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

    Neuronpedia

    INDEX
    Explanations

    punctuation

    np_max-act · gemini-2.0-flash

    tokens that are part of the user's input (i.e., user-role prompt text).

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    New Auto-Interp
    Negative Logits
    項目
    -0.08
    _ser
    -0.07
    攻克
    -0.07
    .master
    -0.07
    这些问题
    -0.07
     homage
    -0.07
    igers
    -0.06
     처음
    -0.06
    どのように
    -0.06
    HEAD
    -0.06
    POSITIVE LOGITS
    _fw
    0.07
    (stdin
    0.07
    ]=='
    0.07
     Johan
    0.07
    肢
    0.06
    orarily
    0.06
     figur
    0.06
    萍
    0.06
     Robbins
    0.06
    0.06
    Act Density 0.052%

    No Known Activations