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

    Neuronpedia

    INDEX
    Explanations

    Measurement of quality

    np_max-act · gemini-2.0-flash

    The neuron fires primarily on descriptive safety‐ and stability‐related terms in chemical contexts (e.g. “safe,” “stable,” “non-reactive,” “low toxicity,” “highly soluble,” etc.).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    IVO
    -0.06
     SECURITY
    -0.06
     lesbienne
    -0.06
    .arg
    -0.06
    ionale
    -0.06
    jen
    -0.06
     pathways
    -0.06
     medicine
    -0.06
    buie
    -0.06
    -One
    -0.06
    POSITIVE LOGITS
     잘
    0.07
     Ticket
    0.07
    嘉
    0.07
    되었다
    0.07
     Partnership
    0.07
     základní
    0.07
     descended
    0.06
    �
    0.06
    Interior
    0.06
     athletics
    0.06
    Act Density 0.042%

    No Known Activations