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

    Neuronpedia

    INDEX
    Explanations

    structured format, underlying hardware, long-term

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Negative Logits
     awoke
    0.41
    ます
    0.41
    isode
    0.41
    🤎
    0.40
    COVID
    0.38
    iov
    0.38
    ⁇
    0.38
     woke
    0.37
    jasmine
    0.37
    discovery
    0.37
    POSITIVE LOGITS
     katva
    0.44
     Relating
    0.41
     späteren
    0.40
     Το
    0.40
     http
    0.39
     Response
    0.39
     Any
    0.39
     Copies
    0.38
     With
    0.38
     Systems
    0.38
    Act Density 0.009%

    No Known Activations