© 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. 25536
    Prev
    Next
    INDEX
    Explanations

    list like items or code

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

    the neuron activates on numeric tokens (digits or standalone numbers) 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_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
     pięk
    -1.34
     detalhe
    -1.30
     brinquedo
    -1.30
     içerisinde
    -1.29
     maravilh
    -1.26
     its
    -1.25
    ようになる
    -1.24
     princípio
    -1.23
    ところに
    -1.23
     início
    -1.23
    POSITIVE LOGITS
     in
    1.39
    ℎ
    1.33
     could
    1.30
     had
    1.27
     by
    1.23
     (
    1.23
    は
    1.22
    というわけで
    1.16
    もし
    1.14
     MemoryStream
    1.14
    Activations Density 0.000%

    No Known Activations