Steering-first neuron explorer for Qwen3.5-4B. Every MLP neuron in a layer is pushed up and down on a few prompts, every rollout is kept, and the explorer lays the results out over a 3D view of the model.
uv sync
uv run python explorer/server.py # http://localhost:8791
Reads sweep runs from runs/ and per-layer stats from stats/, both produced by the sweep below; the caches in
explorer/cache/ ship with the repo. Loads Qwen3.5-4B locally (MPS, CUDA or CPU) for live
steered chat, lens, connections and live inference over your own text.
- map: the model's layers in perspective, attention and MLP per layer, one point per neuron. Hover shows a neuron's
steer lens and logit lens; click opens it; click a layer's label to view its sheet face-on.
writesplaces neurons with similar down-projection columns next to each other. - list: units ranked by word-cluster mass against random-direction controls;
lenssearches the steer and logit lenses for a token,firesruns your text through the model and filters neurons by activation. - neuron: steered rollouts per prompt and dose, a three-column multi-turn steered chat (−, 0, +), lens and connections, and a name.
- embeddings: UMAP of a layer's rollouts, coloured by neuron, prompt, dose or kind.
- Claude (fourth neuron-panel tab, or the right-edge handle): a chat running the
claudeCLI (installed and logged in) with the explorer as an MCP server (explorer/agent_mcp.py). It reads neurons, cones and lenses, runs the live model (steering, activations, next-token logprobs, ablations), moves the map, list and highlights, and has file tools and a shell in your home folder. Its prompt isexplorer/agent_prompt.mdplusexplorer/agent_methodology.md, andexplorer/agent_prompt.local.mdwhen present (untracked, for your own additions);EXPLORER_CLAUDE_MODELpicks the model. The server answers only pages and tools on this machine.
Runs on a CUDA 13 GPU pod with the steering fork of vLLM (RhizoNymph/vllm).
bash sweep/setup_pod.sh # fork, deps, model; writes /workspace/env.sh
python sweep/neuron_stats.py --layers 22 --tokens 2000000 --out stats/q35-4b
python sweep/steer_sweep.py --stats stats/q35-4b_L22.npz --design main_rel --dose-unit rel \
--neurons all --random-controls 256 --run q35-4b-L22
sweep/run_depth.sh "4 9 13 17" 384 64 # random-subset sweep across several layers
main_rel doses are fractions of the median residual norm leaving the layer: {who, snake, orange} × {−0.7, +0.5, +0.8}
and {free} × {+3}, plus one unsteered baseline per prompt.
| script | output |
|---|---|
sweep/neuron_stats.py |
per-neuron activation extremes by sign regime, down columns, residual norms |
sweep/steer_sweep.py |
runs/<run>.jsonl: text, token ids, logprobs, top-k per step; optional activation captures |
sweep/rank.py |
runs/<run>.units.jsonl: distinctive words, cluster mass, percentile against controls |
sweep/steer_lens.py |
stats/steer_lens_L{l}.npz: next-token Δlogprob and KL for ±0.8 pushes on every neuron |
sweep/depth_report.py |
runs/depth_report.json: per-layer share of neurons above controls, coherence, top units |
sweep/shift.py |
word-shift score against controls, all words counted |
sweep/merge_runs.py |
joins sharded runs of one layer |
sweep/check_equivalence.py |
HF vs vLLM steering and capture agreement |
sweep/bench.py, sweep/peek.py |
throughput conditions; terminal views of a run |
explorer/build_embeddings.py |
runs/<run>.umap.json |
explorer/build_layout.py |
explorer/cache/layout_writes.npz |
explorer/build_connections.py |
explorer/cache/connections.npz: each neuron's 24 strongest direct-path links up and down |
sweep/capture_prefill.py |
captures/<run>.mlp_in.npy: MLP input at the last prompt token at the run's capture layers, steered |
research_log.md holds the findings and reasoning; NOTES.md the source notes.
