© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 38873
    Prev
    Next
    INDEX
    Explanations

    enactivism

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

    This neuron fires on document‐structure markers—things like section headings, underlining/separator lines, numbered steps, “Requirements:” labels, and other formatting cues that mark out the structure of a technical or markdown‐style document.

    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
    <eos>
    -1.32
     more
    -1.10
    อีก
    -1.01
     Griechenland
    -0.95
     another
    -0.95
    以上
    -0.95
    tigkeits
    -0.95
    stringBuilder
    -0.94
    ">:
    -0.93
    "';
    -0.90
    POSITIVE LOGITS
    まずは
    1.11
    废话
    0.96
     trước
    0.96
    rô
    0.92
    あっという
    0.89
     centru
    0.89
     offenbar
    0.88
    董事长
    0.88
     прежде
    0.88
    点此
    0.85
    Activations Density 0.174%

    No Known Activations