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

    Neuronpedia

    INDEX
    Explanations

    instances of the Japanese particle "を" indicating the direct object in sentences

    oai_token-act-pair · gpt-4o-miniTriggered by @bot

    direct object marker

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     similit
    -0.40
     ainfi
    -0.39
     espé
    -0.39
     nė
    -0.39
     iſt
    -0.38
     exigences
    -0.38
     enfans
    -0.38
     Jefus
    -0.37
     cser
    -0.37
     plufieurs
    -0.37
    POSITIVE LOGITS
    を
    1.44
    를
    1.07
    을
    0.93
    いを
    0.90
     を
    0.87
    りを
    0.79
     를
    0.74
     을
    0.72
    のを
    0.71
    子を
    0.70
    Act Density 0.003%

    No Known Activations