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

    Neuronpedia

    INDEX
    Explanations

    programming function calls

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Hé
    0.37
     उष्ण
    0.33
     إذ
    0.33
    膂
    0.33
    Montana
    0.33
     الحق
    0.32
     மல
    0.31
     экскур
    0.31
     Hebre
    0.31
    Bern
    0.30
    POSITIVE LOGITS
    ().
    0.50
    ()
    0.50
    .
    0.44
    /
    0.40
    (),
    0.40
    ("")
    0.37
    ();
    0.34
     function
    0.34
     clause
    0.34
    ()`
    0.34
    Act Density 0.567%

    No Known Activations