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

    Neuronpedia

    INDEX
    Explanations

    crucial, let's, close attention, caveats

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    1.21
    가
    1.20
    ↵↵
    1.20
     Girlfriend
    1.18
     adorn
    1.17
    ↵
    1.17
    k
    1.16
     составляет
    1.15
    ER
    1.14
     alegria
    1.13
    POSITIVE LOGITS
    ït
    1.57
    राधिक
    1.53
    當
    1.36
     BSData
    1.35
    ត្ត
    1.32
     hindsight
    1.32
    aliyet
    1.31
    अरविंद
    1.31
    సారి
    1.31
    âce
    1.30
    Act Density 0.752%

    No Known Activations