© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 66949
    Prev
    Next
    INDEX
    Explanations

    date and code identifiers

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

    The neuron fires on digit‐only tokens (numeric sequences like version numbers, timestamps, or other all‐digit strings).

    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
    hhh
    -1.02
    -^
    -1.00
     wedi
    -0.99
    <strong>
    -0.95
    ↵↵
    -0.94
    ghh
    -0.94
    日至
    -0.92
    idez
    -0.91
     fantastic
    -0.91
    cll
    -0.89
    POSITIVE LOGITS
     носить
    1.04
     their
    1.02
    Вариан
    1.01
    ➀
    0.99
    墳
    0.98
    Хотя
    0.97
    Buat
    0.96
    Berita
    0.96
     vrst
    0.94
    lacian
    0.94
    Activations Density 0.007%

    No Known Activations