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

    Neuronpedia

    INDEX
    Explanations

    references to narrow or restricted dimensions or contexts

    oai_token-act-pair · gpt-4o-mini
    New Auto-Interp
    Negative Logits
    baomidou
    -0.84
     SITES
    -0.80
     bénévol
    -0.74
     colchón
    -0.72
     politiet
    -0.72
    istocene
    -0.71
     lørdag
    -0.71
     copertura
    -0.71
     برانيه
    -0.70
     jorden
    -0.70
    POSITIVE LOGITS
     Narrow
    1.17
    Narrow
    1.10
     narrow
    1.02
     narrows
    0.98
     narrowed
    0.96
    narrow
    0.91
     narrowing
    0.90
     Nare
    0.82
     שוליים
    0.82
     narrowly
    0.79
    Act Density 0.004%

    No Known Activations