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

    Neuronpedia

    INDEX
    Explanations

    en

    np_max-act · gemini-2.0-flash

    This neuron detects French verbs and verbal phrases meaning "to put in place / implement" (e.g., "mettre en place" and its variants).

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    New Auto-Interp
    Negative Logits
    ilater
    -0.08
    stackoverflow
    -0.06
    That
    -0.06
    -nav
    -0.06
    enticator
    -0.06
     pastor
    -0.06
    _backend
    -0.06
    (P
    -0.06
    ous
    -0.06
    aybe
    -0.06
    POSITIVE LOGITS
    masının
    0.07
     xây
    0.07
     král
    0.06
    zen
    0.06
     Spoj
    0.06
     monks
    0.06
     série
    0.06
    (sim
    0.06
     měly
    0.06
    ’ın
    0.06
    Act Density 0.012%

    No Known Activations