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

    Neuronpedia

    INDEX
    Explanations

    phrases indicating uniqueness or exceptional experiences

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    never seen before or unique

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    脚注の使い方
    -0.63
     مشين
    -0.55
    UnusedPrivate
    -0.55
    writeFieldEnd
    -0.54
     queſta
    -0.53
     propOrder
    -0.53
    KommentareTeilen
    -0.53
    AsUp
    -0.52
     estekak
    -0.51
    indications
    -0.51
    POSITIVE LOGITS
     anything
    0.67
     seen
    0.60
     unique
    0.57
     another
    0.52
     altro
    0.50
     nothing
    0.49
     Anything
    0.49
     unlike
    0.48
    Anything
    0.47
     unparalleled
    0.46
    Act Density 0.011%

    No Known Activations