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

    Neuronpedia

    INDEX
    Explanations

    demonstrate results

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     simonsen
    0.32
     develops
    0.31
     diverges
    0.31
     pursues
    0.31
     schist
    0.30
     underlies
    0.30
     resolves
    0.29
     possuem
    0.29
     recent
    0.29
     overview
    0.29
    POSITIVE LOGITS
     Honestly
    0.39
     Và
    0.36
     แถม
    0.36
     And
    0.34
     และ
    0.33
    そして
    0.33
     あなた
    0.33
     Και
    0.31
     অথবা
    0.31
     Remember
    0.31
    Act Density 0.001%

    No Known Activations