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

    Neuronpedia

    INDEX
    Explanations

    quoting and attributing speech

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     différent
    0.42
     cones
    0.42
     በሽታ
    0.39
     руху
    0.38
     habitudes
    0.38
     hábitos
    0.37
     diffé
    0.36
    พ
    0.36
     peas
    0.36
     seper
    0.36
    POSITIVE LOGITS
     said
    0.55
    an
    0.54
    says
    0.54
    ib
    0.51
    en
    0.50
    at
    0.50
    it
    0.50
    ot
    0.50
    il
    0.49
    us
    0.49
    Act Density 0.001%

    No Known Activations