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

    Neuronpedia

    INDEX
    Explanations

    instances of markup or special characters indicating document structure

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    aday
    -0.15
    adier
    -0.15
    indre
    -0.15
     Wander
    -0.14
    ianne
    -0.14
    INGS
    -0.13
     vap
    -0.13
    enko
    -0.13
    assen
    -0.13
    mand
    -0.13
    POSITIVE LOGITS
    DOCTYPE
    0.31
    doctype
    0.20
    CDATA
    0.18
    urch
    0.18
     sovere
    0.16
    ENTITY
    0.15
    çħ§
    0.15
    ãĥĥãĥĦ
    0.15
    /ros
    0.15
    sett
    0.14
    Act Density 0.012%

    No Known Activations