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

    Neuronpedia

    INDEX
    Explanations

    comparative phrases indicating preference or contrast

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
     nonetheless
    -0.16
    arken
    -0.15
     nicht
    -0.15
     niet
    -0.15
    ERSHEY
    -0.15
    esco
    -0.14
    Ľi
    -0.14
     nejen
    -0.14
     både
    -0.14
    uly
    -0.14
    POSITIVE LOGITS
     necessarily
    0.34
     merely
    0.27
     vice
    0.26
     relying
    0.25
     being
    0.24
     mere
    0.24
     just
    0.24
     simply
    0.23
     having
    0.23
    being
    0.22
    Act Density 0.090%

    No Known Activations