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

    Neuronpedia

    INDEX
    Explanations

    repellent

    np_max-act · gemini-2.0-flash

    The neuron detects special control or metadata tokens that mark the boundaries and headers (e.g. <|start_header_id|>, question marks in that context) in the transcript.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    *);↵↵
    -0.07
    .pointer
    -0.06
    сте
    -0.06
     query
    -0.06
     goddess
    -0.06
    _find
    -0.06
    ponder
    -0.06
    没
    -0.06
    (direction
    -0.06
    ském
    -0.06
    POSITIVE LOGITS
     repell
    0.07
    .':
    0.07
    	Double
    0.07
     PubMed
    0.06
     Republican
    0.06
     Austin
    0.06
    although
    0.06
     entityManager
    0.06
     Cron
    0.06
     Effect
    0.06
    Act Density 0.003%

    No Known Activations