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

    Neuronpedia

    INDEX
    Explanations

    instruction to insert text

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     refreshments
    0.42
    的老
    0.42
     commentary
    0.41
    правление
    0.41
     кине
    0.41
     pivots
    0.40
     magazines
    0.40
    舞
    0.40
     নিপী
    0.39
     жесто
    0.39
    POSITIVE LOGITS
     Thy
    0.48
    ⓷
    0.47
    PREFIX
    0.47
    ک
    0.46
    WITH
    0.45
    with
    0.44
     sila
    0.43
    Thy
    0.43
    tolist
    0.43
    utiliser
    0.43
    Act Density 0.004%

    No Known Activations