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

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 128175
    Prev
    Next
    INDEX
    Explanations

    The neuron fires on standalone numeric tokens—i.e. it detects numbers (sequences of digits) in the text.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408

    Sinopia

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_22/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
     that
    -1.38
     modelleri
    -1.08
     chré
    -1.00
     trouvent
    -1.00
     özellikleri
    -0.98
     religieux
    -0.98
    gaande
    -0.97
    diğimiz
    -0.97
     hedef
    -0.96
     всё
    -0.96
    POSITIVE LOGITS
    lidae
    1.19
    britanni
    1.17
    вшись
    1.03
     installations
    1.02
    úrese
    1.02
     Forsch
    1.01
     relacionada
    1.00
    房屋
    0.99
     be
    0.98
     Fama
    0.96
    Activations Density 0.003%

    No Known Activations