© 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. 79601
    Prev
    Next
    INDEX
    Explanations

    code documentation tags

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

    This neuron activates on documentation‐style comment markers—especially the “*” prefix and “@” annotation symbols in block comments.

    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
     suav
    -1.16
    Gewicht
    -1.01
    ERe
    -0.98
    zemu
    -0.96
     һәм
    -0.96
    팻
    -0.95
    Bref
    -0.95
     Mientras
    -0.94
     KNOWN
    -0.94
     supri
    -0.94
    POSITIVE LOGITS
     unique
    1.00
     first
    0.96
    Even
    0.96
     -
    0.95
    U
    0.94
     method
    0.94
    V
    0.92
    To
    0.91
     well
    0.90
     does
    0.90
    Activations Density 0.017%

    No Known Activations