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

    Neuronpedia

    INDEX
    Explanations

    actively working, break down

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    ió
    1.80
     bootstrapping
    1.79
    embeddings
    1.76
    orc
    1.76
    待
    1.75
     bây
    1.74
    โมง
    1.73
    olome
    1.72
    subdirectory
    1.71
    Heating
    1.70
    POSITIVE LOGITS
    ча
    1.80
    ع
    1.77
    िक
    1.73
     psyche
    1.71
     поворо
    1.70
    йки
    1.67
    ce
    1.66
    جست
    1.66
     bede
    1.61
    sters
    1.58
    Act Density 0.000%

    No Known Activations