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

    Neuronpedia

    INDEX
    Explanations

    severe

    np_max-act · gemini-2.0-flash

    This neuron activates on the adjective “severe,” marking mentions of high disease or condition severity.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    Fake
    -0.07
     Strap
    -0.07
     Outcome
    -0.07
    jac
    -0.06
    Ey
    -0.06
     Chop
    -0.06
    tryside
    -0.06
     salad
    -0.06
    igraph
    -0.06
    879
    -0.06
    POSITIVE LOGITS
    .scalatest
    0.07
     severe
    0.07
     možnosti
    0.07
    icago
    0.06
     derin
    0.06
     نبود
    0.06
    fillable
    0.06
    `(
    0.06
     рассказ
    0.06
    ђ
    0.06
    Act Density 0.023%

    No Known Activations