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

    Neuronpedia

    INDEX
    Explanations

    discussing issues

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     number
    0.75
     semblance
    0.75
     ஏராளமான
    0.73
     eagerness
    0.72
     unharmed
    0.71
     flurry
    0.70
     약간
    0.69
     яр
    0.68
     never
    0.68
     indistinct
    0.68
    POSITIVE LOGITS
    problems
    1.12
    issues
    1.11
    Issues
    1.08
     problém
    1.06
     aspectos
    1.05
    Ethics
    1.05
     cuestiones
    1.04
     problemen
    1.02
     issues
    1.01
     problèmes
    1.00
    Act Density 1.248%

    No Known Activations