© 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. 67545
    Prev
    Next
    INDEX
    Explanations

    modal verbs followed by verbs

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

    The neuron is strongly activated by common English function words—especially personal pronouns (I, they) and auxiliary/modal verbs (will, can)—rather than content words.

    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
    ’
    -1.43
    nąć
    -1.29
    Ul
    -1.26
    ?”
    -1.22
     your
    -1.20
     nommé
    -1.20
    Frame
    -1.20
    Color
    -1.20
    It
    -1.19
     tijd
    -1.18
    POSITIVE LOGITS
    赓
    1.47
    henswürdigkeiten
    1.37
     although
    1.34
     svých
    1.34
    Begriff
    1.33
    canth
    1.27
     insanların
    1.23
     bVar
    1.23
    número
    1.22
    CCTV
    1.22
    Activations Density 0.157%

    No Known Activations