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

    Neuronpedia

    INDEX
    Explanations

    instances of significant moments or events related to proposals and surprises

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ecast
    -0.15
    ãĥ³ãĥĩãĤ£
    -0.15
    ennen
    -0.14
     Wen
    -0.14
    icode
    -0.14
    ortic
    -0.14
    iddi
    -0.14
    eworld
    -0.13
    mens
    -0.13
    ongan
    -0.13
    POSITIVE LOGITS
     surprise
    0.23
     surprises
    0.20
    ampo
    0.17
     secretly
    0.16
     Surprise
    0.16
     ambush
    0.16
    oir
    0.15
    101
    0.15
     Secret
    0.15
     custom
    0.14
    Act Density 0.028%

    No Known Activations