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

    Neuronpedia

    INDEX
    Explanations

    commands and suggestions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     was
    0.52
     of
    0.42
     during
    0.41
     sticky
    0.40
     coursework
    0.40
     wakeup
    0.40
     During
    0.39
     denotes
    0.39
     sinusoidal
    0.39
     및
    0.39
    POSITIVE LOGITS
    ifrån
    0.47
    olith
    0.47
    ólito
    0.44
    äude
    0.43
     ಗಾ
    0.43
    മല്ല
    0.42
    amacare
    0.42
    CLUD
    0.42
    还能
    0.42
    LOTRE
    0.42
    Act Density 0.009%

    No Known Activations