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

    Neuronpedia

    INDEX
    Explanations

    strategy type possessive

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    op
    0.56
    а
    0.53
    adel
    0.51
    atau
    0.51
    af
    0.47
    is
    0.46
    $\$
    0.46
    inov
    0.45
    о
    0.45
    pot
    0.44
    POSITIVE LOGITS
    使い
    0.53
     uso
    0.50
     muchas
    0.49
    続
    0.49
     évaluation
    0.49
    ğini
    0.49
     deceive
    0.48
    し
    0.48
     aceste
    0.48
     કરીએ
    0.48
    Act Density 0.000%

    No Known Activations