© 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 EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 9560
    Prev
    Next
    INDEX
    Explanations

    modal verbs and knowing/saying

    np_acts-logits-general · gemini-2.5-flash-lite
    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
    𝟐
    -1.38
    maximize
    -1.36
    стон
    -1.35
    🇲
    -1.35
     LAKE
    -1.30
    tison
    -1.27
    šana
    -1.26
    endrait
    -1.25
     anos
    -1.25
    ITY
    -1.25
    POSITIVE LOGITS
     how
    2.45
     where
    1.93
     who
    1.88
     when
    1.67
     decide
    1.51
     whether
    1.48
     Хотя
    1.45
     know
    1.43
     knows
    1.43
     this
    1.42
    Activations Density 0.011%

    No Known Activations