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

    Neuronpedia

    INDEX
    Explanations

    should not

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     planned
    0.49
     diminish
    0.41
     dissu
    0.40
     lead
    0.40
     contrary
    0.40
     bf
    0.39
     বিএনপি
    0.38
     discourage
    0.38
    чним
    0.37
    msford
    0.37
    POSITIVE LOGITS
    廨
    0.46
    Below
    0.43
    谨慎
    0.43
     ajutor
    0.42
     జాగ్ర
    0.41
    ዚያ
    0.40
     BELOW
    0.39
    ابر
    0.38
    lax
    0.38
    燡
    0.38
    Act Density 0.002%

    No Known Activations