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

    Neuronpedia

    INDEX
    Explanations

    comment directives and placeholders

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    …。
    0.35
    此外
    0.32
    Moreover
    0.32
       
    0.32
     (†
    0.31
     furthermore
    0.30
     Messrs
    0.30
    .
    0.30
     meanwhile
    0.30
    Furthermore
    0.30
    POSITIVE LOGITS
     Requires
    0.53
     требует
    0.47
     Reduces
    0.47
     ต้อง
    0.45
     Allows
    0.45
     Creates
    0.44
     Needed
    0.44
     Harus
    0.43
     requires
    0.43
    requires
    0.43
    Act Density 3.931%

    No Known Activations