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

    Neuronpedia

    INDEX
    Explanations

    restricted, especially amounts

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Mad
    0.37
    ülés
    0.37
    лейбол
    0.37
     Mad
    0.37
    bm
    0.37
    γγελμα
    0.37
     regularization
    0.36
    伦敦
    0.35
    td
    0.35
     correctly
    0.35
    POSITIVE LOGITS
     AUR
    0.50
     Ау
    0.49
     Aure
    0.46
     첨
    0.44
     Aura
    0.44
     auk
    0.43
     मुंह
    0.42
     요거
    0.42
    lia
    0.40
    TAU
    0.40
    Act Density 0.000%

    No Known Activations