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

    Neuronpedia

    INDEX
    Explanations

    various dash and hyphen formats used in writing

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    rag
    -0.18
    åķ
    -0.17
    linger
    -0.16
    uters
    -0.15
    erce
    -0.15
    edad
    -0.14
    olina
    -0.14
    auc
    -0.14
    utors
    -0.14
    ecure
    -0.13
    POSITIVE LOGITS
    oret
    0.17
     both
    0.16
     -*-č↵
    0.15
    both
    0.15
     Janet
    0.14
    zo
    0.14
    840
    0.14
    бÑĥдÑĮ
    0.13
    _both
    0.13
     Wyn
    0.13
    Act Density 0.125%

    No Known Activations