© 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. 10-GEMMASCOPE-RES-131K
    7. 98646
    Prev
    Next
    INDEX
    Explanations

    rimraf, battling, Metal, How, Q

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

    The neuron fires on the “/-/” token pattern (slash–dash–slash), as found in package‐URL path segments.

    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_10/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Negative Logits
     asiática
    -2.05
     pizarra
    -2.00
     porque
    -1.97
     мужская
    -1.94
     significativos
    -1.92
     måske
    -1.88
     profundos
    -1.87
    ↑↑↑</
    -1.87
     dlatego
    -1.86
     maternelle
    -1.79
    POSITIVE LOGITS
     Maintaining
    2.61
    ’
    2.59
     Ensuring
    2.33
     Allowing
    2.19
     According
    2.17
     допомогти
    2.14
    This
    2.13
     Similarly
    2.09
     Roughly
    2.09
     crux
    2.06
    Activations Density 0.001%

    No Known Activations