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

    Neuronpedia

    INDEX
    Explanations

    repeating old arguments

    np_max-act · gemini-2.0-flash

    spots meta-commentary about repetition or rehashing of arguments or topics (mentions that something is being repeated or raised again).

    oai_token-act-pair · gpt-5-miniTriggered by @yooniel31
    New Auto-Interp
    Negative Logits
     depend
    -0.07
    istringstream
    -0.07
     você
    -0.07
    .assertNotNull
    -0.06
    íš
    -0.06
     quarry
    -0.06
    ARGE
    -0.06
        
    -0.06
    uff
    -0.06
    awe
    -0.06
    POSITIVE LOGITS
     Bibli
    0.07
    ducation
    0.07
    iphery
    0.06
     Jorge
    0.06
    .populate
    0.06
    crud
    0.06
    SELF
    0.06
    _CITY
    0.06
     Ibn
    0.06
     ElseIf
    0.06
    Act Density 0.140%

    No Known Activations