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

    Neuronpedia

    INDEX
    Explanations

    instances of the word "very."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Jefus
    -0.83
     ſche
    -0.77
     ſch
    -0.77
     auffi
    -0.76
     houſe
    -0.75
     Reſ
    -0.75
     iſt
    -0.75
     judiciales
    -0.73
     againſt
    -0.72
     ſtate
    -0.72
    POSITIVE LOGITS
     very
    1.09
    very
    0.87
     VERY
    0.81
     Very
    0.81
     same
    0.78
    Very
    0.77
     própria
    0.75
    VERY
    0.74
     stesso
    0.67
    same
    0.61
    Act Density 0.188%

    No Known Activations