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

    Neuronpedia

    INDEX
    Explanations

    words and phrases related to conditions and qualifications

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.02
    1:0.01
    2:0.17
    3:0.06
    4:0.06
    5:0.03
    6:0.06
    7:0.32
    8:0.03
    9:0.03
    10:0.11
    11:0.04
    Negative Logits
    haust
    -1.78
    wic
    -1.74
    erity
    -1.63
    uilt
    -1.57
    roxy
    -1.56
    okingly
    -1.56
    rador
    -1.54
    rha
    -1.54
    rina
    -1.53
    orically
    -1.51
    POSITIVE LOGITS
    Reading
    1.40
    ADE
    1.40
     capacitor
    1.38
     Africans
    1.35
     Californ
    1.34
    GBT
    1.32
     Hiroshima
    1.31
     Madagascar
    1.30
     Ukrain
    1.29
    SPONSORED
    1.26
    Act Density 0.088%

    No Known Activations