© 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-3-12B
    5. 31-GEMMASCOPE-2-RES-16K
    7. 6348
    Prev
    Next
    INDEX
    Explanations

    countries: Serbia, Croatia, Yugoslavia

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-12b-pt/resid_post/layer_31_width_16k_l0_medium
    Prompts (Dashboard)
    392,802 prompts, 256 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     япон
    0.78
     боль
    0.74
     ambulatory
    0.73
    ленных
    0.71
     буль
    0.70
    скага
    0.67
     Mexique
    0.66
    韩
    0.66
     Phillies
    0.65
    эль
    0.65
    POSITIVE LOGITS
     Serbia
    1.95
     Serbian
    1.90
    Serbia
    1.84
     Croatian
    1.80
    Croatian
    1.73
     Croatia
    1.69
     Beograd
    1.66
     BiH
    1.64
     Yugoslav
    1.62
     Belgrade
    1.61
    Activations Density 0.216%

    No Known Activations