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

    Neuronpedia

    INDEX
    Explanations

    phrases or terms that suggest utility or helpfulness

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

    identifying usefulness

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    InvalidProtocol
    -0.44
    LookAnd
    -0.42
     casó
    -0.41
    BufferException
    -0.41
     VIOL
    -0.40
     arkas
    -0.40
    OGND
    -0.40
     höch
    -0.38
     mahdollis
    -0.37
     LEYENDO
    -0.37
    POSITIVE LOGITS
     helpful
    1.09
     useful
    1.05
    useful
    1.00
    Useful
    0.96
     Useful
    0.95
    helpful
    0.94
    Helpful
    0.93
     Helpful
    0.91
     útiles
    0.86
     útil
    0.85
    Act Density 0.060%

    No Known Activations