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

    Neuronpedia

    INDEX
    Explanations

    superlative suffixes

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
    _dma
    -0.07
     bic
    -0.07
     equipe
    -0.07
    γκα
    -0.07
     forfe
    -0.07
     negocio
    -0.06
    ORA
    -0.06
    SuppressWarnings
    -0.06
     Technique
    -0.06
     SCHOOL
    -0.06
    POSITIVE LOGITS
    est
    0.08
     safer
    0.07
    ter
    0.07
    (which
    0.07
    的情
    0.06
     fuller
    0.06
    fer
    0.06
     splitted
    0.06
     Ми
    0.06
     richer
    0.06
    Act Density 0.023%

    No Known Activations