© 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
    2. Gemma-2-27B
    3. 34-GEMMASCOPE-RES-131K
    4. 45236
    Prev
    Next
    INDEX
    Explanations

    HTML PUBLIC DOCTYPE, cultivar, inner ear

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

    The neuron fires on formal, document‐structure tokens such as section titles or key legal/technical terms (e.g. “PUBLIC,” “cultivar,” “Segunda,” “División,” “provided,” “notice”) that mark headings, license clauses, or other metadata in technical and legal texts.

    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
     interés
    -0.67
    engeance
    -0.66
     restaurant
    -0.66
    ⑅
    -0.65
     diner
    -0.65
     Erwachsenen
    -0.63
    attori
    -0.63
     teleno
    -0.63
    entran
    -0.63
     Gefahr
    -0.63
    POSITIVE LOGITS
    ستگی
    0.79
     Mete
    0.78
    žiu
    0.75
    ZING
    0.72
    联网
    0.71
    迈
    0.69
    Gravity
    0.68
     Rice
    0.67
     rong
    0.67
     Rusty
    0.66
    Activations Density 0.085%

    No Known Activations