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

    Neuronpedia

    INDEX
    Explanations

    words related to numerical or coded data, possibly indicating values or categories in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     g
    -0.15
    ersh
    -0.14
    ughs
    -0.14
     Vie
    -0.14
    orton
    -0.14
    skirts
    -0.14
     gl
    -0.14
    esper
    -0.14
    ared
    -0.14
    perimental
    -0.13
    POSITIVE LOGITS
     (*((
    0.16
    egg
    0.14
     dilig
    0.14
    lett
    0.14
     parch
    0.14
    Ñĭл
    0.14
    apult
    0.14
     ì§
    0.13
    tty
    0.13
     ReturnType
    0.13
    Act Density 0.030%

    No Known Activations