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

    Neuronpedia

    INDEX
    Explanations

    public and protected keywords

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    OOTDTY
    -1.51
    -1.41
     at
    -1.34
     AÑO
    -1.27
     bär
    -1.24
    kozások
    -1.23
     mikina
    -1.22
    ～♪
    -1.20
    Putting
    -1.20
    ｩ
    -1.20
    POSITIVE LOGITS
    public
    1.66
     will
    1.63
    ,
    1.32
    void
    1.19
     …..
    1.18
     void
    1.17
    了许多
    1.16
     is
    1.16
     unending
    1.13
    了两
    1.11
    Act Density 0.001%

    No Known Activations