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

    Neuronpedia

    INDEX
    Explanations
    No Explanations Found
    New Auto-Interp
    Negative Logits
     Muitos
    1.07
    rouw
    0.98
    𝑥
    0.97
    𝑎
    0.94
    ным
    0.93
     vaste
    0.93
     Sobre
    0.91
     Física
    0.91
     stoked
    0.89
     spéciale
    0.89
    POSITIVE LOGITS
    smaller
    1.39
    sag
    1.36
     casualties
    1.33
    αν
    1.33
    us
    1.28
    𝓪
    1.28
     interpreters
    1.26
     remnants
    1.26
    𝓸
    1.25
    旣
    1.24
    Act Density 0.000%

    No Known Activations