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

    Neuronpedia

    INDEX
    Explanations

    remains unchanged or intact

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     segera
    0.48
     préal
    0.47
     okam
    0.47
     сразу
    0.47
     imediatamente
    0.46
     sofort
    0.46
     immédiatement
    0.46
     odmah
    0.46
     immédi
    0.45
     dès
    0.44
    POSITIVE LOGITS
     indefinitely
    0.89
     intact
    0.85
     throughout
    0.75
     longer
    0.71
     unchanged
    0.70
     despite
    0.64
     dłu
    0.64
     Longer
    0.62
     बरकरार
    0.61
    不变
    0.59
    Act Density 0.027%

    No Known Activations