© 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. 65930
    Prev
    Next
    INDEX
    Explanations

    always, consistently, around, above

    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
     еще
    -1.20
     ještě
    -1.05
     jeszcze
    -0.99
    因
    -0.98
     cương
    -0.94
     ešte
    -0.92
     ancora
    -0.89
    monių
    -0.89
     influential
    -0.88
     potranno
    -0.88
    POSITIVE LOGITS
     sempre
    1.60
    always
    1.48
     consistently
    1.39
     mindig
    1.34
     always
    1.34
     siempre
    1.33
    始终
    1.30
    ตลอด
    1.30
    except
    1.28
     zawsze
    1.27
    Activations Density 0.071%

    No Known Activations