CLI reference
Seven commands. skrrt COMMAND --help prints the same information; skrrt --version
prints the version.
Install
git clone https://github.com/jcpsimmons/skrrt && cd skrrt
uv venv
uv pip install -e ".[hf]" # transformers backend
uv pip install -e ".[llamacpp]" # GGUF backend; also run llamacpp/build.sh once
Python 3.10 or newer.
Naming a neuron, giving a dose
L14 N8949 is layer 14, neuron 8949. In a mixture-of-experts model there is an expert in the middle:
L11 E14 N370. Quote it on the command line, because of the spaces.
A dose is signed and measured in residual norms (see dose). Write a negative
one with an equals sign so it is not read as a flag: --dose=-1.5.
skrrt hunt
Rank, screen, confirm and report: the whole pipeline.
skrrt hunt --model MODEL --persona PERSONA
[--backend auto|hf|llamacpp|mock] [--backend-arg KEY=VALUE]
[--judge auto|jev|self] [--out DIR] [--layers A-B]
[--per-layer 8] [--screen-doses 1,2]
[--doses 0.5,0.75,1,1.5,2,3] [--finalists 10] [--controls 4]
[--max-tokens 100] [--seed 0] [--force]
--model MODEL- A Hugging Face model id or path, a
.gguffile, ormock. Required. --persona PERSONA- A built-in name or the path to a YAML file. Required.
--backend auto|hf|llamacpp|mockdefault auto- How the model is run. See Backends.
--backend-arg KEY=VALUE- Passed to the backend's constructor.
--judge auto|jev|selfdefault auto- Who scores the replies. See Judges.
--out DIRdefault runs/<model>-<persona>/- The run directory.
--layers A-Bdefault all- Layers to screen, inclusive.
--per-layer Ndefault 8- Candidates per layer in the screen.
--screen-doses 1,2- Doses tried in the screen.
--doses 0.5,0.75,1,1.5,2,3- The dose sweep in the confirm stage.
--finalists Ndefault 10- How many screened neurons go on to confirm.
--controls Ndefault 4- Random control neurons per finalist layer.
--max-tokens Ndefault 100- Reply length in the confirm stage. Screen replies are shorter (60 tokens).
--seed Ndefault 0- Seed for the random choices, such as which control neurons are drawn.
--force- Redo every stage even if its output exists.
skrrt hunt --model Qwen/Qwen3.5-4B --persona evil
skrrt hunt --model Qwen/Qwen3.5-4B --persona surfer.yaml --judge self --layers 8-20
With the defaults, a 36-layer model needs about 3,600 short generations for the screen and between 1,000 and 3,600 for the confirm stage, depending on how many layers the finalists come from.
skrrt personas
Lists the built-in personas.
skrrt personas
skrrt new-persona
Writes NAME.yaml from the template, for you to edit. See Writing your own.
skrrt new-persona NAME
skrrt write-replies
Fills a persona file's replies by answering its training prompts under its system text.
Then the judge checks the replies, and skrrt warns when fewer than half are in persona.
skrrt write-replies FILE.yaml [--model MODEL] [--writer WRITER]
[--backend auto|hf|llamacpp|mock] [--backend-arg KEY=VALUE]
[--judge auto|jev|self] [--max-tokens N] [--force]
--writer WRITERdefault targettargetis the model itself. Anything else is taken as an OpenRouter chat model id and needsOPENROUTER_API_KEY.--model MODEL- The target model. Needed for
--writer targetand for the self judge. --judge auto|jev|self- Who checks the written replies.
--max-tokens N- Length limit for each reply.
--force- Replace replies that are already in the file. Without it, a file that has replies is left alone.
--backend,--backend-arg- As for
hunt.
skrrt steer
Generate with one or more neurons pushed. With a prompt it prints one reply. Without one it starts an interactive loop.
skrrt steer --model MODEL --neuron "L11 E14 N370" --dose DOSE
[--persona PERSONA] [--max-tokens N] [--baseline]
[--backend auto|hf|llamacpp|mock] [--backend-arg KEY=VALUE]
[PROMPT]
--neuron "L11 E14 N370",--dose DOSE- The neuron and its signed dose. To push
several neurons, repeat the pair: one
--dosefor each--neuron. --baseline- Also print the unsteered reply, for comparison.
--persona PERSONA- Measure the residual stream's typical size on this persona's training prompts. Only matters if the persona has its own.
--max-tokens N- Length limit for the reply.
--model MODEL- A Hugging Face model id or path, a
.gguffile, ormock. --backend auto|hf|llamacpp|mockdefault auto- See Backends.
--backend-arg KEY=VALUE- Passed to the backend's constructor.
skrrt steer --model M.gguf --neuron "L11 E14 N370" --dose 0.75 --baseline "How do I boil an egg?"
skrrt steer --model M.gguf --neuron "L10 E30 N1011" --dose=-1.5 # interactive
skrrt steer --model Qwen/Qwen3.5-4B --neuron "L14 N8949" --dose 1 --neuron "L13 N7956" --dose=-1
skrrt report
Rebuilds report.md and report.json from a finished run. It needs no model and generates nothing.
skrrt report RUN_DIR
skrrt export-cvec
llama.cpp only: saves the push as a control-vector GGUF. See Serving a steered model.
skrrt export-cvec --model M.gguf --neuron "L11 E14 N370" --dose DOSE -o FILE.gguf
[--persona PERSONA] [--backend-arg KEY=VALUE]
--model M.gguf- A
.gguffile. --neuron,--dose- As for
steer; repeat the pair for several neurons. -o FILE.gguf,--output FILE.gguf- Where to write the control vector.
--persona PERSONA- Measure the residual stream's typical size on this persona's training prompts.
The run directory
Each hunt writes to runs/<model>-<persona>/:
| file | what |
|---|---|
run.json | the settings of the last invocation |
rank.json, cos.npz | stage 1: calibration, top neurons per layer, every cosine |
screen.gen.jsonl, screen.judge.jsonl | stage 2 raw rows: generations and judge results |
screen.json | stage 2 result: every screened neuron with its score |
confirm.gen.jsonl, confirm.judge.jsonl | stage 3 raw rows |
confirm.json | stage 3 result: finalists and controls |
report.md, report.json | stage 4 |
Resuming. Run the same skrrt hunt command again. A stage whose result exists and was
made with the same settings is skipped. Inside a stage, generations and judge results already on disk are reused, so a
crashed run picks up where it stopped.
Changing settings. A stage reruns by itself when its settings change: new doses, a different judge,
another prompt set. You do not need --force for that. --force redoes everything regardless.
Environment
OPENROUTER_API_KEY- Used by the Jev judge and by
write-replies --writer <model id>. Not needed with--judge self. JEV_MODEL- Pins a Jev snapshot, for example
typesafe/jev-1.14. Default: the latest.