© 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. 38656
    Prev
    Next
    INDEX
    Explanations

    initialization and errors

    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
     expects
    -1.43
    really
    -1.35
    Trunk
    -1.32
     "
    -1.31
     eigene
    -1.27
    entiary
    -1.25
     světa
    -1.24
    žka
    -1.23
    ţa
    -1.22
    costs
    -1.21
    POSITIVE LOGITS
    </strong>
    1.88
     erfolgreichen
    1.51
    Are
    1.44
    Where
    1.38
     vollständ
    1.38
    At
    1.37
     zuvor
    1.31
    What
    1.31
    When
    1.30
    Several
    1.27
    Activations Density 0.010%

    No Known Activations