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

    Neuronpedia

    INDEX
    Explanations

    system, input values, groups are

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    。
    0.82
    .
    0.69
    ।
    0.67
    0.64
     transferable
    0.63
    ?
    0.63
     established
    0.61
    name
    0.59
    এ
    0.58
    established
    0.57
    POSITIVE LOGITS
     tengan
    0.64
    যেন
    0.62
     luôn
    0.61
     pourra
    0.60
    ेंसेस
    0.60
     présenter
    0.60
    chsler
    0.59
     preferentially
    0.59
    টেও
    0.59
     nhìn
    0.58
    Act Density 0.384%

    No Known Activations