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

    Neuronpedia

    INDEX
    Explanations

    #f4f4f4 color code

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Hus
    0.37
    Prothorax
    0.36
    หนังสือ
    0.36
    ೊ
    0.35
     ספר
    0.35
    䒺
    0.35
    㢄
    0.34
     রাখে
    0.34
     âgé
    0.34
    oš
    0.34
    POSITIVE LOGITS
     representations
    0.44
     representation
    0.42
     represent
    0.42
     representan
    0.40
     மேலே
    0.40
     прадстаў
    0.39
     REPRESENT
    0.39
     représent
    0.38
    fta
    0.36
     orig
    0.36
    Act Density 0.002%

    No Known Activations