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

    Neuronpedia

    INDEX
    Explanations

    feel

    np_max-act · gemini-2.0-flash

    This neuron is sensitive to occurrences of the word “feel” (and its immediate context), flagging descriptions of how something feels.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    uke
    -0.06
    Sub
    -0.06
    出す
    -0.06
     rectangles
    -0.06
     saint
    -0.06
     взаєм
    -0.06
     Artem
    -0.06
    _two
    -0.06
    acker
    -0.06
    -mails
    -0.06
    POSITIVE LOGITS
    feel
    0.07
    -original
    0.07
     gist
    0.07
     vibe
    0.07
    Characters
    0.07
    eur
    0.07
     countryside
    0.07
     conclus
    0.06
     endPoint
    0.06
    _CNTL
    0.06
    Act Density 0.005%

    No Known Activations