© 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. 10-GEMMASCOPE-RES-131K
    7. 9727
    Prev
    Next
    INDEX
    Explanations

    in turn causes

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

    This neuron detects technical measurements and magnitude‐related descriptors (e.g. “very,” “small,” “variations,” “greater”) that occur in engineering or patent-style parameter descriptions.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/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
     ministres
    -2.03
     ruines
    -2.00
     juges
    -1.95
     pères
    -1.95
     travailleurs
    -1.89
     coelho
    -1.85
     maîtres
    -1.85
     citoyens
    -1.80
     Acesso
    -1.80
    ᾿
    -1.79
    POSITIVE LOGITS
     at
    2.08
    ଵ
    1.83
     or
    1.82
     when
    1.66
     STILL
    1.59
     When
    1.57
     begins
    1.56
    饌
    1.56
     ALWAYS
    1.55
     الجلد
    1.54
    Activations Density 1.220%

    No Known Activations