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

    Neuronpedia

    INDEX
    Explanations

    harmful testimony, exploiting, wound

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    సర్
    0.44
     unlawfully
    0.42
     unwarrant
    0.42
     উত্তম
    0.41
    了起来
    0.40
     ആവശ്യമ
    0.40
     unsett
    0.40
     Variablen
    0.40
     necessaria
    0.40
    SizePolicy
    0.40
    POSITIVE LOGITS
     amazon
    0.41
     Amazon
    0.40
    Phir
    0.38
    edged
    0.37
     орга
    0.36
     führt
    0.36
     bekommt
    0.36
    regon
    0.35
     cryogenic
    0.35
     jag
    0.35
    Act Density 0.000%

    No Known Activations