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

    Neuronpedia

    INDEX
    Explanations

    transformer model architecture

    np_max-act · gemini-2.0-flash
    New Auto-Interp
    Negative Logits
     changing
    -0.08
    Changing
    -0.08
     tenu
    -0.08
    情况
    -0.07
     পরিস্থিত
    -0.07
    changing
    -0.07
     వే�
    -0.07
     Roth
    -0.07
    document
    -0.07
     ప్రమాద
    -0.07
    POSITIVE LOGITS
    -stack
    0.10
     tầng
    0.10
     layers
    0.10
     слоя
    0.10
     слой
    0.10
     terdiri
    0.10
     stacked
    0.10
     �
    0.09
     stack
    0.09
     Layers
    0.09
    Act Density 0.004%

    No Known Activations