© 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. 33027
    Prev
    Next
    INDEX
    Explanations

    you can

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

    The neuron primarily detects numeric tokens (e.g. 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_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
     also
    -1.08
     también
    -1.00
     myös
    -0.96
     также
    -0.94
     acompan
    -0.93
     Republic
    -0.93
     willen
    -0.93
    Sedangkan
    -0.90
     IMP
    -0.88
    ateness
    -0.87
    POSITIVE LOGITS
     simply
    1.69
     just
    1.54
     through
    1.40
     by
    1.28
     wystarczy
    1.21
    只要
    1.20
     merely
    1.18
     simplement
    1.16
     बस
    1.16
     albeit
    1.13
    Activations Density 0.076%

    No Known Activations