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

    Neuronpedia

    INDEX
    Explanations

    negative number operations

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    বঙ্গ
    0.50
    Prime
    0.41
    ransform
    0.41
     primes
    0.41
     fibre
    0.40
     estimator
    0.40
     oaks
    0.40
    畞
    0.40
     sampling
    0.39
    ல்க
    0.39
    POSITIVE LOGITS
    +(-
    0.60
     (−
    0.55
    -(-
    0.52
     (+)
    0.51
     positivas
    0.50
    {(-
    0.49
     +(
    0.48
     (-
    0.48
     negative
    0.47
     positif
    0.47
    Act Density 0.030%

    No Known Activations