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

    Neuronpedia

    INDEX
    Explanations

    words whispered in hushed tones

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     affords
    0.44
     pilih
    0.42
     ensures
    0.41
     operates
    0.40
     mitigate
    0.39
    犾
    0.39
     provides
    0.39
     dipilih
    0.38
     team
    0.38
    arry
    0.37
    POSITIVE LOGITS
     cyberpunk
    0.49
     பரபர
    0.48
    ﺸ
    0.47
    ("[
    0.46
    Hidden
    0.46
    Retour
    0.46
     scandalous
    0.45
     indecent
    0.45
     sogenannte
    0.44
     обновления
    0.44
    Act Density 0.001%

    No Known Activations