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

    Neuronpedia

    INDEX
    Explanations

    technical prefixes and identifiers

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    z
    0.43
    <unused1068>
    0.39
    f
    0.39
    ak
    0.38
    ون
    0.38
    an
    0.38
    ாக
    0.37
    штар
    0.35
     agrad
    0.34
    ला
    0.34
    POSITIVE LOGITS
     to
    0.59
     t
    0.40
     y
    0.34
     k
    0.34
     У
    0.33
     الع
    0.33
    INE
    0.32
    לי
    0.31
     tantamount
    0.31
    ^
    0.31
    Act Density 4.527%

    No Known Activations