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

    Neuronpedia

    INDEX
    Explanations

    code block syntax or structure

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     northward
    0.48
     vocab
    0.47
     arriver
    0.45
     corrige
    0.44
    appell
    0.44
     verstehen
    0.42
     recher
    0.42
     vocabulary
    0.42
     emulation
    0.41
     selatan
    0.41
    POSITIVE LOGITS
    రా
    0.50
     толькі
    0.48
    anio
    0.46
    ଆ
    0.45
    として
    0.43
    して
    0.43
    ス
    0.43
    した
    0.42
    在
    0.42
    回の
    0.41
    Act Density 0.000%

    No Known Activations