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

    Neuronpedia

    INDEX
    Explanations

    contractions indicating possession or descriptions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    ìĦł
    -0.17
    763
    -0.15
     thereby
    -0.14
    Ùĩ
    -0.14
    ipeg
    -0.13
    etc
    -0.13
    -même
    -0.13
     bulunan
    -0.13
     gratuites
    -0.13
    boa
    -0.13
    POSITIVE LOGITS
     why
    0.30
     how
    0.26
    why
    0.21
     precisely
    0.20
     where
    0.20
     what
    0.20
    为ä»Ģä¹Ī
    0.18
    how
    0.18
     right
    0.18
     exactly
    0.17
    Act Density 0.084%

    No Known Activations