© 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. 13863
    Prev
    Next
    INDEX
    Explanations

    full-length, big-budget, old-fashion, low-penetrance

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

    This neuron strongly activates on dash tokens (hyphens or em-dashes) used as separators or line-break markers.

    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
     of
    -1.64
    もあって
    -1.63
    而非
    -1.46
    见状
    -1.45
    日を
    -1.45
    你们的
    -1.43
    Largeur
    -1.37
    two
    -1.35
    四个
    -1.33
     votación
    -1.33
    POSITIVE LOGITS
    '
    1.40
    為に
    1.39
    "
    1.34
    ,
    1.30
     n
    1.30
    ’
    1.29
    );
    1.27
    Ĝ
    1.22
    ほうが
    1.20
     some
    1.20
    Activations Density 0.102%

    No Known Activations