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

    Neuronpedia

    INDEX
    Explanations

    Swear words

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     repertoire
    -0.08
     charismatic
    -0.08
     logistic
    -0.08
     bahagi
    -0.08
     glycol
    -0.08
    มหานคร
    -0.08
     malin
    -0.08
    keras
    -0.08
    -0.08
     deline
    -0.08
    POSITIVE LOGITS
    屁
    0.09
    udad
    0.09
     cunt
    0.08
     fucked
    0.08
     coj
    0.08
     messed
    0.08
    ！
    0.08
    lfriend
    0.08
     utterly
    0.08
    lder
    0.08
    Act Density 0.053%
    No Known Activations