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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating tasks or actions in a sequence

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ovich
    -0.17
     Throughout
    -0.16
    Throughout
    -0.16
    inate
    -0.16
    erman
    -0.15
    tring
    -0.15
    isen
    -0.15
    olini
    -0.15
    tte
    -0.15
     throughout
    -0.14
    POSITIVE LOGITS
     unit
    0.27
    unit
    0.22
    -unit
    0.21
     缴
    0.20
     Unit
    0.20
    	un
    0.19
     UNIT
    0.18
    [unit
    0.18
     unidad
    0.18
    	unit
    0.18
    Act Density 0.062%

    No Known Activations