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

    Neuronpedia

    INDEX
    Explanations

    conversational/opinionated text

    np_max-act · gemini-2.0-flash

    instances of direct address in dialogue—especially the word "you" and character-name mentions used to speak to or about the interlocutor.

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    New Auto-Interp
    Negative Logits
     Coc
    -0.07
     nao
    -0.07
    งน
    -0.07
    dıktan
    -0.07
    untlet
    -0.06
     että
    -0.06
    ｀
    -0.06
    �
    -0.06
    _even
    -0.06
     nakne
    -0.06
    POSITIVE LOGITS
     invoke
    0.07
    obbies
    0.07
     adversity
    0.07
    disconnect
    0.06
     ugl
    0.06
    离
    0.06
    _cores
    0.06
     SEEK
    0.06
    ffic
    0.06
     exploits
    0.06
    Act Density 0.068%

    No Known Activations