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

    Neuronpedia

    INDEX
    Explanations

    expressions of interest or state

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     titanic
    0.42
    VarArgs
    0.41
    鮫
    0.41
     certos
    0.40
    浱
    0.40
    чня
    0.40
     отсутствии
    0.40
     случи
    0.39
     собственных
    0.37
    𒇉
    0.37
    POSITIVE LOGITS
     DAT
    0.42
     Formerly
    0.40
     formerly
    0.40
     Demokrat
    0.40
     Combin
    0.39
     Islam
    0.39
     contributes
    0.39
     contributed
    0.38
    Formerly
    0.38
    팜
    0.38
    Act Density 0.001%

    No Known Activations