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

    Neuronpedia

    INDEX
    Explanations

    increasingly concerned

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    æk
    0.31
    特定の
    0.30
    複数の
    0.30
     serializer
    0.29
     ряда
    0.29
     தேர்ந்த
    0.28
    いくつかの
    0.28
     columnIndex
    0.28
    য়েক
    0.28
     TypeScript
    0.28
    POSITIVE LOGITS
     unbelievably
    0.34
     cannot
    0.32
     desperately
    0.31
     दिखाकर
    0.31
     importantly
    0.30
     WITHOUT
    0.30
     ironically
    0.30
     MUST
    0.29
     -
    0.29
     tanpa
    0.29
    Act Density 0.208%

    No Known Activations