© 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-2-27B
    5. 22-GEMMASCOPE-RES-131K
    7. 28459
    Prev
    Next
    INDEX
    Explanations

    maintain / maintaining / maintained

    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
    Procedimento
    -1.10
     Ternyata
    -1.08
     Eğer
    -1.06
     Semoga
    -1.05
     Sebelum
    -1.05
     Akhir
    -1.05
    itieren
    -1.02
    setStart
    -1.02
    ïdes
    -1.00
     Fakta
    -1.00
    POSITIVE LOGITS
     throughout
    1.29
    throughout
    1.16
     on
    1.13
     maintains
    1.08
     Maintain
    1.06
    maintained
    1.06
     tradition
    1.02
     momentum
    1.02
     vigil
    1.02
    maintain
    1.00
    Activations Density 0.068%

    No Known Activations