© 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. 11034
    Prev
    Next
    INDEX
    Explanations

    should / must / ought

    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
     Whilst
    -1.63
     faktisk
    -1.41
     whilst
    -1.40
     nemlig
    -1.39
     overras
    -1.37
     ujedno
    -1.34
    Whilst
    -1.33
     filós
    -1.32
     moje
    -1.30
     poft
    -1.30
    POSITIVE LOGITS
     should
    1.75
     —
    1.48
     shouldn
    1.47
     must
    1.28
     должен
    1.27
    我们
    1.23
     ought
    1.21
     this
    1.21
     --
    1.20
    ควร
    1.19
    Activations Density 0.023%

    No Known Activations