© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 34972
    Prev
    Next
    INDEX
    Explanations

    information, services, communications

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

    The neuron fires on the typical legal‐compliance and privacy‐policy phrasing—words like “service,” “information,” “you/your,” “for,” etc., that appear in formal policy clauses.

    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
     not
    -1.30
     to
    -1.15
     will
    -1.13
     instead
    -1.09
     every
    -1.09
     tonight
    -1.09
     just
    -1.05
     should
    -1.04
     now
    -1.03
     from
    -1.00
    POSITIVE LOGITS
     our
    1.43
     Manfaat
    1.40
     amélior
    1.32
    containsKey
    1.27
    ignement
    1.27
    我們
    1.25
    Mű
    1.23
     you
    1.22
     väli
    1.22
     parfois
    1.20
    Activations Density 0.011%

    No Known Activations