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

    Neuronpedia

    INDEX
    Explanations

    risks related to implementation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ऍ
    0.36
     sobreviv
    0.35
    leqslant
    0.35
    qquad
    0.34
    _{
    0.34
    供給
    0.34
    KERNEL
    0.34
     kernel
    0.34
    காற்று
    0.33
    VAE
    0.33
    POSITIVE LOGITS
     Groom
    0.42
     Obst
    0.39
    不明
    0.39
    》(
    0.39
     особа
    0.39
     ши
    0.38
     Whiting
    0.38
     obstructed
    0.38
     Goo
    0.38
    aing
    0.37
    Act Density 0.000%

    No Known Activations