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

    Neuronpedia

    INDEX
    Explanations

    globes, Addresses, beast

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     top
    0.51
     respective
    0.42
     instrumental
    0.42
     pens
    0.41
     do
    0.41
     k
    0.41
     s
    0.40
     National
    0.39
     Compass
    0.39
     Spine
    0.39
    POSITIVE LOGITS
    ۸
    0.81
    <unused2023>
    0.81
    ۴
    0.80
    ٨
    0.80
    quantidade
    0.80
    <unused2100>
    0.79
     获取
    0.79
    <unused234>
    0.79
    २
    0.79
    ječ
    0.79
    Act Density 0.000%

    No Known Activations