© 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. 9477
    Prev
    Next
    INDEX
    Explanations

    UI elements, technical terms, biological groups

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

    user interface elements such as buttons, search boxes, controls, and interface components.

    oai_token-act-pair · claude-3-7-sonnet-20250219Triggered by @neilrathi

    The neuron activates on domain-specific technical nouns (e.g. UI elements like “button,” hardware terms like “pin,” experimental parameters like “intensity” or “strains”).

    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
     литература
    -1.09
     from
    -1.08
     obicei
    -1.06
     mewakili
    -1.05
     koszulka
    -1.04
    émission
    -1.04
     gikk
    -1.03
    p
    -1.01
     синд
    -1.01
    from
    -1.01
    POSITIVE LOGITS
    Bienvenido
    1.35
    𓁹
    1.26
    
    1.24
    imágenes
    1.23
     Anhäng
    1.22
     này
    1.18
    Buona
    1.16
     then
    1.13
    cardigan
    1.13
    です
    1.13
    Activations Density 0.098%

    No Known Activations