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

    Neuronpedia

    INDEX
    Explanations

    say "new"

    np_max-act-logits · gemini-2.0-flashTriggered by @majianfei030706
    New Auto-Interp
    Negative Logits
     SOL
    -0.08
     lt
    -0.08
    (cf
    -0.07
    visitor
    -0.07
     fö
    -0.07
    .say
    -0.07
     mentioned
    -0.06
    少了
    -0.06
     variant
    -0.06
    .primary
    -0.06
    POSITIVE LOGITS
    Life
    0.07
    ܛ
    0.07
    _Word
    0.07
    Triple
    0.07
    uke
    0.07
     bank
    0.07
    па
    0.06
    adelphia
    0.06
    pur
    0.06
    建筑工程
    0.06
    Act Density 0.047%

    No Known Activations