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

    Neuronpedia

    INDEX
    Explanations

    override followed by code definition

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    擢
    0.81
    हाइड
    0.75
    ieszka
    0.71
     flatness
    0.68
    効率
    0.68
     Fermi
    0.67
    ટે
    0.67
    racial
    0.66
    高效
    0.66
     দারিদ্র
    0.66
    POSITIVE LOGITS
     val
    2.27
    val
    2.24
    fun
    1.82
     fun
    1.78
     println
    1.55
    Val
    1.52
     Val
    1.48
    println
    1.44
     var
    1.40
    var
    1.39
    Act Density 0.058%

    No Known Activations