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

    Neuronpedia

    INDEX
    Explanations

    goes against

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ulfide
    0.44
     wrong
    0.42
     fully
    0.41
    olda
    0.40
     Wrong
    0.39
     opposing
    0.39
    fully
    0.39
     succesfully
    0.39
     adversarial
    0.38
    south
    0.38
    POSITIVE LOGITS
    麃
    0.44
    Kern
    0.43
    Hand
    0.42
     hand
    0.41
     без
    0.41
     चालक
    0.38
     Dipl
    0.38
    वायु
    0.38
    義
    0.37
     Hand
    0.37
    Act Density 0.003%

    No Known Activations