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

    Neuronpedia

    INDEX
    Explanations

    knock knock joke

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
    0.50
     ◆
    0.48
     BOX
    0.46
    𝗿
    0.46
    𝙧
    0.44
     ⚫
    0.44
    𝘄
    0.43
     »
    0.42
     ▼
    0.42
     ►
    0.42
    POSITIVE LOGITS
    Possible
    0.52
    Begin
    0.47
    Kap
    0.44
    Remember
    0.44
    Quantity
    0.43
    Upon
    0.43
    GRE
    0.40
    While
    0.39
    quant
    0.39
    Quant
    0.38
    Act Density 0.001%

    No Known Activations