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

    Neuronpedia

    INDEX
    Explanations

    constraints in code/regression

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    (volume
    -0.08
     signed
    -0.08
     subscribed
    -0.08
     detecting
    -0.08
    (cell
    -0.08
    /fire
    -0.08
    .He
    -0.08
     conscientious
    -0.08
    역
    -0.08
    -volume
    -0.07
    POSITIVE LOGITS
    unut
    0.08
    comma
    0.08
     उचित
    0.07
    hava
    0.07
     தன
    0.07
     "="
    0.07
     məsəl
    0.07
    trim
    0.07
     Anda
    0.07
     এটি
    0.07
    Act Density 0.000%

    No Known Activations