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

    Neuronpedia

    INDEX
    Explanations

    automated pruning

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     аре
    0.49
    չ
    0.46
    том
    0.45
    çada
    0.45
    сторон
    0.44
     singleRun
    0.43
    ния
    0.43
     سته
    0.42
    рон
    0.42
     забо
    0.41
    POSITIVE LOGITS
    at
    0.53
    ਰ
    0.52
     ലാ
    0.50
    indicators
    0.46
    an
    0.46
     My
    0.45
    ah
    0.44
    er
    0.43
    ipas
    0.43
    osk
    0.41
    Act Density 0.001%

    No Known Activations