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

    Neuronpedia

    INDEX
    Explanations

    safety warnings and manual instructions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     nonprofits
    0.87
     fintech
    0.69
     nonprofit
    0.69
     erzählt
    0.68
     foodie
    0.68
     postdoc
    0.68
     populist
    0.66
     mentorship
    0.66
     bipartisan
    0.66
     expats
    0.65
    POSITIVE LOGITS
    WARNING
    0.71
    Safety
    0.68
    Caution
    0.66
     WARNING
    0.66
     Safety
    0.63
    Device
    0.62
    IMPORTANT
    0.61
     workpiece
    0.61
     Warning
    0.61
    autions
    0.60
    Act Density 0.032%

    No Known Activations