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

    Neuronpedia

    INDEX
    Explanations

    fight scenes

    np_max-act · gemini-2.0-flash

    detailed descriptions of combat scenarios involving magic and mythology.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8

    The neuron activates on tokens that are part of the assistant’s generated story or narrative responses.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     cannons
    -0.07
    empre
    -0.07
     vend
    -0.07
     uninstall
    -0.06
    emade
    -0.06
    __);↵↵
    -0.06
     Bret
    -0.06
     Mattis
    -0.06
     institutes
    -0.06
    .fft
    -0.06
    POSITIVE LOGITS
    Geo
    0.07
     WARRANTIES
    0.07
     завдання
    0.06
    Leader
    0.06
    .protobuf
    0.06
     FORMAT
    0.06
    (href
    0.06
    .lwjgl
    0.06
     нуж
    0.06
     cgi
    0.06
    Act Density 0.013%

    No Known Activations