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

    Neuronpedia

    INDEX
    Explanations

    modal verbs

    np_max-act · gemini-2.0-flash

    This neuron mainly fires on occurrences of the modal auxiliary “would,” marking conditional or future-tense constructions.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     Spare
    -0.07
     substring
    -0.07
    issa
    -0.07
     offsetX
    -0.07
     Melissa
    -0.07
     encountering
    -0.07
     viewBox
    -0.06
     divides
    -0.06
    .df
    -0.06
     exponential
    -0.06
    POSITIVE LOGITS
    .head
    0.07
    .SizeMode
    0.06
     Ông
    0.06
    those
    0.06
    leri
    0.06
    	score
    0.06
    ピ
    0.06
    ドラ
    0.06
     scp
    0.06
     shimmer
    0.06
    Act Density 0.287%

    No Known Activations