© 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. 25192
    Prev
    Next
    INDEX
    Explanations

    like, for example

    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
    Not in Any Lists

    No Comments

    Negative Logits
     också
    -1.24
     også
    -1.23
     myös
    -1.14
     tudi
    -1.14
     arī
    -1.13
     also
    -1.12
    也是
    -1.09
     alfo
    -1.07
     также
    -1.05
    encing
    -1.02
    POSITIVE LOGITS
     would
    1.72
     some
    1.47
     nor
    1.42
     niektórych
    1.34
     often
    1.33
     sometimes
    1.32
    Some
    1.30
     others
    1.29
     sondern
    1.28
     might
    1.27
    Activations Density 0.045%

    No Known Activations