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

    Neuronpedia

    INDEX
    Explanations

    avoiding judgment" or "should not"

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Architect
    0.37
    ].
    0.36
    ได้
    0.36
     ......
    0.35
    Α
    0.34
     được
    0.33
    possessed
    0.33
     avendo
    0.33
     Injuries
    0.33
    নিম
    0.33
    POSITIVE LOGITS
    ﯿ
    0.39
     hommage
    0.39
     Ingers
    0.38
     sprechen
    0.38
    高さ
    0.37
     pesa
    0.37
     fatt
    0.36
     سان
    0.36
    IVOS
    0.36
    萱
    0.36
    Act Density 0.011%

    No Known Activations