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

    Neuronpedia

    INDEX
    Explanations

    action/state changes

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     actually
    0.54
     BOTH
    0.50
     certes
    0.45
     OTHER
    0.44
     aslında
    0.42
    ITHER
    0.41
     (
    0.41
     ALSO
    0.41
     *
    0.40
     যেহেতু
    0.40
    POSITIVE LOGITS
    有所
    0.47
     prevails
    0.39
     predominate
    0.39
     prevail
    0.38
    をご覧ください
    0.37
     suffice
    0.36
     கூறினார்
    0.35
     sluč
    0.35
    做得
    0.35
    லைக்
    0.35
    Act Density 0.009%

    No Known Activations