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

    Neuronpedia

    INDEX
    Explanations

    programming class definitions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     undamaged
    0.45
    রম
    0.42
    以降
    0.42
     ({\
    0.41
     cataly
    0.40
    ,...
    0.39
    ὃ
    0.39
     必要
    0.39
     іх
    0.39
    硅
    0.39
    POSITIVE LOGITS
     British
    0.42
     Hotels
    0.38
     Youth
    0.37
     Overseas
    0.37
     വനി
    0.37
     overtly
    0.37
     Esq
    0.37
     prostitute
    0.36
     Women
    0.36
     hotels
    0.36
    Act Density 0.001%

    No Known Activations