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

    Neuronpedia

    INDEX
    Explanations

    back

    np_max-act · gemini-2.0-flash

    This neuron selectively activates on the token “back” when it appears in instructions about triple back-ticks.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     �
    -0.07
     Champions
    -0.07
     ル
    -0.07
     formats
    -0.06
    2
    -0.06
    بوب
    -0.06
     Modifications
    -0.06
     Chromium
    -0.06
    $k
    -0.06
     champions
    -0.06
    POSITIVE LOGITS
    .isAuthenticated
    0.07
     loi
    0.07
    BOOT
    0.06
     rural
    0.06
    ]])↵
    0.06
    Git
    0.06
    머
    0.06
    ');↵↵↵
    0.06
    familia
    0.06
    [__
    0.06
    Act Density 0.001%

    No Known Activations