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

    Neuronpedia

    INDEX
    Explanations

    error message content

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     βιβ
    0.40
    clamps
    0.40
    Referring
    0.39
    DEP
    0.39
    Bracket
    0.39
    ಲಿನ
    0.39
     підпри
    0.39
    ourcen
    0.39
     وګ
    0.39
     வழங்கு
    0.39
    POSITIVE LOGITS
     message
    0.68
    getMessage
    0.58
    message
    0.54
     mensagem
    0.47
    Message
    0.45
     mensaje
    0.45
    原因
    0.45
     संदेश
    0.45
    消息
    0.44
    の原因
    0.43
    Act Density 0.004%

    No Known Activations