© 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. 16367
    Prev
    Next
    INDEX
    Explanations

    improving or reducing conditions

    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
     to
    -1.59
    ebly
    -1.21
    再去
    -1.10
    の高い
    -1.10
     ทำ
    -1.10
     fã
    -1.07
    找个
    -1.07
    mbal
    -1.06
    lysm
    -1.06
    机遇
    -1.05
    POSITIVE LOGITS
     and
    1.43
     or
    1.21
    Răspunde
    1.12
    หรือ
    1.12
    tablissements
    1.10
    台灣
    1.06
     people
    1.02
     delapan
    1.00
     their
    1.00
    estero
    0.99
    Activations Density 0.072%

    No Known Activations