© 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
    2. Gemma-2-27B
    3. 22-GEMMASCOPE-RES-131K
    4. 127868
    Prev
    Next
    INDEX
    Explanations

    The neuron responds strongly to the apostrophe character (single‐quote) in words, i.e. contractions or possessives.

    oai_token-act-pair · o4-mini
    Triggered by @jyhe0408

    earl or eleanor

    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_22/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Negative Logits
     vuitton
    -1.45
     sopa
    -1.33
     Réunion
    -1.31
     })}
    -1.28
     venez
    -1.28
     produkter
    -1.26
     TRANSPORTATION
    -1.22
     munici
    -1.20
     hech
    -1.20
    Географи
    -1.19
    POSITIVE LOGITS
     and
    2.03
     then
    1.57
     I
    1.29
     didn
    1.27
     how
    1.23
     why
    1.22
    '
    1.20
     that
    1.19
     who
    1.17
     fanden
    1.16
    Activations Density 0.001%

    No Known Activations