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

    Neuronpedia

    INDEX
    Explanations

    Making choices

    np_max-act · gemini-2.0-flash

    The neuron flags tokens that are clearly French (i.e. picks up French-language words).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    .Cookie
    -0.07
     wins
    -0.07
    @Configuration
    -0.06
    finder
    -0.06
    packing
    -0.06
    -connect
    -0.06
     Τζ
    -0.06
    děpodob
    -0.06
    Dismiss
    -0.06
    ogany
    -0.06
    POSITIVE LOGITS
     कड
    0.06
     unlink
    0.06
    .music
    0.06
    rab
    0.06
    auer
    0.06
    官方
    0.06
     říj
    0.06
    Intl
    0.06
    _ylabel
    0.06
     ammon
    0.06
    Act Density 0.095%

    No Known Activations