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

    Neuronpedia

    INDEX
    Explanations

    Expressing negative sentiment/controversy

    np_max-act · gemini-2.0-flash

    discussions about architectural projects and their controversies.

    oai_token-act-pair · gpt-4o-miniTriggered by @xinyanhu8

    The main thing this neuron does is spot subjective or evaluative language—words and phrases that express personal feelings, reactions, opinions, or directives.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    ená
    -0.08
    _launcher
    -0.07
    ework
    -0.07
     Passport
    -0.06
    RETURN
    -0.06
     сті
    -0.06
     soruml
    -0.06
     tenemos
    -0.06
     economist
    -0.06
    _bind
    -0.06
    POSITIVE LOGITS
    errorMessage
    0.07
     τρο
    0.07
    .property
    0.07
     informant
    0.06
     DUI
    0.06
    jquery
    0.06
     OnTriggerEnter
    0.06
    _BAL
    0.06
     Společ
    0.06
     horsepower
    0.06
    Act Density 0.127%

    No Known Activations