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

    Neuronpedia

    INDEX
    Explanations

    specific references to primary focus areas or key subjects within discussions

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Head Attr Weights
    0:0.02
    1:0.02
    2:0.15
    3:0.22
    4:0.15
    5:0.05
    6:0.04
    7:0.05
    8:0.04
    9:0.06
    10:0.08
    11:0.07
    Negative Logits
    "[
    -1.49
    )[
    -1.40
    Interstitial
    -1.40
    aneously
    -1.39
    )."
    -1.37
    angles
    -1.36
    udden
    -1.35
     ['
    -1.35
    ?」
    -1.33
     ('
    -1.32
    POSITIVE LOGITS
     disclaim
    1.79
     responsibly
    1.78
    Disclaimer
    1.66
    PDATED
    1.64
     THANK
    1.63
    OIL
    1.59
     optimism
    1.59
     cautiously
    1.52
     caveats
    1.48
    Hopefully
    1.45
    Act Density 0.005%

    No Known Activations