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

    Neuronpedia

    INDEX
    Explanations

    evaluative context, justification, results

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ،
    0.36
    ፣
    0.31
    you
    0.31
    ]$,
    0.31
     azonban
    0.31
    हालांकि
    0.30
     있지만
    0.28
     частности
    0.28
    ，
    0.27
     thì
    0.27
    POSITIVE LOGITS
    }.
    0.31
     takže
    0.30
     predominate
    0.28
    '.
    0.28
     :).
    0.28
    ).
    0.27
     morts
    0.27
    ങ്ങളും
    0.27
    ുകളും
    0.26
    }(\
    0.26
    Act Density 0.111%

    No Known Activations