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

    Neuronpedia

    INDEX
    Explanations

    relative to positions

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    blockSize
    0.68
     shingles
    0.66
     savoury
    0.64
    ন্দরী
    0.64
     Seamless
    0.64
    getConfig
    0.64
     பாதி
    0.63
     corrosion
    0.62
     getConfig
    0.61
    Seamless
    0.61
    POSITIVE LOGITS
     relative
    2.55
    相对于
    2.54
    relative
    2.42
     respecto
    2.32
     отношению
    2.25
     rispetto
    2.22
    Relative
    2.21
     Relative
    2.18
     wzglę
    2.18
     terhadap
    2.07
    Act Density 0.224%

    No Known Activations