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

    Neuronpedia

    INDEX
    Explanations

    medical scans

    np_max-act · gemini-2.0-flash

    This neuron activates on pathology‐style descriptive adjectives ending in “-ic” or “-oid” (e.g., necrotic, hemorrhagic, rhabdoid).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    usi
    -0.07
     External
    -0.07
    SKTOP
    -0.07
     бизнес
    -0.07
     Month
    -0.07
     Heat
    -0.06
    \D
    -0.06
     narrowly
    -0.06
    iability
    -0.06
    Van
    -0.06
    POSITIVE LOGITS
    swagger
    0.06
     参
    0.06
     BLL
    0.06
    .sh
    0.06
     对
    0.06
    .Is
    0.06
     {},↵
    0.06
    (NS
    0.06
     ARM
    0.06
    .Unicode
    0.06
    Act Density 0.017%

    No Known Activations