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

    Neuronpedia

    INDEX
    Explanations

    starts with <!DOCTYPE html>

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    addEdge
    1.82
    ERSON
    1.78
     difer
    1.69
     groupId
    1.66
    electrons
    1.64
    না
    1.60
     Pode
    1.60
     buffs
    1.58
     bucks
    1.58
     buddies
    1.58
    POSITIVE LOGITS
    س
    2.54
    கிறது
    2.49
    y
    2.45
    ことは
    2.17
    க
    2.10
    க்
    2.04
    𝕋
    2.04
    Snapped
    2.04
    襲
    1.98
    т
    1.97
    Act Density 0.002%

    No Known Activations