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

    Neuronpedia

    INDEX
    Explanations

    programming or code

    np_max-act · gemini-2.0-flash

    The neuron lights up on imperative or instructional action words (like “Fine,” “Plot,” “Upload,” etc.) that introduce steps or commands in technical how-to explanations.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     く
    -0.07
    ynomial
    -0.07
     despite
    -0.07
     shuts
    -0.07
     آورد
    -0.06
    ennon
    -0.06
    NoArgsConstructor
    -0.06
     neoliberal
    -0.06
     Cache
    -0.06
     },
↵
    -0.06
    POSITIVE LOGITS
    ARY
    0.07
    ным
    0.06
    ері
    0.06
    _cre
    0.06
    (updated
    0.06
     snatch
    0.06
     Lat
    0.06
    siniz
    0.06
    督
    0.06
    ньому
    0.06
    Act Density 0.160%

    No Known Activations