© 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 EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 34-GEMMASCOPE-RES-131K
    7. 108030
    Prev
    Next
    INDEX
    Explanations

    Swar, Sver, Svar, Schwar prefixes

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron strongly activates on single, capitalized tokens that begin with the “Sv/Sw/Suvar…” letter sequence—i.e. proper names or place names starting with S-v or S-w.

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_34/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
     captain
    -0.75
     eigentlich
    -0.73
     Kinders
    -0.71
     UNA
    -0.70
    owns
    -0.70
     pour
    -0.70
    pour
    -0.70
     entrenamiento
    -0.69
    ndr
    -0.69
     sò
    -0.68
    POSITIVE LOGITS
    บล
    0.75
    یان
    0.72
     facilities
    0.71
    Promote
    0.70
    Галерея
    0.70
    Psychological
    0.68
     Shortly
    0.68
    ANTES
    0.68
    Татьяна
    0.67
    猎
    0.66
    Activations Density 0.016%

    No Known Activations