© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Jacobian LensNEW
    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsinterp-engineNEWAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 45121
    Prev
    Next
    INDEX
    Explanations

    here we go. talking about

    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
     want
    -1.38
     cartaz
    -1.33
     using
    -1.31
     chosen
    -1.27
     TODAY
    -1.23
     would
    -1.23
     requested
    -1.21
     given
    -1.20
     wanted
    -1.20
     ~~~
    -1.20
    POSITIVE LOGITS
    ieuses
    1.52
    ień
    1.25
     that
    1.22
    jaard
    1.20
    Tahun
    1.19
    rfan
    1.19
    áculo
    1.18
    řit
    1.18
    mino
    1.17
     akhirnya
    1.17
    Activations Density 0.060%

    No Known Activations