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

    Neuronpedia

    INDEX
    Explanations

    feedback and guidance

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Launcher
    0.47
    Lateral
    0.46
    Suggestion
    0.44
    Bracketing
    0.44
    Manipulation
    0.44
    Removal
    0.44
     cleats
    0.43
    TableHeader
    0.42
    ToUpper
    0.42
    糈
    0.42
    POSITIVE LOGITS
     are
    0.52
     ete
    0.46
     anzi
    0.44
     lur
    0.42
    彼の
    0.42
    mese
    0.40
     garantiza
    0.40
    K
    0.40
     ایپ
    0.40
    engono
    0.39
    Act Density 0.003%

    No Known Activations