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

    Neuronpedia

    INDEX
    Explanations

    the Japanese particle "は" indicating the topic of a sentence

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

    topic marker は

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    PerformLayout
    -0.53
    sprit
    -0.45
    存于互联网档案馆
    -0.45
     Jefus
    -0.44
    plural
    -0.44
    っちゃう
    -0.43
    っちゃった
    -0.43
    BagConstraints
    -0.43
    澍
    -0.43
     Sols
    -0.43
    POSITIVE LOGITS
    は
    0.98
    는
    0.81
    ]]=
    0.68
    은
    0.65
    らは
    0.65
     は
    0.65
    是
    0.64
    等は
    0.63
    年は
    0.63
    )}=
    0.62
    Act Density 0.005%

    No Known Activations