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

    Neuronpedia

    INDEX
    Explanations

    conditional phrases indicating a sequence of events

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     purpoſe
    -0.85
     ſeveral
    -0.76
     poffible
    -0.75
     ftate
    -0.75
     Theſe
    -0.74
     ſtate
    -0.72
     ſche
    -0.71
     ſame
    -0.71
     himſelf
    -0.69
     houſe
    -0.68
    POSITIVE LOGITS
     then
    2.28
     THEN
    1.56
    then
    1.56
     entonces
    1.47
     então
    1.43
    THEN
    1.32
     luego
    1.26
     dann
    1.24
     allora
    1.16
     Then
    1.14
    Act Density 0.182%

    No Known Activations