© 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-3-27B
    5. 31-GEMMASCOPE-2-RES-262K
    7. 138021
    Prev
    Next
    INDEX
    Explanations

    sponsor

    np_acts-logits-general · gemini-2.5-flash-lite

    sponsored content

    np_acts-logits-general · gemini-2.5-flash-lite
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-2-27b-pt/resid_post/layer_31_width_262k_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
     subjug
    0.43
     Languages
    0.41
    +}$
    0.40
     seminal
    0.39
     linguistic
    0.38
    |$,
    0.38
     languages
    0.37
    }}^{(\
    0.37
     hostility
    0.37
     кана
    0.37
    POSITIVE LOGITS
     sponsor
    1.56
     Sponsor
    1.50
    sponsor
    1.44
     sponsors
    1.33
     sponsored
    1.32
     sponsoring
    1.28
     patrocin
    1.27
     Sponsors
    1.23
    Sponsored
    1.22
    sponsored
    1.17
    Activations Density 0.226%

    No Known Activations