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

    Neuronpedia

    INDEX
    Explanations

    Past tense and conditionals

    np_max-act · gemini-2.0-flash

    The neuron detects conditional or causal cue words (e.g. if, when, because, then) that introduce logical or explanatory relationships.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
    ep
    -0.07
     imped
    -0.07
     bases
    -0.06
    ีซ
    -0.06
    ecial
    -0.06
    ío
    -0.06
     specials
    -0.06
     politik
    -0.06
    ):\
    -0.06
     DJs
    -0.06
    POSITIVE LOGITS
    さんが
    0.07
     LatLng
    0.07
    ивши
    0.07
    .helpers
    0.07
     leh
    0.07
    _daily
    0.07
     ylabel
    0.07
    .getCurrentUser
    0.07
    Mont
    0.06
    bons
    0.06
    Act Density 0.076%

    No Known Activations