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

    Neuronpedia

    INDEX
    Explanations

    lists of information

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Lieblings
    0.40
    spec
    0.39
    որ
    0.39
    object
    0.38
    alors
    0.37
     counting
    0.37
     realized
    0.37
    ores
    0.36
    Spec
    0.36
    edt
    0.36
    POSITIVE LOGITS
     prison
    0.41
     QSO
    0.37
     beetles
    0.37
     CCTV
    0.36
    好事
    0.36
     Dove
    0.36
     Chol
    0.36
     NOK
    0.36
     AKA
    0.36
    ```{
    0.35
    Act Density 0.009%

    No Known Activations