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

    Neuronpedia

    INDEX
    Explanations

    writing greetings and advice

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     components
    0.46
     functional
    0.45
     consultant
    0.44
     dalla
    0.43
     fascia
    0.42
    ུ་
    0.41
    არის
    0.40
     connector
    0.40
     connectors
    0.40
     bulldog
    0.39
    POSITIVE LOGITS
    预测
    0.51
    环境
    0.50
    𝙀
    0.47
    异步
    0.47
    大约
    0.47
    ខ្ញ
    0.46
     这些
    0.46
     不过
    0.46
    𝓊
    0.46
    攻击
    0.46
    Act Density 0.001%

    No Known Activations