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

    Neuronpedia

    INDEX
    Explanations

    redirecting or steering

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ಬ
    1.51
     cohorts
    1.46
    seau
    1.44
    ست
    1.40
    ême
    1.39
    ょう
    1.38
     ills
    1.38
     TCE
    1.38
     porte
    1.36
    ének
    1.35
    POSITIVE LOGITS
    이기
    1.64
    ro
    1.58
     asunto
    1.46
    이면
    1.42
    ;'>
    1.40
    thead
    1.39
    ্যোগ
    1.39
    да
    1.38
    uole
    1.34
     चलकर
    1.34
    Act Density 0.174%

    No Known Activations