© 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. 84108
    Prev
    Next
    INDEX
    Explanations

    modal verbs and 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
     opini
    -1.51
     soms
    -1.46
    鳚
    -1.44
     manier
    -1.39
    sometimes
    -1.39
     sond
    -1.38
     goud
    -1.38
     interrom
    -1.37
     inond
    -1.35
     aboli
    -1.35
    POSITIVE LOGITS
     would
    1.89
     that
    1.75
     should
    1.47
     could
    1.43
     must
    1.41
     whatsoever
    1.34
     can
    1.32
     abierta
    1.23
     or
    1.23
     можно
    1.23
    Activations Density 0.035%

    No Known Activations