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

    Neuronpedia

    INDEX
    Explanations

    spelled words

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    >∕
    0.27
    專業
    0.26
    引っ
    0.26
    収入
    0.26
    ᖁ
    0.26
    詮
    0.26
    沍
    0.26
     erstmal
    0.25
     drawdown
    0.25
     xAxis
    0.25
    POSITIVE LOGITS
     spelling
    0.39
     vowels
    0.38
    e
    0.37
     Spelling
    0.31
     e
    0.29
     spelled
    0.29
    C
    0.28
     consonants
    0.28
     misspelled
    0.28
     p
    0.27
    Act Density 0.083%

    No Known Activations