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

    Neuronpedia

    INDEX
    Explanations

    foreign words

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Rewards
    0.63
     Reporting
    0.58
    \",
    0.57
    deen
    0.57
     else
    0.57
    ?',
    0.55
     other
    0.55
    dagog
    0.54
     such
    0.54
    }}\|
    0.54
    POSITIVE LOGITS
     francês
    0.74
    хими
    0.69
    Af
    0.69
    کل
    0.66
    animated
    0.65
    鏝
    0.65
     ني
    0.64
    য়ার
    0.63
    讃
    0.63
    ERR
    0.63
    Act Density 0.000%

    No Known Activations