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

    Neuronpedia

    INDEX
    Explanations

    pronouns and names

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     SHOULD
    0.52
     our
    0.45
     plufieurs
    0.45
     unsere
    0.43
     devemos
    0.43
    是否
    0.43
     ควร
    0.42
     rumoured
    0.42
    <0x0D>
    0.42
     changes
    0.42
    POSITIVE LOGITS
     chuckle
    0.49
     работают
    0.46
     yaşayan
    0.45
    says
    0.44
    さんと
    0.44
    asına
    0.44
     affectionately
    0.44
    Д
    0.43
    сет
    0.43
    他和
    0.42
    Act Density 0.014%

    No Known Activations