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

    Neuronpedia

    INDEX
    Explanations

    better to utilize or equip

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    prime
    1.26
    statusCode
    1.14
    better
    1.12
    metro
    1.06
    best
    1.05
    atically
    1.04
    momentum
    1.04
    meth
    1.03
     भर
    1.02
    sk
    1.01
    POSITIVE LOGITS
    ة
    1.65
     suited
    1.34
    س
    1.27
    ed
    1.26
     behaved
    1.23
     able
    1.20
     understood
    1.20
     equipped
    1.19
     localizada
    1.19
    een
    1.18
    Act Density 0.051%

    No Known Activations