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

    Neuronpedia

    INDEX
    Explanations

    previously

    np_max-act · gemini-2.0-flash

    This neuron detects novelty-claim phrases about something never existing or being said/seen before (e.g. “never been said before,” “never seen before”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     أف
    -0.07
    马
    -0.07
     sour
    -0.06
    NotificationCenter
    -0.06
     jab
    -0.06
    Nothing
    -0.06
    plugins
    -0.06
    _REQUIRED
    -0.06
    (buffer
    -0.06
    .gf
    -0.06
    POSITIVE LOGITS
     annotated
    0.07
     �
    0.07
     mond
    0.07
     \<^
    0.06
     carcin
    0.06
     espec
    0.06
     할인
    0.06
     misogyn
    0.06
    _AspNet
    0.06
     Registers
    0.06
    Act Density 0.020%

    No Known Activations