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

    Neuronpedia

    INDEX
    Explanations

    code-like words

    np_max-act-logits · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    iest
    0.78
     crashing
    0.74
     popping
    0.72
    श्विक
    0.71
    轆
    0.70
     repeating
    0.67
     నెల
    0.66
     survey
    0.65
    translate
    0.64
    uring
    0.63
    POSITIVE LOGITS
     ಪ್ರತ
    0.70
     张
    0.66
     identity
    0.65
    계
    0.65
     பாரம்பரிய
    0.65
     কাটা
    0.65
     Identity
    0.64
    Ƌ
    0.64
    Mg
    0.64
     प्
    0.63
    Act Density 0.000%

    No Known Activations