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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating a lack of action or helplessness

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     whoever
    -0.17
     somewhere
    -0.17
    å±ħ
    -0.15
     ÑĥÑģÑĸ
    -0.15
    991
    -0.14
     whichever
    -0.14
     Whoever
    -0.14
    FTER
    -0.14
    zel
    -0.14
     withd
    -0.14
    POSITIVE LOGITS
     nothing
    1.05
    nothing
    0.94
     Nothing
    0.93
     NOTHING
    0.88
    Nothing
    0.85
     nada
    0.72
     nichts
    0.68
     rien
    0.67
     ниÑĩего
    0.55
     anything
    0.50
    Act Density 0.298%

    No Known Activations