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

    Neuronpedia

    INDEX
    Explanations

    explaining actions and states

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ấu
    0.53
    不過
    0.52
     dru
    0.52
     walnuts
    0.51
    юць
    0.50
    藠
    0.50
     называ
    0.50
    の為
    0.50
     Spro
    0.49
     walks
    0.49
    POSITIVE LOGITS
    v
    0.58
    T
    0.50
    S
    0.49
     ဆ
    0.48
    se
    0.48
     स्टैट
    0.47
    p
    0.47
    f
    0.46
    U
    0.46
    zinha
    0.46
    Act Density 0.000%

    No Known Activations