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

    Neuronpedia

    INDEX
    Explanations

    product series

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     hurtful
    0.71
     causal
    0.71
     causality
    0.69
    urricular
    0.64
     Boolean
    0.64
     solutes
    0.63
     pathways
    0.61
     literacy
    0.61
    얽
    0.61
    boolean
    0.61
    POSITIVE LOGITS
     Series
    1.48
     series
    1.33
    シリーズ
    1.26
    Series
    1.18
     серия
    1.16
    系列
    1.13
     Serie
    1.10
     시리즈
    1.10
     серии
    1.08
     รุ่น
    1.07
    Act Density 0.300%

    No Known Activations