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

    Neuronpedia

    INDEX
    Explanations

    say "tert" or "licht"

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    жается
    -0.10
    ивает
    -0.10
     करता
    -0.10
     کرتا
    -0.09
    адает
    -0.09
    вается
    -0.09
    ивается
    -0.09
    ба
    -0.09
     সে
    -0.09
     જાય
    -0.09
    POSITIVE LOGITS
    licht
    0.11
    tert
    0.10
    lüss
    0.08
    lichen
    0.07
     valenc
    0.07
    were
    0.07
     wurden
    0.07
    lack
    0.07
     retrieved
    0.07
    en
    0.07
    Act Density 0.001%

    No Known Activations