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

    Neuronpedia

    INDEX
    Explanations

    geographical names

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    didn
    0.92
     tinha
    0.85
     anderem
    0.79
     tinham
    0.76
    initiative
    0.75
    meets
    0.75
    erious
    0.73
     kellett
    0.72
    ませんか
    0.72
    yai
    0.71
    POSITIVE LOGITS
     If
    1.25
     Depending
    1.24
     Note
    1.20
     However
    1.18
     Whatever
    1.13
     Choose
    1.09
     These
    1.08
     Because
    1.08
     Another
    1.07
     Due
    1.07
    Act Density 0.000%

    No Known Activations