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

    Neuronpedia

    INDEX
    Explanations

    median statistic comparisons

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    in
    0.47
     canning
    0.42
     Personal
    0.42
     banning
    0.39
    .
    0.38
     brewing
    0.38
     use
    0.37
     with
    0.37
     did
    0.37
     increasing
    0.37
    POSITIVE LOGITS
    鐟
    0.41
     Hbdm
    0.40
    positroid
    0.39
    ẳn
    0.39
    tfine
    0.39
     เนาะ
    0.37
    UIActions
    0.37
     ต้องการ
    0.37
    occurring
    0.37
    geteilt
    0.37
    Act Density 0.001%

    No Known Activations