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

    Neuronpedia

    INDEX
    Explanations

    light, lightgreen, lightgray

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    簧
    0.50
     Yer
    0.50
     E
    0.48
    ла
    0.48
     Fertil
    0.48
    ist
    0.48
    andescent
    0.47
    чить
    0.47
    امل
    0.47
    uce
    0.47
    POSITIVE LOGITS
     modulator
    0.56
     vess
    0.55
     conversion
    0.54
     produkter
    0.54
     médioc
    0.54
     marshall
    0.54
     meriye
    0.54
    HARAD
    0.53
     enquiry
    0.53
     моду
    0.53
    Act Density 0.001%

    No Known Activations