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

    Neuronpedia

    INDEX
    Explanations

    morphological variations of the word "role" and related terms

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    i
    -0.40
    o
    -0.25
    oth
    -0.23
    ez
    -0.21
    iš
    -0.20
    eed
    -0.20
    iado
    -0.20
    ek
    -0.19
    ei
    -0.19
    eh
    -0.19
    POSITIVE LOGITS
    s
    0.26
    ska
    0.20
    scape
    0.19
    sand
    0.19
    suit
    0.18
    sak
    0.18
    sing
    0.18
    spe
    0.17
    sat
    0.17
    sx
    0.17
    Act Density 0.068%

    No Known Activations