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

    Neuronpedia

    INDEX
    Explanations

    glad or happy expressions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     መካከል
    0.38
    에게
    0.35
    ወስ
    0.35
    ֹ
    0.35
    波動
    0.33
    াহীন
    0.33
     jaunâtre
    0.33
     อาจ
    0.33
     প্রথমে
    0.32
     아마
    0.32
    POSITIVE LOGITS
     glad
    0.63
     to
    0.60
     Glad
    0.55
     bahwa
    0.54
     that
    0.51
     أن
    0.50
     they
    0.50
    Glad
    0.47
     That
    0.47
     أنها
    0.46
    Act Density 0.012%

    No Known Activations