© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it
    Neuronpedia
    Jacobian LensNEW
    Natural Language
    Autoencoders
    MAY 26    Assistant AxisJAN 26Circuit TracerAPR 26SteerSAE EvalsExportsinterp-engineNEW Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 24949
    Prev
    Next
    INDEX
    Explanations

    possessive constructions

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

    The neuron consistently lights up on structural markers and the very first content words of list items or new sections—e.g. colons, item‐numbers, and the initial capitalized words at the start of a bullet or numbered point.

    oai_token-act-pair · o4-mini
    Triggered 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
    Negative Logits
    très
    -1.31
     Corbu
    -1.29
    茳
    -1.23
    tront
    -1.23
     ongles
    -1.23
    peut
    -1.20
    archiwizowane
    -1.19
    ſt
    -1.18
     oreille
    -1.18
     nke
    -1.18
    POSITIVE LOGITS
     when
    1.65
     they
    1.55
     I
    1.45
    When
    1.42
     It
    1.33
     these
    1.33
    The
    1.32
     When
    1.30
     This
    1.30
     you
    1.27
    Activations Density 0.011%

    No Known Activations