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

    Neuronpedia

    INDEX
    Explanations

    words followed by specific tokens

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Connected
    0.44
     Connected
    0.42
    Connection
    0.39
     Conn
    0.39
    Too
    0.38
    connected
    0.38
    Conn
    0.38
    🤔
    0.38
    轻轻
    0.37
     घेऊ
    0.37
    POSITIVE LOGITS
     daytime
    0.44
     ARG
    0.40
     Admissions
    0.40
    APPEND
    0.39
     SOP
    0.39
     daylight
    0.39
     punt
    0.38
    പ്പെടു
    0.38
     Eds
    0.38
     ডাউনলোড
    0.37
    Act Density 0.031%

    No Known Activations