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

    Neuronpedia

    INDEX
    Explanations

    method signatures or event handlers

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     from
    -1.21
     people
    -0.94
     /
    -0.94
     specific
    -0.93
     dvo
    -0.92
     regarding
    -0.88
     small
    -0.87
     navideña
    -0.87
     every
    -0.86
     designated
    -0.86
    POSITIVE LOGITS
    WHY
    1.14
     WHY
    1.03
    歌詞
    1.01
     綺麗
    0.95
     gemessen
    0.91
     noastre
    0.91
     rn
    0.88
     RIP
    0.86
    jahres
    0.84
     véhic
    0.84
    Act Density 0.002%

    No Known Activations