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

    Neuronpedia

    INDEX
    Explanations

    auxiliary verbs

    np_max-act · gemini-2.0-flash

    The neuron is strongly activated by occurrences of “to be” auxiliaries (e.g. is, are, be, will be).

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    ЎыџN
    -0.07
     participating
    -0.06
     wandering
    -0.06
    Day
    -0.06
     vet
    -0.06
    团
    -0.06
     Мор
    -0.06
     portrait
    -0.06
     typed
    -0.06
     differed
    -0.05
    POSITIVE LOGITS
    infeld
    0.07
    inheritDoc
    0.07
    орг
    0.06
    ibs
    0.06
     Switzerland
    0.06
    FW
    0.06
     ніж
    0.06
    /sw
    0.06
    latent
    0.06
    ystone
    0.06
    Act Density 0.305%

    No Known Activations