© 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. 22-GEMMASCOPE-RES-131K
    7. 104311
    Prev
    Next
    INDEX
    Explanations

    numbers and measurements

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

    The neuron activates on numeric tokens (individual digits 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_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
     luzes
    -1.90
    轳
    -1.84
     hilfre
    -1.84
     geprü
    -1.83
     schnellen
    -1.82
     Hinblick
    -1.81
    不
    -1.78
    傒
    -1.77
    irical
    -1.76
    糨
    -1.76
    POSITIVE LOGITS
    大学の
    1.70
    『
    1.63
    槩
    1.52
    躂
    1.48
    这一
    1.46
    自己的
    1.45
    MENTE
    1.44
    ть
    1.44
     افتراض
    1.44
    참고
    1.43
    Activations Density 0.022%

    No Known Activations