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

    Neuronpedia

    INDEX
    Explanations

    -

    np_max-act · gemini-2.0-flash

    This neuron fires on shell command options (tokens like “-”, “-s”, “-f”) associated with Unix file-operation commands (e.g. ln, rm).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    _contains
    -0.07
    INCLUDED
    -0.07
    ρών
    -0.07
    astes
    -0.06
    “In
    -0.06
    suz
    -0.06
    ulo
    -0.06
    otts
    -0.06
    \Admin
    -0.06
     Diff
    -0.06
    POSITIVE LOGITS
     mực
    0.07
    -_
    0.07
     eius
    0.07
    UNIT
    0.07
    .Gray
    0.07
     xứ
    0.07
    .WHITE
    0.06
    ों
    0.06
     breathable
    0.06
     getpid
    0.06
    Act Density 0.003%

    No Known Activations