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

    Neuronpedia

    INDEX
    Explanations

    offering style and focus options

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     الفرق
    0.69
     Limitations
    0.67
     Protecting
    0.67
     Protected
    0.67
     protecting
    0.67
     کیوں
    0.66
    켐
    0.66
     protection
    0.65
     Effect
    0.64
    ียง
    0.64
    POSITIVE LOGITS
    Dancing
    0.81
    Make
    0.78
    Speaker
    0.76
     sugger
    0.74
    Zen
    0.74
    Flowers
    0.72
    Yosh
    0.70
    Mad
    0.70
     antena
    0.70
    Ish
    0.70
    Act Density 0.013%

    No Known Activations