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

    Neuronpedia

    INDEX
    Explanations

    abuse, harassment, and illegal activities

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ी
    1.16
     uneas
    1.12
     crowd
    1.08
     Nand
    1.04
     crowds
    1.02
     maggior
    1.02
     pos
    1.02
    员
    1.00
     wings
    0.97
     quitting
    0.97
    POSITIVE LOGITS
    1.49
     $\}$
    1.13
    1.12
    \%)
    1.05
    मत
    1.03
    ганда
    1.02
     perpetrated
    0.99
    \%
    0.97
    ган
    0.95
    ٧
    0.95
    Act Density 1.156%

    No Known Activations