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

    Neuronpedia

    INDEX
    Explanations

    instances of the word "true" in various contexts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ronics
    -0.20
    trib
    -0.17
    lot
    -0.17
    ses
    -0.17
    shire
    -0.16
    ummings
    -0.15
    recht
    -0.15
    ãĥªãĥ¼ãĤº
    -0.15
    truth
    -0.15
    ized
    -0.14
    POSITIVE LOGITS
    /false
    0.42
    -blue
    0.30
    st
    0.28
    -life
    0.23
    caller
    0.23
    blood
    0.21
    -blood
    0.21
    fully
    0.20
    -bel
    0.19
     believers
    0.19
    Act Density 0.061%

    No Known Activations