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

    Neuronpedia

    INDEX
    Explanations

    instances of the word "always" and its variations indicating frequency and certainty

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Gracie
    -0.89
    writeFieldEnd
    -0.78
    InjectAttribute
    -0.74
     MainAxisSize
    -0.72
    nonatomic
    -0.72
     Halley
    -0.70
     nonatomic
    -0.70
    ]")]
    -0.69
     StatelessWidget
    -0.69
    anyahu
    -0.69
    POSITIVE LOGITS
     can
    0.98
     Can
    0.89
     CAN
    0.87
    can
    0.86
    Can
    0.79
     توانند
    0.77
    CAN
    0.73
    也可
    0.72
     potrà
    0.70
     può
    0.69
    Act Density 0.167%

    No Known Activations