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

    Neuronpedia

    INDEX
    Explanations

    guide, convert, seek

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Phe
    0.40
    विजय
    0.40
    jectory
    0.40
     Faithful
    0.40
     Hers
    0.40
    дко
    0.39
     apostle
    0.39
    bum
    0.38
    avour
    0.38
     Stade
    0.38
    POSITIVE LOGITS
     deserve
    0.43
    Global
    0.43
     trillion
    0.43
     Global
    0.42
     billion
    0.39
     千
    0.38
     quit
    0.38
     halted
    0.38
    global
    0.38
     blanket
    0.38
    Act Density 0.052%

    No Known Activations