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

    Neuronpedia

    INDEX
    Explanations

    references to sinks or sink-related concepts

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    doms
    -0.68
     SX
    -0.63
    ³³
    -0.62
    OTO
    -0.62
    æ³
    -0.61
     Unch
    -0.61
    çĶŁ
    -0.61
    ACP
    -0.61
    OGR
    -0.61
    RFC
    -0.60
    POSITIVE LOGITS
    hole
    1.34
    holes
    1.24
    vier
    0.95
     sink
    0.93
    maid
    0.92
    ucer
    0.86
    fish
    0.83
    gel
    0.82
    pipe
    0.82
    mers
    0.81
    Act Density 0.002%

    No Known Activations