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

    Neuronpedia

    INDEX
    Explanations

    repeated phrases emphasizing the concept of 'these' or 'those' in relation to various subjects

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     informaci
    -0.70
     Kleidung
    -0.67
     gouttes
    -0.63
    zieży
    -0.61
    addComponent
    -0.60
     pouvoirs
    -0.59
     Öffentlichkeit
    -0.57
    firstChild
    -0.57
     barnet
    -0.57
    PathVariable
    -0.56
    POSITIVE LOGITS
     these
    1.20
    These
    1.16
     Theses
    1.12
     theses
    1.12
     These
    1.09
     THESE
    1.08
     Theſe
    1.04
    these
    1.01
     kinds
    0.99
    Эти
    0.97
    Act Density 0.138%

    No Known Activations