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

    Neuronpedia

    INDEX
    Explanations

    system resources and performance

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     not
    0.57
     pick
    0.49
     can
    0.49
     We
    0.48
    be
    0.46
     All
    0.46
     Peach
    0.46
     would
    0.45
     for
    0.45
     It
    0.45
    POSITIVE LOGITS
     پڑے
    0.55
     जनगणना
    0.52
     salles
    0.48
    ʼn
    0.48
    propane
    0.48
    가의
    0.48
    ⤟
    0.48
    –
    0.48
    subplots
    0.48
     변
    0.48
    Act Density 0.000%

    No Known Activations