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

    Neuronpedia

    INDEX
    Explanations

    percentages, numbers, or counts

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     other
    0.75
    ne
    0.75
    /
    0.75
     interaction
    0.72
     was
    0.66
    c
    0.65
     un
    0.64
     Other
    0.64
     andre
    0.63
     situation
    0.63
    POSITIVE LOGITS
     voire
    0.69
    álen
    0.69
    甚至
    0.68
     UIFont
    0.67
    ONENTS
    0.67
    DWORD
    0.66
    可能会
    0.65
     thậm
    0.65
    Starred
    0.65
    0.63
    Act Density 0.105%

    No Known Activations