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

    Neuronpedia

    INDEX
    Explanations

    colons

    np_max-act · gemini-2.0-flash

    floating-point decimal numbers (standalone numeric tokens).

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    New Auto-Interp
    Negative Logits
    .ColumnHeadersHeightSizeMode
    -0.06
     Ses
    -0.06
    职业
    -0.06
    dir
    -0.06
     рев
    -0.06
    Regs
    -0.06
     kadınlar
    -0.05
    herit
    -0.05
     dosp
    -0.05
     Sew
    -0.05
    POSITIVE LOGITS
     strips
    0.08
     GOOGLE
    0.07
    queeze
    0.07
    %↵↵
    0.07
     wool
    0.07
    _SAVE
    0.07
     então
    0.07
    becue
    0.07
    理解
    0.07
    _q
    0.07
    Act Density 0.211%

    No Known Activations