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

    Neuronpedia

    INDEX
    Explanations

    code function arguments

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     kabar
    0.42
     irgende
    0.39
     ständig
    0.38
    蠵
    0.37
     prothorax
    0.36
    臓
    0.36
     Sainsbury
    0.36
    瘍
    0.36
     sentir
    0.36
    訫
    0.36
    POSITIVE LOGITS
    and
    0.42
    with
    0.39
    ate
    0.37
     for
    0.35
     with
    0.34
    之后
    0.34
     metadata
    0.33
     decrease
    0.33
    _
    0.33
    ond
    0.33
    Act Density 0.024%

    No Known Activations