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

    Neuronpedia

    INDEX
    Explanations

    code structure and syntax

    np_acts-logits-general · gemini-2.5-flash-lite

    This neuron detects numeric literals—especially multi-digit or floating-point number tokens.

    oai_token-act-pair · gpt-5-miniTriggered by @vetterc0
    New Auto-Interp
    Negative Logits
     Veneto
    0.47
     Intra
    0.42
     Tetra
    0.42
     Izv
    0.41
    文献
    0.39
     návr
    0.39
    兜
    0.38
     sleeves
    0.38
     drafts
    0.38
     eszköz
    0.38
    POSITIVE LOGITS
    ਰੀ
    0.43
     il
    0.42
    keun
    0.42
    ]]]
    0.42
    kw
    0.41
    ły
    0.40
    Br
    0.40
    alaikumsalam
    0.40
    kele
    0.40
     لري
    0.40
    Act Density 0.000%

    No Known Activations