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

    Neuronpedia

    INDEX
    Explanations

    void and for in code, rice in food

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     barrage
    0.30
     】,
    0.28
    されていた
    0.27
     ဒီ
    0.27
     listened
    0.27
     drooping
    0.27
     To
    0.27
     straightened
    0.27
     Gone
    0.26
     gone
    0.26
    POSITIVE LOGITS
    ब
    0.30
    ractable
    0.29
     zł
    0.29
    ্ু
    0.29
    improve
    0.28
    using
    0.28
    새
    0.27
    ingan
    0.27
    inox
    0.27
    íts
    0.27
    Act Density 0.000%

    No Known Activations