© 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. 22-GEMMASCOPE-RES-131K
    7. 13568
    Prev
    Next
    INDEX
    Explanations

    specialized, specific, or themed topics

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

    The neuron activates on domain‐specific content words—especially substantive nouns and jargon that denote specialized topics or projects.

    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
     سبتمبر
    -1.00
    gown
    -1.00
    vintage
    -1.00
    lounge
    -0.93
     نوفمبر
    -0.90
    varu
    -0.90
    ложена
    -0.89
    💯
    -0.89
     JULY
    -0.88
    鏁
    -0.88
    POSITIVE LOGITS
     specialized
    1.62
     such
    1.55
     specific
    1.52
     specialization
    1.35
     specialised
    1.25
     específico
    1.20
     speciality
    1.20
     specialty
    1.20
     specialize
    1.17
     spécialistes
    1.13
    Activations Density 0.127%

    No Known Activations