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

    Neuronpedia

    INDEX
    Explanations

    words related to culpability and related actions or states

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    culminate or exculpate

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     étoient
    -0.59
    ſelf
    -0.57
     pinulongan
    -0.56
     avoient
    -0.53
     présidenti
    -0.52
     femei
    -0.52
     Händen
    -0.51
    onee
    -0.50
     neceſſ
    -0.50
    iano
    -0.49
    POSITIVE LOGITS
     cul
    2.14
     Cul
    2.09
    Cul
    2.02
    cul
    1.79
     CUL
    1.77
    CUL
    1.44
     Culver
    1.20
     culvert
    1.12
     cull
    1.11
     Kul
    1.10
    Act Density 0.011%

    No Known Activations