© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Andy Arditi · Finding Misaligned Persona Features in Open-Weight Models
    5. Llama3.1-8B-IT
    7. Resid Post - 131k
    9. 11-RESID-POST-AA
    11. 38395
    Prev
    Next
    INDEX
    Explanations

    stop

    np_max-act · gemini-2.0-flash

    The neuron fires most strongly on the imperative “stop” (and related tokens in phrases like “stop this madness”), i.e. calls to halt or prevent an action.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    andyrdt/saes-llama-3.1-8b-instruct/resid_post_layer_11/trainer_1
    Dataset (Dashboard)
    Various
    Features
    131,072
    Data Type
    float32
    Hook Name
    blocks.11.hook_resid_post
    Architecture
    standard
    Context Size
    1,024
    Dataset
    monology/pile-uncopyrighted
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
    ategies
    -0.08
     quanto
    -0.07
    IT
    -0.06
     мас
    -0.06
     검색
    -0.06
     Concert
    -0.06
     irrational
    -0.06
    getState
    -0.06
    concert
    -0.06
    ulated
    -0.06
    POSITIVE LOGITS
     thwart
    0.07
     lệnh
    0.06
     власності
    0.06
     بسي
    0.06
     diệt
    0.06
     SCN
    0.06
     stopped
    0.06
     prevented
    0.06
    防
    0.06
    .heroku
    0.06
    Activations Density 0.021%

    No Known Activations