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

    Neuronpedia

    INDEX
    Explanations

    causal and conditional phrases in text

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     Há»į
    -0.16
     loro
    -0.16
    iero
    -0.15
     jejÃŃ
    -0.14
     __("
    -0.14
    ange
    -0.13
     nữa
    -0.13
    aje
    -0.13
     há»į
    -0.13
    ãģ¾ãģŁãģ¯
    -0.13
    POSITIVE LOGITS
     there
    0.26
     when
    0.21
    there
    0.21
     many
    0.21
     although
    0.20
     some
    0.20
     during
    0.20
     since
    0.19
     unlike
    0.19
     whereas
    0.19
    Act Density 0.442%

    No Known Activations