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

    Neuronpedia

    INDEX
    Explanations

    detoxification and other specific concepts

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     savior
    0.45
     cherry
    0.42
     seizing
    0.42
     avsl
    0.41
     Average
    0.40
     gateway
    0.40
     killer
    0.40
     seizure
    0.40
     disability
    0.40
     average
    0.40
    POSITIVE LOGITS
     heutigen
    0.41
    Vector
    0.39
    崧
    0.39
     ഗു
    0.39
     ബ
    0.38
     сьогодні
    0.38
    erosene
    0.38
    ос
    0.38
    贰百
    0.37
     উৎপাদন
    0.37
    Act Density 0.000%

    No Known Activations