© 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. 54419
    Prev
    Next
    INDEX
    Explanations

    Chief Judge

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

    The neuron is specialized for spotting structured “metadata” tokens—dates, digits, agency tags (e.g. “/TASS/”), reference numbers, bylines, erratum codes, etc.—rather than normal running text.

    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
    associated
    -1.02
     soos
    -0.94
    ֯
    -0.90
    ünsche
    -0.85
     matic
    -0.84
    を作って
    -0.83
     Robson
    -0.83
    謝謝
    -0.83
    袛
    -0.82
    λών
    -0.81
    POSITIVE LOGITS
     with
    1.09
     as
    0.97
     является
    0.84
    bcryptjs
    0.84
     farmland
    0.83
     comes
    0.79
    óp
    0.79
    킴
    0.78
    越來越
    0.77
    tiftung
    0.76
    Activations Density 0.006%

    No Known Activations