© 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. 73549
    Prev
    Next
    INDEX
    Explanations

    numbered list items

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

    The neuron strongly activates on document‐structure markers—things like section titles, headings, and numbered list items.

    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
     connection
    -1.18
     vasto
    -1.16
     is
    -1.14
     will
    -1.09
     would
    -1.08
     was
    -1.06
    kuuta
    -1.05
     base
    -1.05
     selben
    -1.04
     capacity
    -1.03
    POSITIVE LOGITS
     heartbreak
    1.17
    叹息
    1.12
    versial
    1.05
     Ді
    1.05
    Brak
    1.05
    Those
    1.03
    <>
    1.03
    不思議
    1.01
    第一个
    1.00
     sympathique
    1.00
    Activations Density 0.032%

    No Known Activations