© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 58401
    Prev
    Next
    INDEX
    Explanations

    markdown headings and content

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

    This neuron detects Markdown headings that consist of (or begin with) a numeric value, often indicating section numbers or priorities.

    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
     órganos
    -1.87
     of
    -1.62
     fijne
    -1.59
     immense
    -1.59
     deliciously
    -1.49
     prachtige
    -1.47
     trabajadores
    -1.47
    begins
    -1.45
     assures
    -1.45
     velmi
    -1.44
    POSITIVE LOGITS
    There
    1.98
     There
    1.96
     When
    1.91
    what
    1.70
    Getting
    1.66
    What
    1.62
     Extremely
    1.60
     Because
    1.58
     Before
    1.55
    Some
    1.53
    Activations Density 0.003%

    No Known Activations