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

    Neuronpedia

    INDEX
    Explanations

    debug=true in python

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     fal
    1.23
    getto
    1.23
     IEquatable
    1.23
    voud
    1.22
    ce
    1.21
    延長
    1.19
    Lavender
    1.18
    فران
    1.18
    valids
    1.17
    refund
    1.17
    POSITIVE LOGITS
     cual
    1.08
     сі
    0.98
    கோ
    0.96
    ных
    0.95
     disorganized
    0.94
    ため
    0.93
    el
    0.89
    казать
    0.88
     घट
    0.87
    卷
    0.87
    Act Density 0.001%

    No Known Activations