© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 80763
    Prev
    Next
    INDEX
    Explanations

    numbers and decimals

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

    The neuron strongly activates on sequences of digits (including zeros) and decimal points—that is, numeric tokens (numbers and decimal numbers).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     can
    -1.98
     to
    -1.79
     trov
    -1.55
     può
    -1.45
    ）。
    -1.43
     berücksich
    -1.42
    ことができます
    -1.41
    aziland
    -1.36
     cannot
    -1.36
    quilibrium
    -1.33
    POSITIVE LOGITS
    -
    1.48
     октя
    1.46
     ноябре
    1.45
     belie
    1.45
    UNS
    1.39
     eases
    1.33
    ,
    1.28
    wamy
    1.27
    sla
    1.27
     ,
    1.26
    Activations Density 0.017%

    No Known Activations