© 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. 34-GEMMASCOPE-RES-131K
    7. 25911
    Prev
    Next
    INDEX
    Explanations

    templates, jig, University

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

    The neuron fires on lines of formal document metadata or headings—things like paper titles, section headers, institutional affiliations, and other all-caps or specially formatted label tokens.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
    Teléfono
    -0.78
    clav
    -0.74
    cop
    -0.73
     crouch
    -0.69
     usai
    -0.69
    GrantedAuthority
    -0.69
    oblast
    -0.68
     godt
    -0.68
    Marshaller
    -0.68
    Craig
    -0.66
    POSITIVE LOGITS
    nefs
    0.90
    밍
    0.83
     Szw
    0.77
    Taco
    0.76
     vort
    0.76
     vore
    0.73
     Burkina
    0.71
     ブラウス
    0.71
    术
    0.71
    stoma
    0.71
    Activations Density 0.014%

    No Known Activations