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

    Neuronpedia

    INDEX
    Explanations

    evaluative adjectives indicating correctness or sufficiency

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    නී
    0.37
    ஆர்
    0.36
    гей
    0.36
    荕
    0.36
    rup
    0.35
    𝔾
    0.34
    危险
    0.33
     கிஷோர்
    0.33
    🐌
    0.32
    ર્સ
    0.32
    POSITIVE LOGITS
     enough
    0.45
     Enough
    0.44
     banget
    0.42
     للغاية
    0.38
    无比
    0.38
    .
    0.38
     ENO
    0.37
     hale
    0.37
     genoeg
    0.36
     codeword
    0.36
    Act Density 0.042%

    No Known Activations