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

    Neuronpedia

    INDEX
    Explanations

    code punctuation

    np_max-act · gemini-2.0-flash

    This neuron detects the “.” token used for code punctuation (e.g. property/method access and string‐concatenation) in programming languages.

    oai_token-act-pair · o4-miniTriggered by @xinyanhu8
    New Auto-Interp
    Negative Logits
     pirates
    -0.07
     judgments
    -0.07
     reefs
    -0.07
     Feedback
    -0.06
    ώ
    -0.06
    ifs
    -0.06
    occer
    -0.06
    quivo
    -0.06
    !<
    -0.06
     Related
    -0.06
    POSITIVE LOGITS
     бра
    0.06
    .Ret
    0.06
    ?";↵
    0.06
    .OrdinalIgnoreCase
    0.06
     becer
    0.06
     *}↵↵
    0.06
     позвол
    0.06
     backpage
    0.06
    ));
↵
    0.06
    させ
    0.06
    Act Density 0.015%

    No Known Activations