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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating redundancy or excess

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ferred
    -0.15
    огÑĢам
    -0.15
    rong
    -0.14
    panion
    -0.14
    strcasecmp
    -0.14
    eger
    -0.14
    inals
    -0.13
    udies
    -0.13
    erne
    -0.13
    æk
    -0.13
    POSITIVE LOGITS
     say
    0.20
    say
    0.19
    isay
    0.18
    days
    0.18
     stays
    0.17
     distractions
    0.17
    lessly
    0.16
     mention
    0.16
     sagen
    0.16
    .say
    0.16
    Act Density 0.009%

    No Known Activations