© 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. 92941
    Prev
    Next
    INDEX
    Explanations

    inner div or North Pole

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

    The neuron activates on numeric tokens—i.e. it detects any literal containing digits.

    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
    菊
    -1.05
     then
    -1.02
     потім
    -0.98
    äumt
    -0.94
    alisis
    -0.94
     實
    -0.90
     detailed
    -0.84
    ！）
    -0.82
    洶
    -0.82
    omaly
    -0.82
    POSITIVE LOGITS
     見える
    0.95
    νει
    0.90
    belline
    0.90
     Sloveniji
    0.90
    combs
    0.89
    horts
    0.88
     instalaciones
    0.88
     Naciones
    0.87
    ’
    0.85
    unzel
    0.85
    Activations Density 0.007%

    No Known Activations