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

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-3-27B
    5. 53-GEMMASCOPE-2-RES-65K
    7. 5419
    Prev
    Next
    INDEX
    Explanations

    action verbs followed by direct objects

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

    refusing to generate harmful responses

    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_53_width_65k_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
     Partly
    0.42
    これを
    0.35
     ။
    0.33
     although
    0.32
     Partially
    0.32
     unterschiedlich
    0.31
    这点
    0.31
     meantime
    0.31
     partly
    0.31
    いくつかの
    0.30
    POSITIVE LOGITS
     ANYTHING
    0.42
     anything
    0.41
     ashore
    0.40
    imize
    0.38
     exclusivamente
    0.38
    ilize
    0.38
     tillbaka
    0.37
     cierta
    0.36
     определенных
    0.36
     certain
    0.36
    Activations Density 0.090%

    No Known Activations