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

    Neuronpedia

    INDEX
    Explanations

    was

    np_max-act · gemini-2.0-flash

    This neuron detects positive, subjective evaluative words and intensifiers (e.g., “awesome,” “incredible,” “great,” “fun”).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    iconductor
    -0.07
     tz
    -0.07
    orneys
    -0.06
    Rendering
    -0.06
    .setX
    -0.06
    دی
    -0.06
    question
    -0.06
    uild
    -0.06
    CTX
    -0.06
    cı
    -0.06
    POSITIVE LOGITS
    股
    0.07
     misses
    0.06
     Eş
    0.06
    ありがとう
    0.06
     outrageous
    0.06
     bella
    0.06
    "user
    0.06
    育
    0.06
     seemed
    0.06
     sessions
    0.06
    Act Density 0.035%

    No Known Activations