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

    Neuronpedia

    INDEX
    Explanations

    improved model or version

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    枭
    0.42
     επισ
    0.39
     madness
    0.39
     climbing
    0.39
     dangerous
    0.38
     sirven
    0.38
     hazardous
    0.38
     embezz
    0.38
     گزار
    0.37
     للخ
    0.37
    POSITIVE LOGITS
    behavior
    0.43
    extensions
    0.40
    model
    0.40
    UTCTime
    0.38
     MODEL
    0.38
    Behavior
    0.37
    管制
    0.37
    ijk
    0.37
     Behavior
    0.37
     libert
    0.36
    Act Density 0.000%

    No Known Activations