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

    Neuronpedia

    INDEX
    Explanations

    harmless

    np_max-act · gemini-2.0-flash

    statements that convey benign or neutral sentiments in potentially sensitive contexts.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     thrilling
    -0.06
     数据
    -0.06
    anın
    -0.06
     �
    -0.06
    laş
    -0.06
    \"\
    -0.06
     král
    -0.06
     artillery
    -0.06
    _counter
    -0.06
    يلة
    -0.06
    POSITIVE LOGITS
     investig
    0.07
     noreferrer
    0.06
     threaded
    0.06
     brightly
    0.06
     forged
    0.06
    ought
    0.06
     Californ
    0.06
    WR
    0.06
     Những
    0.06
     unlikely
    0.06
    Act Density 0.003%

    No Known Activations