© 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. 22-GEMMASCOPE-RES-131K
    7. 65581
    Prev
    Next
    INDEX
    Explanations

    programming structures, markup, and code definitions

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

    This neuron is sensitive to HTML table markup (e.g. “<tr>”, “<tbody>”) and other table‐related structural tokens.

    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
     as
    -1.75
     their
    -1.74
     from
    -1.71
     which
    -1.53
     these
    -1.50
     what
    -1.48
     because
    -1.48
     by
    -1.44
     When
    -1.42
     where
    -1.38
    POSITIVE LOGITS
     terrific
    1.43
    僰
    1.30
    耵
    1.30
     wondrous
    1.28
     dix
    1.27
     détect
    1.27
    typically
    1.25
    generally
    1.24
     commencer
    1.23
     insanely
    1.22
    Activations Density 0.000%

    No Known Activations