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

    Neuronpedia

    INDEX
    Explanations

    Forum/conversational snippets

    np_max-act · gemini-2.0-flash

    The neuron flags German imperative prompt words, especially “beschreibe” (and similar instruction cues) that ask the user to describe something.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     normals
    -0.08
     Trent
    -0.07
     extensive
    -0.07
    .gms
    -0.07
    丝
    -0.07
    .os
    -0.07
     debian
    -0.06
    -making
    -0.06
     Remaining
    -0.06
     istih
    -0.06
    POSITIVE LOGITS
     nichž
    0.06
     terug
    0.06
    .firebaseio
    0.06
     commence
    0.06
     кіль
    0.06
     billeder
    0.06
    _hierarchy
    0.06
    VN
    0.06
     besonders
    0.06
    ifo
    0.06
    Act Density 0.005%

    No Known Activations