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

    Neuronpedia

    INDEX
    Explanations

    significantly increases

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     dismay
    0.44
    စွာ
    0.42
    ly
    0.41
    gover
    0.40
     resisting
    0.40
     defens
    0.39
    ｌ
    0.39
     digitized
    0.39
     govern
    0.38
     decentralized
    0.38
    POSITIVE LOGITS
    щі
    0.43
     devuelve
    0.38
     mauris
    0.38
     συμφ
    0.37
    ariest
    0.37
    undle
    0.37
    יל
    0.36
    setMax
    0.36
    ⊢
    0.36
     unicorn
    0.35
    Act Density 0.001%

    No Known Activations