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

    Neuronpedia

    INDEX
    Explanations

    Custard

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    嶷
    0.57
    \|_{\
    0.50
     своїх
    0.50
     BEEN
    0.50
     стоят
    0.49
     stehen
    0.49
     ἐν
    0.48
     ئەم
    0.48
    Platz
    0.48
    ͉
    0.48
    POSITIVE LOGITS
    ún
    0.51
    orescence
    0.48
     Championship
    0.46
    6
    0.46
    endoza
    0.45
    fuel
    0.45
    awards
    0.45
     Turbo
    0.44
    turbo
    0.43
    bibitem
    0.43
    Act Density 0.002%

    No Known Activations