© 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. 84484
    Prev
    Next
    INDEX
    Explanations

    code and parsing

    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
     in
    -2.55
     at
    -2.33
     for
    -2.25
     on
    -2.20
    {
    -2.19
     as
    -2.14
     provides
    -2.05
     or
    -2.03
     with
    -1.94
     an
    -1.82
    POSITIVE LOGITS
     luscious
    2.11
     ravaged
    1.98
     incríveis
    1.96
     ludicrous
    1.91
    ⓹
    1.89
    捽
    1.88
     delightfully
    1.88
     unbelievably
    1.87
     semelhantes
    1.87
     différentes
    1.86
    Activations Density 0.007%

    No Known Activations