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

    Neuronpedia

    INDEX
    Explanations

    technical complexity

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     murderous
    0.44
     betray
    0.43
     traitor
    0.43
     Skoda
    0.42
     delusional
    0.41
     grotesque
    0.41
     psychopath
    0.41
     elucid
    0.40
     sarcastic
    0.40
     cowardly
    0.40
    POSITIVE LOGITS
    8
    0.47
    2
    0.46
    arranty
    0.45
    4
    0.43
    ayette
    0.43
    6
    0.43
    7
    0.42
    1
    0.41
    IA
    0.41
    9
    0.40
    Act Density 0.000%

    No Known Activations