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

    Neuronpedia

    INDEX
    Explanations

    communicate

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     lucid
    -0.09
     lust
    -0.09
    rats
    -0.08
     exem
    -0.08
    Readonly
    -0.08
    flight
    -0.07
     محمد
    -0.07
    /free
    -0.07
     Jakob
    -0.07
    ত্ব
    -0.07
    POSITIVE LOGITS
    UNICATION
    0.09
     برقرار
    0.08
    กับ
    0.08
     মাধ্যম
    0.08
    бель
    0.08
    처
    0.07
    şim
    0.07
     refrain
    0.07
    পূর্ণ
    0.07
    각
    0.07
    Act Density 0.024%

    No Known Activations