© 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. 22-GEMMASCOPE-RES-131K
    7. 26090
    Prev
    Next
    INDEX
    Explanations

    structured attributes and values

    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
     which
    -1.13
     everyone
    -1.01
     Ա
    -0.93
     for
    -0.91
    めました
    -0.91
    的東西
    -0.89
     each
    -0.89
     where
    -0.88
     whose
    -0.87
     Հ
    -0.86
    POSITIVE LOGITS
     farlo
    1.32
     pamamagitan
    1.09
     hacerlo
    1.06
    Dazu
    0.97
    Doing
    0.94
    Doe
    0.93
     bourgeoisie
    0.91
     ऐसा
    0.91
     путем
    0.90
    Sunt
    0.88
    Activations Density 0.002%

    No Known Activations