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

    Neuronpedia

    INDEX
    Explanations

    sweet,

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     in
    0.35
     always
    0.34
    0.34
     now
    0.33
     and
    0.32
     V
    0.31
     the
    0.31
      
    0.30
    ,
    0.29
     a
    0.29
    POSITIVE LOGITS
     PlayerSelector
    0.43
    <unused330>
    0.40
    marshalO
    0.39
    Ꭻ
    0.39
    <unused344>
    0.38
    <unused326>
    0.38
    <unused730>
    0.38
    ategor
    0.37
    <unused641>
    0.37
    <unused393>
    0.37
    Act Density 0.001%

    No Known Activations