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

    Neuronpedia

    INDEX
    Explanations

    on

    np_max-act · gemini-2.0-flash

    This neuron selectively activates on the preposition “on” in scientific‐style titles of the form “Effect of X on Y.”

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    保護
    -0.08
     harvested
    -0.06
    .enterprise
    -0.06
    abling
    -0.06
     отношения
    -0.06
     玩
    -0.06
    .addTo
    -0.06
     surprised
    -0.06
     finds
    -0.06
     полі
    -0.06
    POSITIVE LOGITS
    pgsql
    0.07
     [...]↵↵
    0.07
    ственная
    0.06
     GX
    0.06
    ncy
    0.06
     egy
    0.06
     φ
    0.06
     tempt
    0.06
    �
    0.06
    ward
    0.06
    Act Density 0.027%

    No Known Activations