© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 64584
    Prev
    Next
    INDEX
    Explanations

    been interrupted, enhanced skin, Sandy railway

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

    The neuron strongly activates on numeric tokens—digits, number sequences, year-like numbers, and related symbols (e.g. hyphens in dates).

    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
     over
    -0.96
     to
    -0.95
     in
    -0.90
     about
    -0.87
     consommate
    -0.84
    дено
    -0.83
     Catholics
    -0.82
    詐欺
    -0.81
     a
    -0.81
    temperatura
    -0.81
    POSITIVE LOGITS
     obsah
    1.00
    pathy
    1.00
     飲み
    0.97
    ENCIES
    0.95
     dutch
    0.95
     legales
    0.95
     خواهد
    0.94
    PTION
    0.93
    HOW
    0.93
     любы
    0.93
    Activations Density 0.003%

    No Known Activations