© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 75292
    Prev
    Next
    INDEX
    Explanations

    molecular spintronics, regional anesthesia, [differences

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

    This neuron detects numeric tokens (standalone numbers in text, such as page numbers, equation coefficients, years, etc.).

    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
     jesu
    -1.13
     from
    -1.06
     salade
    -1.00
     himmel
    -0.98
     död
    -0.94
     vezet
    -0.94
     kanton
    -0.93
     berge
    -0.91
     annen
    -0.90
     vek
    -0.90
    POSITIVE LOGITS
     ২০১
    1.02
     Hadoop
    0.92
     Mandatory
    0.90
    inWeight
    0.90
     impos
    0.89
     scoprire
    0.89
     Mindfulness
    0.88
    đồng
    0.88
     finanzi
    0.88
     and
    0.87
    Activations Density 0.000%

    No Known Activations