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

    Neuronpedia

    INDEX
    Explanations

    expressions of gratitude or appreciation in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    careful and clever actions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    security
    -0.46
     safety
    -0.46
     security
    -0.44
    Security
    -0.43
     sécurité
    -0.40
     for
    -0.39
    SECURITY
    -0.39
    safety
    -0.39
     looks
    -0.38
     risks
    -0.38
    POSITIVE LOGITS
    verifyException
    0.77
     carefully
    0.74
    UnusedPrivate
    0.71
     careful
    0.70
     cleverly
    0.69
     clever
    0.69
    ďaka
    0.69
    hyrchwyd
    0.69
     cuidadosamente
    0.67
     judicious
    0.67
    Act Density 0.104%

    No Known Activations