© 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. 10-GEMMASCOPE-RES-131K
    7. 11506
    Prev
    Next
    INDEX
    Explanations

    URLs and abbreviations

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

    URLs, specifically blog post URLs with date paths and domain names.

    oai_token-act-pair · claude-4-5-sonnetTriggered by @bfgcn7frf5

    This neuron detects numeric tokens (e.g. digits or numbers often found in URLs, dates, or metadata).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     you
    -2.31
     It
    -1.70
     Instead
    -1.64
    嶇
    -1.59
     не
    -1.52
     we
    -1.47
     Using
    -1.47
    ступления
    -1.45
     Not
    -1.43
    CHAFT
    -1.43
    POSITIVE LOGITS
    \
    2.03
    3
    1.80
    8
    1.75
    櫚
    1.63
     Ausdrücke
    1.63
    a
    1.58
    姈
    1.56
     liknande
    1.55
    9
    1.54
     QUART
    1.53
    Activations Density 0.042%

    No Known Activations