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

    Neuronpedia

    INDEX
    Explanations

    code fragments and technical terms

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    5
    0.43
    3
    0.43
    Mother
    0.42
    4
    0.42
    [
    0.41
    8
    0.41
     Mother
    0.39
    突然
    0.39
    为
    0.38
    *
    0.38
    POSITIVE LOGITS
     swipes
    0.50
     configs
    0.47
     שה
    0.46
    arrayList
    0.46
     apopt
    0.45
     analogs
    0.44
     noms
    0.44
     یا
    0.44
    achtige
    0.43
     הס
    0.43
    Act Density 0.007%

    No Known Activations