© 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. 79806
    Prev
    Next
    INDEX
    Explanations

    described or previously published

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

    The neuron fires on numeric tokens (both integers and decimals) wherever they appear 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
    ícias
    -1.00
    什么呢
    -0.98
    ują
    -0.95
     ort
    -0.95
    typeid
    -0.90
     الحكومة
    -0.89
    ândia
    -0.89
    嬸
    -0.89
    atization
    -0.88
     verschillende
    -0.87
    POSITIVE LOGITS
     previously
    1.84
     recently
    1.55
     published
    1.52
     previous
    1.50
     recent
    1.41
     reported
    1.29
     described
    1.27
    previously
    1.24
     from
    1.21
     недавно
    1.16
    Activations Density 0.104%

    No Known Activations