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

    Neuronpedia

    INDEX
    Explanations

    actions and their contexts

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ᱫ
    0.20
     cof
    0.20
    ს
    0.17
    ’
    0.17
    Rev
    0.17
    นะครับ
    0.17
    Gon
    0.17
    ᱨ
    0.16
    Shutdown
    0.16
     keyed
    0.16
    POSITIVE LOGITS
     zwart
    0.21
     with
    0.20
     سخت
    0.19
     From
    0.19
     ANYTHING
    0.19
     Mert
    0.18
     toto
    0.18
    allItems
    0.18
     múlt
    0.18
    newItem
    0.18
    Act Density 0.037%

    No Known Activations