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

    Neuronpedia

    INDEX
    Explanations

    words related to highlighting or stressing concepts and ideas

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    zon
    -0.15
    ish
    -0.15
    omb
    -0.14
    PIO
    -0.14
     kä
    -0.14
    ack
    -0.14
    ndl
    -0.14
    slu
    -0.13
    iska
    -0.13
    ange
    -0.13
    POSITIVE LOGITS
    phasis
    0.23
     importance
    0.18
     Importance
    0.17
    phas
    0.16
     emphasis
    0.16
    ãĤ·ãĥ¼
    0.15
    248
    0.14
    ái
    0.14
    IID
    0.14
    pars
    0.14
    Act Density 0.024%

    No Known Activations