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

    Neuronpedia

    INDEX
    Explanations

    open-weights community

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    discrimin
    0.49
     Discrimin
    0.46
     Fores
    0.46
    Discrimin
    0.42
     Inner
    0.40
     fores
    0.40
     मो
    0.40
     foreshadow
    0.40
     arbitrary
    0.39
     inner
    0.39
    POSITIVE LOGITS
     bungalow
    0.57
     Bungal
    0.54
     bungalows
    0.51
    Balt
    0.49
    bung
    0.44
     Baltic
    0.43
     hyd
    0.43
     balt
    0.40
    Hyd
    0.39
     ब्रैकेट
    0.39
    Act Density 0.003%

    No Known Activations