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

    Neuronpedia

    INDEX
    Explanations

    warnings and disclaimers

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    Fixture
    0.70
    ረ
    0.66
    ülle
    0.64
    સે
    0.64
    ometown
    0.62
    جی
    0.62
    tokenize
    0.62
    References
    0.61
    yesters
    0.61
    绮
    0.61
    POSITIVE LOGITS
     warning
    4.64
     warnings
    4.41
     Warning
    4.26
     warned
    4.12
    warning
    4.07
    Warning
    4.06
     warn
    4.02
     warns
    3.93
     Warnings
    3.92
    警告
    3.88
    Act Density 0.549%

    No Known Activations