© 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. 22561
    Prev
    Next
    INDEX
    Explanations

    which followed by adjective/noun

    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_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
     tráiler
    -1.40
     aparcamiento
    -1.38
     acogedor
    -1.37
     jaune
    -1.35
     botellas
    -1.34
    rovna
    -1.34
     estacionamiento
    -1.33
     macetas
    -1.33
     envió
    -1.31
     increíble
    -1.30
    POSITIVE LOGITS
     which
    1.92
     who
    1.80
     we
    1.61
     how
    1.49
     whose
    1.42
     I
    1.31
     can
    1.31
    f
    1.28
    c
    1.23
    B
    1.20
    Activations Density 0.134%

    No Known Activations