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

    Neuronpedia

    INDEX
    Explanations

    technical language and code

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     Sink
    0.72
    島
    0.71
     Adding
    0.70
     Couple
    0.70
     এলো
    0.69
     पसर
    0.68
     Но
    0.68
     மாய
    0.66
     FLASH
    0.66
     nana
    0.65
    POSITIVE LOGITS
     የእ
    0.70
    pea
    0.69
     layers
    0.68
     came
    0.68
     }}^{\
    0.67
     travellers
    0.66
     তা
    0.64
     technical
    0.64
     painfully
    0.64
     inwards
    0.63
    Act Density 0.050%

    No Known Activations