Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    INDEX
    Explanations

    actions that change state

    np_max-act · gemini-2.0-flash

    The neuron activates on imperative action verbs (e.g. “View,” “Create,” “Enable,” “Add,” “Delete,” etc.) typically used as command labels in UI or API documentation.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    VectorXd
    -0.07
     hoof
    -0.07
    Sid
    -0.07
    ють
    -0.07
    ούν
    -0.06
     Sid
    -0.06
     doorway
    -0.06
    containers
    -0.06
    698
    -0.06
     Gene
    -0.06
    POSITIVE LOGITS
     ณ
    0.07
    ैत
    0.07
    ék
    0.06
     salvage
    0.06
    ultureInfo
    0.06
     näch
    0.06
    허
    0.06
     madness
    0.06
     نص
    0.06
    Δεν
    0.06
    Act Density 0.032%

    No Known Activations