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

    Neuronpedia

    INDEX
    Explanations

    code constructors and foreign models

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     AFR
    0.43
    )$:
    0.43
     VCO
    0.41
     Partners
    0.39
     apprezz
    0.39
     CO
    0.38
    淆
    0.37
     Single
    0.36
    ాలా
    0.36
    UTIONS
    0.36
    POSITIVE LOGITS
    レイ
    0.43
    model
    0.43
     модель
    0.42
    梩
    0.41
     model
    0.40
     defend
    0.40
    模型
    0.39
     demon
    0.39
    モデル
    0.39
    demon
    0.38
    Act Density 0.001%

    No Known Activations