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

    Neuronpedia

    INDEX
    Explanations

    generating text or keywords

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Channels
    0.39
    粎
    0.39
    identity
    0.38
    ក្រ
    0.38
    Dutch
    0.37
     réf
    0.36
    reflect
    0.36
    reality
    0.36
    স্তর
    0.36
     عندك
    0.36
    POSITIVE LOGITS
     море
    0.41
     тик
    0.39
     सीखने
    0.39
     Legendary
    0.38
     эмо
    0.38
     sahib
    0.38
    iderm
    0.38
    িনি
    0.37
     തയ്യാ
    0.37
    за
    0.37
    Act Density 0.000%

    No Known Activations