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

    Neuronpedia

    INDEX
    Explanations

    -ing states

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    গার
    0.74
     unripe
    0.67
     dyed
    0.66
    sensitivity
    0.66
     applique
    0.65
    Sensitivity
    0.64
     loopt
    0.64
    heed
    0.64
     Poor
    0.63
    ચાર
    0.63
    POSITIVE LOGITS
     liberating
    2.11
     satisfying
    2.01
     gratifying
    1.99
     comforting
    1.95
     invigorating
    1.93
     frustrating
    1.93
     rewarding
    1.92
     exhilarating
    1.91
     exhausting
    1.91
     soothing
    1.91
    Act Density 0.655%

    No Known Activations