© 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. 19808
    Prev
    Next
    INDEX
    Explanations

    click on the link

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

    The neuron activates on numeric tokens, detecting numbers (including integers and decimals) in the text.

    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
     Ahnung
    -1.80
     you
    -1.75
    ти
    -1.68
    2
    -1.67
     my
    -1.67
     tudi
    -1.63
    :
    -1.62
     sly
    -1.60
     Kennt
    -1.59
     kurzen
    -1.56
    POSITIVE LOGITS
     and
    2.22
     who
    2.02
    ！（
    1.62
     what
    1.59
     seluruh
    1.54
    !),
    1.54
     how
    1.52
     tenido
    1.50
    يدة
    1.49
     käyttää
    1.48
    Activations Density 0.081%

    No Known Activations