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

    Neuronpedia

    INDEX
    Explanations

    tf.js code

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     ""))
    0.44
    Idents
    0.41
     Analyzer
    0.41
     консу
    0.40
    ToList
    0.39
     Perfecto
    0.39
    Neo
    0.39
    Consistent
    0.38
    шты
    0.38
     Modifier
    0.37
    POSITIVE LOGITS
    vaise
    0.44
     Castile
    0.43
     undefined
    0.42
     diapers
    0.41
    undefined
    0.41
     burial
    0.41
     rufis
    0.40
    efined
    0.40
    getClassName
    0.40
    elsk
    0.40
    Act Density 0.003%

    No Known Activations