© 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. 95408
    Prev
    Next
    INDEX
    Explanations

    code snippets

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

    The neuron activates for numeric page‐ID or index segments in URLs (i.e. digit tokens used as path components).

    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
    拶
    -1.24
     lanzó
    -1.21
    dición
    -1.18
     oublié
    -1.17
     seleção
    -1.15
     allmäh
    -1.11
     legjobb
    -1.10
     langsam
    -1.10
     laissé
    -1.09
    伀
    -1.09
    POSITIVE LOGITS
     at
    1.13
    1.09
    ？”
    1.09
     sienten
    1.06
    .”
    1.00
    ");
    1.00
     Sympathi
    0.98
    get
    0.97
     continua
    0.96
    is
    0.96
    Activations Density 0.004%
    No Known Activations