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

    Neuronpedia

    INDEX
    Explanations

    references to quantities, particularly "few" and "years."

    oai_token-act-pair · gpt-4o-miniTriggered by @bot
    New Auto-Interp
    Negative Logits
    -за
    -0.16
    798
    -0.15
    stag
    -0.14
    acades
    -0.13
     Stuff
    -0.13
    flows
    -0.13
    -era
    -0.13
     stuff
    -0.13
    -sama
    -0.13
    gio
    -0.13
    POSITIVE LOGITS
     dozen
    0.42
     hundred
    0.33
     thousand
    0.28
     málo
    0.26
    /all
    0.24
     Hundred
    0.22
     of
    0.22
     extra
    0.20
     different
    0.20
    est
    0.19
    Act Density 0.047%

    No Known Activations