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

    Neuronpedia

    INDEX
    Explanations

    anger, annoyance, upset

    np_acts-logits-general · gemini-2.5-flash-lite

    the presence of anger-related words or strong angry emotion (tokens expressing anger/frustration).

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    New Auto-Interp
    Negative Logits
     parach
    0.43
     adventurers
    0.42
     parachute
    0.42
     nimble
    0.42
    W
    0.42
     log
    0.42
     adventurous
    0.42
     সুবিধ
    0.42
     bokeh
    0.40
     Adventure
    0.40
    POSITIVE LOGITS
     anger
    1.63
     angry
    1.59
     angrily
    1.57
    愤怒
    1.49
     colère
    1.47
    怒
    1.34
     गुस्सा
    1.31
     enraged
    1.30
     गुस्से
    1.30
    Angry
    1.30
    Act Density 0.162%

    No Known Activations