© 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. 12448
    Prev
    Next
    INDEX
    Explanations

    rather than, instead of, or versus

    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
    包括
    -1.08
     membentuk
    -0.92
     養
    -0.91
    ಏ
    -0.88
     in
    -0.88
     their
    -0.86
     using
    -0.85
     menghi
    -0.84
     мода
    -0.84
     specified
    -0.83
    POSITIVE LOGITS
     rather
    2.31
     instead
    1.98
    而不是
    1.84
    rather
    1.68
     versus
    1.62
     anstatt
    1.61
     plutôt
    1.60
    instead
    1.45
     whereas
    1.43
     chứ
    1.40
    Activations Density 0.074%

    No Known Activations