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

    Neuronpedia

    INDEX
    Explanations

    instances of the word "but" and variations that signal contrast or objections within the text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    oire
    -0.70
    ampa
    -0.69
    onomy
    -0.61
     Pigs
    -0.60
    },"
    -0.59
     Mehran
    -0.58
    velt
    -0.58
    wake
    -0.57
     Highest
    -0.57
     anthem
    -0.56
    POSITIVE LOGITS
    tons
    1.10
    chery
    1.06
     nonetheless
    1.02
     otherwise
    0.98
     nevertheless
    0.96
     alas
    0.95
     lacks
    0.93
     lacked
    0.91
     lacking
    0.90
    still
    0.90
    Act Density 0.079%

    No Known Activations