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

    Neuronpedia

    INDEX
    Explanations

    pwd

    np_max-act · gemini-2.0-flash

    The neuron primarily activates on shell interaction tokens—that is, command names (like “pwd”) and filesystem path strings (e.g. “/home/user”) in the transcript.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     lash
    -0.07
     travail
    -0.07
     моя
    -0.06
     LIMIT
    -0.06
     thresholds
    -0.06
     walked
    -0.06
    omial
    -0.06
     sculpture
    -0.06
     deck
    -0.06
     sü
    -0.06
    POSITIVE LOGITS
     pwd
    0.08
    /Q
    0.07
     woodworking
    0.07
    _pwm
    0.07
    PWD
    0.07
     GetLastError
    0.07
     Rt
    0.06
     đ
    0.06
    _BG
    0.06
    JsonIgnore
    0.06
    Act Density 0.002%

    No Known Activations