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

    Neuronpedia

    INDEX
    Explanations

    subtract or substitute

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ↵
    1.53
    i
    1.03
     wichtig
    0.95
    ,
    0.91
    x
    0.88
    ي
    0.82
    t
    0.73
    IS
    0.72
     DAG
    0.72
    ↵↵
    0.71
    POSITIVE LOGITS
     as
    0.98
    د
    0.96
    ある
    0.93
    ご
    0.81
     유사
    0.80
    大
    0.80
    あまり
    0.80
    リ
    0.78
    お
    0.76
    見
    0.76
    Act Density 0.007%

    No Known Activations