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

    Neuronpedia

    INDEX
    Explanations

    given conditional

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    忔
    0.39
    这不是
    0.39
    Nicht
    0.38
    нике
    0.38
     impractical
    0.38
    嶈
    0.37
     разно
    0.37
    Mismatch
    0.37
    NoStop
    0.36
    вин
    0.36
    POSITIVE LOGITS
     yani
    0.51
     অর্থাৎ
    0.43
     estão
    0.42
    都已经
    0.39
     están
    0.38
     sebagainya
    0.38
     heps
    0.38
     saja
    0.37
     (
    0.37
     tutte
    0.36
    Act Density 0.032%

    No Known Activations