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

    Neuronpedia

    INDEX
    Explanations

    pattern completions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    い
    0.52
     Москвы
    0.45
    嶌
    0.45
     postaje
    0.44
    ק
    0.44
    硐
    0.44
     Colored
    0.44
    пь
    0.44
     postérieure
    0.43
    IN
    0.42
    POSITIVE LOGITS
    ner
    0.49
     haben
    0.46
     troubling
    0.45
    iler
    0.45
    roller
    0.45
     sullen
    0.45
     τ
    0.45
     blatant
    0.45
     wenn
    0.43
     para
    0.43
    Act Density 0.046%

    No Known Activations