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

    Neuronpedia

    INDEX
    Explanations

    building foundation

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    гана
    0.43
     filaments
    0.43
     Dever
    0.42
    ㎝
    0.40
    olecules
    0.38
    食材
    0.37
    ofil
    0.37
    （
    0.37
    filament
    0.36
     වේ
    0.36
    POSITIVE LOGITS
    𝗸
    0.42
     aviso
    0.39
     lagging
    0.38
    Healing
    0.38
     objet
    0.37
     darts
    0.37
     Часто
    0.37
     правильно
    0.37
     posteriori
    0.36
     healing
    0.36
    Act Density 0.001%

    No Known Activations