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

    Neuronpedia

    INDEX
    Explanations

    Affection, care, appreciation

    np_max-act · gemini-2.0-flash

    The neuron fires on verbs of perception or affection in past or perfect constructions—e.g. “has seen,” “loved him.”

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     integral
    -0.07
     себе
    -0.06
     вперед
    -0.06
     intl
    -0.06
    射
    -0.06
     Integral
    -0.06
    -0.06
    ump
    -0.06
     huyết
    -0.06
     internals
    -0.06
    POSITIVE LOGITS
     sessiz
    0.07
    .gson
    0.07
    akeup
    0.06
    然
    0.06
     Jak
    0.06
     Kaz
    0.06
    TARGET
    0.06
     Fac
    0.06
    .soft
    0.06
     CascadeType
    0.06
    Act Density 0.026%

    No Known Activations