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

    Neuronpedia

    INDEX
    Explanations

    import or artifact

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     of
    0.69
     Of
    0.57
     Snowflake
    0.55
     Nights
    0.54
     [
    0.53
     People
    0.53
     Region
    0.52
     Rusty
    0.52
     on
    0.51
     Date
    0.51
    POSITIVE LOGITS
    .
    0.61
     تضم
    0.52
    5
    0.52
     suas
    0.51
     dinding
    0.49
     thuốc
    0.49
     vervolgens
    0.48
    ।
    0.48
    ını
    0.48
    。
    0.48
    Act Density 0.031%

    No Known Activations