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

    Neuronpedia

    INDEX
    Explanations

    prepositions or phrases indicating purpose or relationship

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.02
    1:0.04
    2:0.08
    3:0.29
    4:0.02
    5:0.03
    6:0.06
    7:0.11
    8:0.06
    9:0.07
    10:0.09
    11:0.08
    Negative Logits
    abbling
    -1.23
    ield
    -1.23
    opter
    -1.15
    iflower
    -1.13
    esters
    -1.13
    NAS
    -1.13
     Omaha
    -1.13
    dies
    -1.10
    OGR
    -1.09
    ificial
    -1.07
    POSITIVE LOGITS
     propri
    1.34
     probabilities
    1.34
    REDACTED
    1.33
     timestamp
    1.31
     perspect
    1.30
     particulars
    1.29
     totality
    1.25
     conclud
    1.25
     derog
    1.22
     ali
    1.20
    Act Density 0.012%

    No Known Activations