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

    Neuronpedia

    INDEX
    Explanations

    bold instructions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     theme
    0.48
    㷌
    0.45
     setTheme
    0.44
     Wink
    0.43
     주제
    0.42
     Theme
    0.41
    ुलर
    0.40
     темати
    0.40
     thema
    0.39
    筍
    0.39
    POSITIVE LOGITS
     endangered
    0.97
     threatened
    0.94
    threat
    0.85
     rischio
    0.85
     zagro
    0.82
     risk
    0.82
     endanger
    0.81
     risked
    0.78
    risk
    0.77
     jeopard
    0.76
    Act Density 0.001%

    No Known Activations