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

    Neuronpedia

    INDEX
    Explanations

    signature, attraction, verification

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     тоже
    0.78
    verdad
    0.76
    dreams
    0.74
     наверное
    0.72
    К
    0.71
    happiness
    0.70
     ♪
    0.70
     врач
    0.69
     adalah
    0.69
     właśnie
    0.69
    POSITIVE LOGITS
     ambiguities
    0.71
     transactional
    0.66
     subsequent
    0.63
     questionable
    0.63
     verification
    0.62
     Labels
    0.62
     diret
    0.61
     Verification
    0.60
     가운데
    0.59
     readability
    0.59
    Act Density 0.101%

    No Known Activations