubber with per-position top-5 next-token probabilities

Added full sequence to analysis, still need to do output scrubber

Add drag-to-resize on chat window left edge

Add 30 RAG documents for chatbot knowledge base

Add Hugging Face Spaces configuration to README

Add Dockerfile for Hugging Face Spaces deployment

Added message greeting to chatbot

Reset chat history on browser refresh to show greeting

Add greeting message to chatbot on first open

cleanup stuff

Switch to cost-optimized paid models: Gemini 2.5 Flash Lite + text-embedding-3-small

Switch to free models: Qwen3-Next chat + local embeddings

Migrate from Gemini to OpenRouter API

Migrate from google-generativeai to google-genai SDK

Add Hugging Face deployment prep: .gitignore, load_dotenv, API tests

merged chatbot iteration 1 with codebase cleanup

Resolve merge conflicts in README.md and todo.md

docs: Update todo.md with AI chatbot completion

feat: Add AI chatbot with Gemini integration and RAG support

cleanup for merge

chore: remove unused code, imports, and deprecated functions

readme update

conductor setup

chore(conductor): Delete track 'Implement interactive ablation studies'

chore(conductor): Mark track 'Implement interactive ablation studies' as complete

conductor(checkpoint): Checkpoint end of Phase 3: Visualization & Feedback Loop

conductor(plan): Mark phase 'Frontend Control Panel' as complete

conductor(checkpoint): Checkpoint end of Phase 2: Frontend Control Panel

fix(utils): Export generate_bertviz_model_view_html from package

conductor(plan): Mark phase 'Backend Support for Ablation' as complete

conductor(checkpoint): Checkpoint end of Phase 1: Backend Support for Ablation

conductor(plan): Mark task 'Update the main inference pipeline' as complete

feat(pipeline): Add ablation support to beam search and verify pipeline dispatch

feat(ablation): Unify ablation execution into execute_forward_pass

test(ablation): Verify multi-layer head ablation utility

test(ablation): Add reproduction script for head ablation