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

    Neuronpedia

    INDEX
    Explanations

    avoiding problems

    np_max-act-logits · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     Mom
    -0.07
    rames
    -0.07
    μές
    -0.07
    omas
    -0.07
    anging
    -0.07
     nedeniyle
    -0.06
    .J
    -0.06
    ionální
    -0.06
    τομα
    -0.06
    .render
    -0.06
    POSITIVE LOGITS
    shelf
    0.06
    -top
    0.06
     vyk
    0.06
     dlg
    0.06
     relieve
    0.06
     Gast
    0.06
    てる
    0.06
     ASN
    0.05
    出现
    0.05
     radar
    0.05
    Act Density 0.113%

    No Known Activations