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

    Neuronpedia

    INDEX
    Explanations

    asking about user intent

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     genü
    0.60
     lze
    0.58
     asla
    0.56
     niemals
    0.56
     그러면
    0.55
     отдельных
    0.55
     способности
    0.55
     waarmee
    0.53
     drugih
    0.53
     voldoende
    0.53
    POSITIVE LOGITS
     trying
    2.31
     attempting
    1.85
     seeking
    1.82
     looking
    1.82
     striving
    1.77
    trying
    1.75
     aiming
    1.75
     interested
    1.74
     hoping
    1.70
     experimenting
    1.64
    Act Density 0.988%

    No Known Activations