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

    Neuronpedia

    INDEX
    Explanations

    encoder blocks

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    rant
    0.48
    Men
    0.47
     основные
    0.46
    Band
    0.46
    Ни
    0.46
    Следу
    0.46
    Й
    0.46
    То
    0.45
    ຣ
    0.45
     основной
    0.45
    POSITIVE LOGITS
     opgenomen
    0.43
     aldehydes
    0.41
     creates
    0.41
     realizan
    0.40
     assunto
    0.40
    OPENJDK
    0.39
     formularios
    0.39
     MSRP
    0.39
     an
    0.39
     rene
    0.39
    Act Density 0.012%

    No Known Activations