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

    Neuronpedia

    INDEX
    Explanations

    concessive phrases indicating a contrast or counterargument

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    iginal
    -0.15
    pq
    -0.15
    sez
    -0.14
    aille
    -0.14
    uchs
    -0.14
    çļĦåľ°
    -0.14
     ÑģооÑĤвеÑĤ
    -0.13
    plusplus
    -0.13
    ubre
    -0.13
    浦
    -0.13
    POSITIVE LOGITS
    ness
    0.25
    forth
    0.20
    NESS
    0.20
    amt
    0.16
    ening
    0.16
    theless
    0.16
    atten
    0.15
     ÙģØ¥ÙĨ
    0.15
    obel
    0.15
    umber
    0.14
    Act Density 0.020%

    No Known Activations